NFL — additional Python functions

Hand-written wrappers, loaders, and helpers in sportsdataverse.nfl not covered by the generated API-endpoint reference above.

Play-by-play, schedule & rosters

espn_nfl_game_rosters(game_id: 'int', raw=False, return_as_pandas=False, **kwargs) -> 'pl.DataFrame'

espn_nfl_game_rosters() - Pull the game by id.

Parameters

Parameter	Type	Default	Description
game_id	int		Unique game_id, can be obtained from espn_nfl_schedule().
raw		False
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe.

Returns

Polars dataframe of game roster data with columns: 'athlete_id', 'athlete_uid', 'athlete_guid', 'athlete_type', 'first_name', 'last_name', 'full_name', 'athlete_display_name', 'short_name', 'weight', 'display_weight', 'height', 'display_height', 'age', 'date_of_birth', 'slug', 'jersey', 'linked', 'active', 'alternate_ids_sdr', 'birth_place_city', 'birth_place_state', 'birth_place_country', 'headshot_href', 'headshot_alt', 'experience_years', 'experience_display_value', 'experience_abbreviation', 'status_id', 'status_name', 'status_type', 'status_abbreviation', 'hand_type', 'hand_abbreviation', 'hand_display_value', 'draft_display_text', 'draft_round', 'draft_year', 'draft_selection', 'player_id', 'starter', 'valid', 'did_not_play', 'display_name', 'ejected', 'athlete_href', 'position_href', 'statistics_href', 'team_id', 'team_guid', 'team_uid', 'team_slug', 'team_location', 'team_name', 'team_nickname', 'team_abbreviation', 'team_display_name', 'team_short_display_name', 'team_color', 'team_alternate_color', 'is_active', 'is_all_star', 'team_alternate_ids_sdr', 'logo_href', 'logo_dark_href', 'game_id'

col_name	type	description
athlete_id	integer	Unique athlete identifier (ESPN).
athlete_uid	character	ESPN athlete UID (universal identifier).
athlete_guid	character	ESPN athlete GUID.
athlete_type	character	Athlete type / class.
first_name	character	First name of player
last_name	character	Last name of player
full_name	character	Full name as per NFL.com
athlete_display_name	character	Athlete display name (full).
short_name	character	Player short name (i.e. "F.Last")
weight	double	Official weight, in pounds
display_weight	character	Player weight in display format (e.g. '180 lbs').
height	double	Official height, in inches
display_height	character	Player height in display format (e.g. '6-2').
age	integer	Age as of last pipeline build, rounded to one decimal. Pipeline is built on a weekly basis.
date_of_birth	character	Date of birth (YYYY-MM-DD).
debut_year	integer	Year of professional debut.
slug	character	URL-safe identifier.
jersey	character	Jersey number worn by the player.
linked	logical	TRUE if the record is linked to a related entity.
active	logical	TRUE if the row represents an active record (player / team / season).
alternate_ids_sdr	character	Alternate ids sdr.
birth_place_city	character	Birth place city.
birth_place_state	character	Birth place state.
birth_place_country	character	Birth place country.
headshot_href	character	Link to ESPN Headshot of Player
headshot_alt	character	Alternative-text label for the headshot.
projections_href	character	ESPN API hyperlink reference URL for the player's statistical projection resource.
contracts_href	character	ESPN API hyperlink reference URL for the full list of the player's historical contract records.
experience_years	integer	Experience years.
college_athlete_href	character	ESPN API hyperlink reference URL for the athlete's college profile resource.
contract_href	character	ESPN API hyperlink reference URL for the player's full contract resource.
contract_option_type	integer	Contract option type.
contract_salary	integer	Contract salary.
contract_bonus	integer	Signing or roster bonus amount (in dollars) associated with the player's current contract.
contract_years_remaining	integer	Contract years remaining.
contract_signed_through	integer	Final year of the player's current contract, expressed as a four-digit season year.
contract_season_href	character	ESPN API hyperlink reference URL for the specific season-level contract detail resource.
contract_team_href	character	ESPN API hyperlink reference URL for the team associated with the player's contract.
contract_active	logical	Contract active.
status_id	character	Status identifier.
status_name	character	Status label.
status_type	character	Status type.
status_abbreviation	character	Status abbreviation.
contract_salary_remaining	integer	Contract salary remaining.
draft_display_text	character	Draft display text.
draft_round	integer	Round that player was drafted in
draft_year	integer	Year that player was drafted
draft_selection	integer	Draft selection.
draft_team_href	character	ESPN API hyperlink reference URL for the team that originally drafted this player.
draft_pick_href	character	ESPN API hyperlink reference URL for the draft-pick record associated with this player.
hand_type	character	Hand type.
hand_abbreviation	character	Hand abbreviation.
hand_display_value	character	Hand display value.
starter	logical	TRUE if the player was in the starting lineup; FALSE otherwise.
jersey_right	character	Secondary or alternate jersey number display string used by ESPN (e.g., for special-game uniforms).
valid	logical	Valid.
did_not_play	logical	TRUE if the player did not appear in the game.
display_name	character	Full name of player
athlete_href	character	ESPN API hyperlink reference URL for the athlete resource.
position_href	character	ESPN API hyperlink reference URL for the player's positional classification resource.
statistics_href	character	ESPN API hyperlink reference URL for the player's career statistics resource.
team_id	integer	Unique team identifier.
order	integer	Display order within the result set.
home_away	character	Game venue label ('home' or 'away').
winner	logical	Winner.
team_guid	character	ESPN team GUID.
team_uid	character	ESPN universal team identifier (UID format 's:40~l:...~t:...').
team_slug	character	URL-safe team identifier (e.g. 'lasvegas-aces' / 'aces').
team_location	character	Team city or location string.
team_name	character	Full team display name (e.g. 'Las Vegas Aces').
team_nickname	character	Team nickname.
team_abbreviation	character	Short team abbreviation (e.g. 'LAS').
team_display_name	character	Full team display name.
team_short_display_name	character	Short team display name (e.g. 'Aces').
team_color	character	Team primary color (hex without leading '#').
team_alternate_color	character	Team alternate color (hex without leading '#').
is_active	logical	Active contract
is_all_star	logical	Is all star.
team_alternate_ids_sdr	character	SportsDataverse SDR alternate identifier for the team, used for cross-source joins.
logo_href	character	Team or league logo URL.
logo_dark_href	character	Logo URL for dark backgrounds.
game_id	integer	Ten digit identifier for NFL game.

Example

from sportsdataverse.nfl import espn_nfl_game_rosters
rosters = espn_nfl_game_rosters(game_id=401220403)
rosters.shape

# Pandas round-trip with home/away split

rosters_pd = espn_nfl_game_rosters(game_id=401220403, return_as_pandas=True)
home = rosters_pd[rosters_pd["home_away"] == "home"]
away = rosters_pd[rosters_pd["home_away"] == "away"]

espn_nfl_player_stats(athlete_id: 'int', season: 'int', *, season_type: 'str' = 'regular', total: 'bool' = False, raw: 'bool' = False, return_as_pandas: 'bool' = False, **kwargs: 'Any') -> 'pl.DataFrame | pd.DataFrame | dict[str, Any]'

Pull an NFL athlete's ESPN season stat line as one wide row.

See sportsdataverse.wbb.espn_wbb_player_stats for full documentation of the wide return shape, the {category}_{stat} stat columns (for football: passing_*, rushing_*, receiving_*, scoring_*, ...), the athlete / team metadata blocks, and the season_type / total parameters. For the richer multi-category web-v3 payload use sportsdataverse.nfl.espn_nfl_player_stats_v3.

Parameters

Parameter	Type	Default	Description
athlete_id	int		ESPN NFL athlete identifier (e.g. 3139477 for Patrick Mahomes).
season	int		Season year, used in the core-v2 path.
season_type	str	'regular'	"regular" (type 2) or "postseason" (type 3).
total	bool	False	Forward-compat totals passthrough.
raw	bool	False	If True, returns the raw core-v2 statistics JSON dict.
return_as_pandas	bool	False	If True, returns a pandas DataFrame; else polars.

Returns

A single-row wide DataFrame (polars by default). When raw=True returns the raw statistics JSON dict.

col_name	type	description
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
season_type	character	REG or POST indicating if the timeframe belongs to regular or post season.
total	logical	The sum of each team's score in the game. Equals h_score + v_score. Is NA for games which haven't yet been played. Convenient for evaluating over/under total bets.
athlete_id	integer	Unique athlete identifier (ESPN).
athlete_uid	character	ESPN athlete UID (universal identifier).
athlete_guid	character	ESPN athlete GUID.
athlete_type	character	Athlete type / class.
first_name	character	First name of player
last_name	character	Last name of player
full_name	character	Full name as per NFL.com
display_name	character	Full name of player
short_name	character	Player short name (i.e. "F.Last")
weight	double	Official weight, in pounds
display_weight	character	Player weight in display format (e.g. '180 lbs').
height	double	Official height, in inches
display_height	character	Player height in display format (e.g. '6-2').
age	integer	Age as of last pipeline build, rounded to one decimal. Pipeline is built on a weekly basis.
date_of_birth	character	Date of birth (YYYY-MM-DD).
jersey	character	Jersey number worn by the player.
slug	character	URL-safe identifier.
active	logical	TRUE if the row represents an active record (player / team / season).
position_id	integer	Unique position identifier.
position_name	character	Listed roster position ('Guard', 'Forward', 'Center').
position_display_name	character	Position display name.
position_abbreviation	character	Position abbreviation ('G' / 'F' / 'C').
college_name	character	Official college (usually the last one attended)
status_id	integer	Status identifier.
status_name	character	Status label.
general_fumbles	double	Total number of times the player fumbled the ball regardless of who recovered it.
general_fumbles_lost	double	Number of fumbles by the player that were subsequently recovered by the opposing team.
general_fumbles_forced	double	Number of fumbles the player forced from opposing ball carriers.
general_fumbles_forced_primary	double	Number of fumbles forced where the player was credited as the primary forcer rather than an assist.
general_fumbles_recovered	double	Number of fumbles (own or opponent's) recovered by the player.
general_fumbles_recovered_yards	double	Total yards gained on fumble recoveries returned by the player.
general_fumbles_touchdowns	double	Number of touchdowns scored by the player on fumble recoveries.
general_games_played	double	Games Played.
general_offensive_two_pt_returns	double	Number of two-point conversion attempts the player's offense returned defensively for a score.
general_offensive_fumbles_touchdowns	double	Number of touchdowns scored by recovering an offensive team fumble (e.g., a fumble recovered in the end zone).
general_defensive_fumbles_touchdowns	double	Number of touchdowns scored by recovering or returning an opponent's fumble on the defensive side.
passing_avg_gain	double	Average yards gained per passing attempt including sacks.
passing_completion_pct	double	Percentage of pass attempts that resulted in a completed reception (completions divided by attempts).
passing_completions	double	Pass completions (split from CFBD's C/ATT field).
passing_espnqb_rating	double	ESPN's proprietary composite quarterback rating blending efficiency, usage, and situational performance into a single metric.
passing_interception_pct	double	Percentage of pass attempts that resulted in an interception (interceptions divided by attempts).
passing_interceptions	double	Total number of passes thrown that were intercepted by the opposing defense.
passing_long_passing	double	Longest completed pass in yards recorded by the player during the period.
passing_net_passing_yards	double	Passing yards minus yards lost on sacks, giving a net aerial production figure.
passing_net_passing_yards_per_game	double	Net passing yards (after sack yardage deduction) per game played.
passing_net_total_yards	double	Combined net rushing and net passing yards accumulated by the player.
passing_net_yards_per_game	double	Net total offensive yards (rushing plus net passing) per game.
passing_passing_attempts	double	Total number of pass attempts thrown by the player.
passing_passing_big_plays	double	Number of passing plays that gained 20 or more yards.
passing_passing_first_downs	double	Number of pass completions that resulted in a first down.
passing_passing_fumbles	double	Number of times the player fumbled while in the act of passing or being sacked.
passing_passing_fumbles_lost	double	Number of passing-related fumbles that were recovered by the opposing team.
passing_passing_touchdown_pct	double	Percentage of pass attempts that resulted in a touchdown (touchdowns divided by attempts).
passing_passing_touchdowns	double	Total number of touchdown passes thrown by the player.
passing_passing_yards	double	Total aerial yards gained on completions thrown by the player.
passing_passing_yards_after_catch	double	Total yards gained by receivers after catching the ball on passes thrown by the player.
passing_passing_yards_at_catch	double	Total yards gained through the air (prior to the catch) on completions thrown by the player.
passing_passing_yards_per_game	double	Gross passing yards per game played by the quarterback.
passing_qb_rating	double	Traditional NFL passer rating computed from completion percentage, yards per attempt, touchdown percentage, and interception percentage on a roughly 0–158.3 scale.
passing_sacks	double	Total number of times the player was sacked behind the line of scrimmage while attempting to pass.
passing_sack_yards_lost	double	Total yards lost by the player as a result of being sacked behind the line of scrimmage.
passing_net_passing_attempts	double	Total pass attempts minus sacks taken, representing meaningful dropbacks.
passing_team_games_played	double	Number of games played by the player's team during which the player accumulated passing statistics.
passing_total_offensive_plays	double	Total number of offensive plays (pass attempts, rushes, and sacks) run with the player as the primary ball handler.
passing_total_points_per_game	double	Average total points scored by the player's team per game.
passing_total_touchdowns	double	Total touchdowns (passing, rushing, and receiving) scored by or credited to the player.
passing_total_yards	double	Combined total of passing, rushing, and receiving yards accumulated by the player.
passing_total_yards_from_scrimmage	double	Total yards from scrimmage (rushing plus receiving) credited to the player in addition to passing yards.
passing_two_point_pass_convs	double	Number of successful two-point conversion attempts thrown by the player.
passing_two_pt_pass	double	Number of two-point conversion passes the player successfully completed.
passing_two_pt_pass_attempts	double	Number of two-point conversion pass attempts thrown by the player regardless of outcome.
passing_yards_from_scrimmage_per_game	double	Yards from scrimmage (rushing plus receiving) per game for a player who also has passing statistics recorded.
passing_yards_per_completion	double	Average yards gained per completed pass attempt.
passing_yards_per_game	double	Gross passing yards per game (equivalent to passing_passing_yards_per_game; alternate label).
passing_yards_per_pass_attempt	double	Gross passing yards divided by total pass attempts, not penalizing for sacks.
passing_net_yards_per_pass_attempt	double	Net passing yards divided by total pass attempts including sacks, penalizing passers for yardage lost in the pocket.
passing_qbr	double	ESPN Quarterback Rating (QBR) for the player in this game.
passing_adj_qbr	double	ESPN's Adjusted Total Quarterback Rating, which adjusts raw QBR for opponent strength and clutch situations on a 0–100 scale.
passing_quarterback_rating	double	Alternate or supplemental quarterback rating value; may represent a different calculation context from passing_qb_rating (e.g., situational or opponent-adjusted).
rushing_avg_gain	double	Average yards gained per rushing attempt by the player.
rushing_espnrb_rating	double	ESPN's proprietary composite running back rating blending efficiency and usage metrics.
rushing_long_rushing	double	Longest single rushing play in yards recorded by the player during the period.
rushing_net_total_yards	double	Combined net rushing and receiving yards accumulated by the player.
rushing_net_yards_per_game	double	Net total offensive yards per game for the player.
rushing_rushing_attempts	double	Total number of rushing attempts carried by the player.
rushing_rushing_big_plays	double	Number of rushing plays that gained 10 or more yards.
rushing_rushing_first_downs	double	Number of rushing attempts that resulted in a first down.
rushing_rushing_fumbles	double	Number of times the player fumbled while carrying the ball on a rushing play.
rushing_rushing_fumbles_lost	double	Number of rushing fumbles by the player that were recovered by the opposing team.
rushing_rushing_touchdowns	double	Total number of rushing touchdowns scored by the player.
rushing_rushing_yards	double	Total yards gained by the player on all rushing attempts.
rushing_rushing_yards_per_game	double	Rushing yards per game played by the player.
rushing_stuffs	double	Number of rushing attempts where the player was tackled at or behind the line of scrimmage.
rushing_stuff_yards_lost	double	Total yards lost on rushing plays where the player was tackled behind the line of scrimmage (stuffed).
rushing_team_games_played	double	Number of games played by the player's team during which the player accumulated rushing statistics.
rushing_total_offensive_plays	double	Total number of offensive plays run with the player active in a rushing role.
rushing_total_points_per_game	double	Average total points scored by the player's team per game.
rushing_total_touchdowns	double	Total touchdowns (rushing and receiving) scored by the player.
rushing_total_yards	double	Combined total rushing and receiving yards accumulated by the player.
rushing_total_yards_from_scrimmage	double	Total yards from scrimmage (rushing plus receiving) credited to the player.
rushing_two_point_rush_convs	double	Number of successful two-point conversion rushes scored by the player.
rushing_two_pt_rush	double	Number of two-point conversion rushes the player successfully converted.
rushing_two_pt_rush_attempts	double	Number of two-point conversion rush attempts by the player regardless of outcome.
rushing_yards_from_scrimmage_per_game	double	Total yards from scrimmage per game for the player.
rushing_yards_per_game	double	Rushing yards per game (equivalent label to rushing_rushing_yards_per_game).
rushing_yards_per_rush_attempt	double	Average yards gained per rushing attempt by the player.
receiving_avg_gain	double	Average yards gained per reception by the player.
receiving_espnwr_rating	double	ESPN's proprietary composite wide receiver / pass-catcher rating blending efficiency and usage metrics.
receiving_long_reception	double	Longest single reception in yards recorded by the player during the period.
receiving_net_total_yards	double	Combined net rushing and receiving yards accumulated by the player.
receiving_net_yards_per_game	double	Net total offensive yards (rushing plus receiving) per game for the player.
receiving_receiving_big_plays	double	Number of receptions that gained 20 or more yards.
receiving_receiving_first_downs	double	Number of receptions that resulted in a first down.
receiving_receiving_fumbles	double	Number of times the player fumbled after making a reception.
receiving_receiving_fumbles_lost	double	Number of post-reception fumbles by the player that were recovered by the opposing team.
receiving_receiving_targets	double	Total number of times the player was the intended target of a pass attempt.
receiving_receiving_touchdowns	double	Total number of touchdown receptions credited to the player.
receiving_receiving_yards	double	Total yards gained by the player on all receptions.
receiving_receiving_yards_after_catch	double	Total yards gained by the player after making contact with the ball (yards after catch).
receiving_receiving_yards_at_catch	double	Total air yards at the point of the catch (depth of target) on receptions by the player.
receiving_receiving_yards_per_game	double	Receiving yards per game played by the player.
receiving_receptions	double	Total number of passes successfully caught by the player.
receiving_team_games_played	double	Number of games played by the player's team during which the player accumulated receiving statistics.
receiving_total_offensive_plays	double	Total number of offensive plays run during which the player was active on the field.
receiving_total_points_per_game	double	Average total points scored by the player's team per game (context for the receiver's role).
receiving_total_touchdowns	double	Total touchdowns (rushing and receiving) scored by the player.
receiving_total_yards	double	Combined total rushing and receiving yards accumulated by the player.
receiving_total_yards_from_scrimmage	double	Total yards from scrimmage (rushing plus receiving) credited to the player.
receiving_two_point_rec_convs	double	Number of successful two-point conversion receptions caught by the player.
receiving_two_pt_reception	double	Number of two-point conversion passes the player successfully caught.
receiving_two_pt_reception_attempts	double	Number of two-point conversion targets thrown to the player regardless of outcome.
receiving_yards_from_scrimmage_per_game	double	Total yards from scrimmage per game for the player.
receiving_yards_per_game	double	Receiving yards per game (equivalent label to receiving_receiving_yards_per_game).
receiving_yards_per_reception	double	Average yards gained per reception, also known as yards per catch.
defensive_assist_tackles	double	Number of assisted tackles credited to the player (helped bring down the ball carrier but was not the primary tackler).
defensive_avg_interception_yards	double	Average yards gained per interception returned by the player.
defensive_avg_sack_yards	double	Average yards lost per sack the player recorded against the opposing quarterback.
defensive_avg_stuff_yards	double	Average yards lost per run stuff (tackle behind the line of scrimmage) recorded by the player.
defensive_blocked_field_goal_touchdowns	double	Number of touchdowns scored by the player after blocking an opponent's field goal attempt and returning it.
defensive_blocked_punt_touchdowns	double	Number of touchdowns scored by the player after blocking an opponent's punt and returning it.
defensive_hurries	double	Number of times the player pressured the quarterback into an early or errant throw without recording a sack.
defensive_kicks_blocked	double	Total number of kicks (field goals or extra points) the player blocked.
defensive_long_interception	double	Longest single interception return in yards recorded by the player.
defensive_misc_touchdowns	double	Number of defensive touchdowns scored via miscellaneous means not captured by other specific categories.
defensive_passes_batted_down	double	Number of passes the player knocked down at the line of scrimmage without recording an interception.
defensive_passes_defended	double	Total number of passes the player broke up or deflected, including both pass deflections and interceptions.
defensive_qb_hits	double	Number of times the player legally hit the quarterback during or just after a pass attempt.
defensive_two_pt_returns	double	Number of two-point conversion attempts the player's defense returned for a defensive conversion score.
defensive_sacks	double	Sacks credited to the player.
defensive_sack_yards	double	Total yards lost by the opposing offense as a result of the player's sacks.
defensive_safeties	double	Number of safeties recorded by the player (tackling the ball carrier in their own end zone).
defensive_solo_tackles	double	Number of unassisted tackles credited solely to the player.
defensive_stuffs	double	Number of times the player tackled a ball carrier for a loss on a rushing play.
defensive_stuff_yards	double	Total yards lost by the offense on run stuffs recorded by the player.
defensive_tackles_for_loss	double	Total number of tackles resulting in a loss of yards for the opposing offense.
defensive_tackles_yards_lost	double	Total yards lost by the opposing offense on the player's tackles for loss.
defensive_team_games_played	double	Number of games played by the player's team in which the player appeared on the defensive side.
defensive_total_tackles	double	Combined total of solo tackles and assisted tackles recorded by the player.
defensive_yards_allowed	double	Total yards allowed by the player's defense during games the player appeared in.
defensive_points_allowed	double	Total points allowed by the player's team during games the player appeared in.
defensive_one_pt_safeties_made	double	Number of one-point safeties recorded by the player's defense (scored when the opposing team is downed in their own end zone during a try).
defensive_missed_field_goal_return_td	double	Number of touchdowns scored by returning a missed field goal attempt.
defensive_blocked_punt_ez_rec_td	double	Number of touchdowns scored by recovering a blocked punt in the end zone.
defensive_interceptions_interceptions	double	Total number of passes intercepted by the player.
defensive_interceptions_interception_touchdowns	double	Number of touchdowns scored by the player on interception returns.
defensive_interceptions_interception_yards	double	Total yards gained by the player on interception returns.
scoring_defensive_points	double	Total points scored by the player's defense via safeties, defensive touchdowns, and blocked kick returns.
scoring_field_goals	double	Total number of field goals successfully made by the player during the period.
scoring_kick_extra_points	double	Total number of extra point attempts (PAT kicks) by the player.
scoring_kick_extra_points_made	double	Total number of extra points successfully kicked through the uprights by the player.
scoring_misc_points	double	Points scored via miscellaneous methods not captured by other scoring categories.
scoring_passing_touchdowns	double	Total passing touchdowns thrown by the player, contributing to their total scoring line.
scoring_receiving_touchdowns	double	Total receiving touchdowns scored by the player.
scoring_return_touchdowns	double	Total touchdowns scored by the player on kick or punt returns.
scoring_rushing_touchdowns	double	Total rushing touchdowns scored by the player.
scoring_total_points	double	Total points contributed by the player across all scoring methods (touchdowns, PATs, field goals, etc.).
scoring_total_points_per_game	double	Average total points contributed by the player per game.
scoring_total_touchdowns	double	Total touchdowns scored or thrown by the player across all methods.
scoring_total_two_point_convs	double	Total number of successful two-point conversions scored or thrown by the player.
scoring_two_point_pass_convs	double	Number of successful two-point conversion passes thrown by the player.
scoring_two_point_rec_convs	double	Number of successful two-point conversion receptions caught by the player.
scoring_two_point_rush_convs	double	Number of successful two-point conversion rushes scored by the player.
scoring_one_pt_safeties_made	double	Number of one-point safeties recorded, scored when the defense stops an offense that is attempting a two-point conversion in their own end zone.
team_id	integer	Unique team identifier.
team_uid	character	ESPN universal team identifier (UID format 's:40~l:...~t:...').
team_guid	character	ESPN team GUID.
team_slug	character	URL-safe team identifier (e.g. 'lasvegas-aces' / 'aces').
team_location	character	Team city or location string.
team_name	character	Full team display name (e.g. 'Las Vegas Aces').
team_abbreviation	character	Short team abbreviation (e.g. 'LAS').
team_display_name	character	Full team display name.
team_short_display_name	character	Short team display name (e.g. 'Aces').
team_color	character	Team primary color (hex without leading '#').
team_alternate_color	character	Team alternate color (hex without leading '#').
team_is_active	logical	TRUE if the team is currently active.
team_logo_href	character	Default team logo URL; team_detail = TRUE only.

Example

from sportsdataverse.nfl import espn_nfl_player_stats
df = espn_nfl_player_stats(athlete_id=3139477, season=2023)
df.select(["full_name", "team_display_name", "passing_passing_yards"])

espn_nfl_schedule(dates=None, week=None, season_type=None, groups=None, limit=500, return_as_pandas=False, **kwargs) -> 'pl.DataFrame'

espn_nfl_schedule - look up the NFL schedule for a given season

Parameters

Parameter	Type	Default	Description
dates	int	None	Used to define different seasons. 2002 is the earliest available season.
week	int	None	Week of the schedule.
season_type	int	None	2 for regular season, 3 for post-season, 4 for off-season.
groups		None
limit	int	500	number of records to return, default: 500.
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe.

Returns

Polars dataframe containing schedule dates for the requested season. Returns None if no games

col_name	type	description
id	character	ID of the player in the 'name' column.
uid	character	ESPN UID string.
date	character	Date in YYYY-MM-DD format.
attendance	integer	Reported attendance.
time_valid	logical	Whether the start time is confirmed.
neutral_site	logical	Neutral site.
conference_competition	logical	Conference competition.
play_by_play_available	logical	Whether play-by-play data is available.
recent	logical	Whether the game is recent.
start_date	character	Start date (YYYY-MM-DD).
broadcast	character	Broadcast information string.
highlights	character	Game highlight urls.
notes_type	character	Notes type.
notes_headline	character	Notes headline.
broadcast_market	character	Broadcast market label (e.g. 'national', 'home').
broadcast_name	character	Broadcast name.
type_id	character	Type identifier (numeric).
type_abbreviation	character	Play type abbreviation.
venue_id	character	Unique venue identifier.
venue_full_name	character	Venue full name.
venue_address_city	character	Venue address city.
venue_address_state	character	Venue address state / region.
venue_address_country	character	Two-letter ISO country code or country name for the country where the venue is located.
venue_indoor	logical	Whether the home venue is indoors.
status_clock	double	Game clock in seconds.
status_display_clock	character	Status display clock.
status_period	integer	Current period.
status_type_id	character	Unique identifier for status type.
status_type_name	character	Status type name.
status_type_state	character	Status state (pre/in/post).
status_type_completed	logical	Whether the game is complete.
status_type_description	character	Status type description.
status_type_detail	character	Status type detail.
status_type_short_detail	character	Status type short detail.
status_is_tbd_flex	logical	Boolean flag indicating whether the game's broadcast slot is designated as a flex/TBD window.
format_regulation_periods	integer	Format regulation periods.
home_id	character	Unique identifier for home.
home_uid	character	Home team's uid.
home_location	character	Home team's location.
home_name	character	Home team display name.
home_abbreviation	character	Home team's abbreviation.
home_display_name	character	Home team display name.
home_short_display_name	character	Home short display name.
home_color	character	Home team primary color hex.
home_alternate_color	character	Color code (hex) for home alternate.
home_is_active	logical	Home team's is active.
home_venue_id	character	Unique identifier for home venue.
home_logo	character	Home team logo URL.
home_score	character	The number of points the home team scored. Is NA for games which haven't yet been played.
home_current_rank	integer	Current AP or ESPN power ranking of the home team at the time of the game.
home_linescores	list	Comma-separated or serialized quarter-by-quarter score totals for the home team.
home_records	character	Serialized win-loss-tie record(s) for the home team (e.g., overall, home, away, conference).
away_id	character	Unique identifier for away.
away_uid	character	Away team's uid.
away_location	character	Away team's location.
away_name	character	Away team display name.
away_abbreviation	character	Away team's abbreviation.
away_display_name	character	Away team display name.
away_short_display_name	character	Away short display name.
away_color	character	Away team primary color hex.
away_alternate_color	character	Color code (hex) for away alternate.
away_is_active	logical	Away team's is active.
away_venue_id	character	Unique identifier for away venue.
away_logo	character	Away team logo URL.
away_score	character	The number of points the away team scored. Is NA for games which haven't yet been played.
away_current_rank	integer	Current AP or ESPN power ranking of the away team at the time of the game.
away_linescores	list	Comma-separated or serialized quarter-by-quarter score totals for the away team.
away_records	character	Serialized win-loss-tie record(s) for the away team (e.g., overall, home, away, conference).
game_id	integer	Ten digit identifier for NFL game.
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
season_type	integer	REG or POST indicating if the timeframe belongs to regular or post season.
week	integer	Season week.

Example

from sportsdataverse.nfl import espn_nfl_schedule
sched = espn_nfl_schedule(dates=20240908)

# Specific week of regular season (``season_type=2``)

wk1 = espn_nfl_schedule(dates=2024, week=1, season_type=2)

# Pandas round-trip

sched_pd = espn_nfl_schedule(dates=20240908, return_as_pandas=True)

Dataset loaders

load_combine(return_as_pandas=False) -> 'pl.DataFrame'

Load NFL Combine information

Parameters

Parameter	Type	Default	Description
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe.

Returns

Polars dataframe containing NFL combine data available.

col_name	type	description
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
draft_year	double	Year that player was drafted
draft_team	character	Team that drafted player
draft_round	double	Round that player was drafted in
draft_ovr	double	Overall draft pick selection. This can be a little bit patchy, since MFL does not report this number.
pfr_id	character	Pro-Football-Reference ID for player
cfb_id	character	Sports Reference (CFB) ID for player
player_name	character	Full name of player
pos	character	Position as tracked by FP
school	character	College of player
ht	character	Height of player (feet and inches)
wt	double	Weight of player (lbs)
forty	double	Player's 40 yard dash time at combine (seconds)
bench	double	Reps benched by player at combine
vertical	double	Player's vertical jump at combine (inches)
broad_jump	double	Player's broad jump at combine (inches)
cone	double	Player's 3 cone drill time at combine (seconds)
shuttle	double	Player's shuttle run time at combine (seconds)

Example

from sportsdataverse.nfl import load_nfl_combine
combine = load_nfl_combine()
combine.shape

# Filter by draft year and position

import polars as pl
qbs_2024 = (
    load_nfl_combine()
    .filter((pl.col("season") == 2024) & (pl.col("pos") == "QB"))
)

load_contracts(return_as_pandas=False) -> 'pl.DataFrame'

Load NFL Historical contracts information

Parameters

Parameter	Type	Default	Description
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe.

Returns

Polars dataframe containing historical contracts available.

col_name	type	description
player	character	Player name
position	character	Primary position as reported by NFL.com
team	character	NFL team. Uses official abbreviations as per NFL.com
is_active	logical	Active contract
year_signed	integer	Year the contract was signed
years	integer	Contract length
value	double	Total contract value
apy	double	Average money per contract year
guaranteed	double	Total guaranteed money
apy_cap_pct	double	Average money per contract year as percentage of the team's salary cap at signing
inflated_value	double	Total contract value inflated to account for the rise of the salary cap
inflated_apy	double	Average money per contract year inflated to account for the rise of the salary cap
inflated_guaranteed	double	Total guaranteed money inflated to account for the rise of the salary cap
player_page	character	Player's OverTheCap url
otc_id	integer	Over the Cap ID for player
gsis_id	character	Game Stats and Info Service ID: the primary ID for play-by-play data.
date_of_birth	character	Date of birth (YYYY-MM-DD).
height	character	Official height, in inches
weight	character	Official weight, in pounds
college	character	Official college (usually the last one attended)
draft_year	integer	Year that player was drafted
draft_round	integer	Round that player was drafted in
draft_overall	integer	Overall draft selection number.
draft_team	character	Team that drafted player
cols	double	Placeholder column retained in the contracts loader output schema; contains no meaningful data in this context.

Example

from sportsdataverse.nfl import load_nfl_contracts
contracts = load_nfl_contracts()
contracts.shape

# Pandas round-trip with sort by APY

contracts_pd = load_nfl_contracts(return_as_pandas=True)
contracts_pd.sort_values("apy", ascending=False).head()

load_depth_charts(seasons: 'List[int]', return_as_pandas=False) -> 'pl.DataFrame'

Load NFL Depth Chart data for selected seasons

Parameters

Parameter	Type	Default	Description
seasons	list		Used to define different seasons. 2001 is the earliest available season.
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe.

Returns

Polars dataframe containing depth chart data available for the requested seasons.

col_name	type	description
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
club_code	character	Three-letter team abbreviation identifying the NFL club on the depth chart row.
week	integer	Season week.
game_type	character	The most recent game type of that season that a player appeared on the roster.
depth_team	character	Numeric depth rank indicating whether the player is listed as the starter (1), backup (2), or further reserve on the depth chart.
last_name	character	Last name of player
first_name	character	First name of player
football_name	character	Common player name (i.e. in most cases common_first_name last_name)
formation	character	Offensive or defensive formation context in which the depth chart position applies (e.g., 'Shotgun', 'Nickel').
gsis_id	character	Game Stats and Info Service ID: the primary ID for play-by-play data.
jersey_number	character	Jersey number. Often useful for joins by name/team/jersey.
position	character	Primary position as reported by NFL.com
elias_id	character	Elias Sports Bureau identifier for the player, used by the NFL for official statistical tracking.
depth_position	character	Positional grouping label used to place the player on the team's official depth chart (e.g., 'QB', 'WR1', 'ILB').
full_name	character	Full name as per NFL.com

Example

from sportsdataverse.nfl import load_nfl_depth_charts
depth = load_nfl_depth_charts(seasons=[2024])

# Multi-season range

depth = load_nfl_depth_charts(seasons=range(2020, 2025))

load_draft_picks(return_as_pandas=False) -> 'pl.DataFrame'

Load NFL Draft picks information

Parameters

Parameter	Type	Default	Description
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe.

Returns

Polars dataframe containing NFL Draft picks data available.

col_name	type	description
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
round	integer	Draft round
pick	integer	Draft overall pick
team	character	NFL team. Uses official abbreviations as per NFL.com
gsis_id	character	Game Stats and Info Service ID: the primary ID for play-by-play data.
pfr_player_id	character	ID from Pro Football Reference
cfb_player_id	character	ID from College Football Reference
pfr_player_name	character	Player's name as recorded by PFR
hof	logical	Whether player has been selected to the Pro Football Hall of Fame
position	character	Primary position as reported by NFL.com
category	character	Broader category of player positions
side	character	O for offense, D for defense, S for special teams
college	character	Official college (usually the last one attended)
age	integer	Age as of last pipeline build, rounded to one decimal. Pipeline is built on a weekly basis.
to	integer	Final season played in NFL
allpro	integer	Number of AP First Team All-Pro selections as recorded by PFR
probowls	integer	Number of Pro Bowls
seasons_started	integer	Number of seasons recorded as primary starter for position
w_av	integer	Weighted Approximate Value
car_av	logical	Career Approximate Value
dr_av	integer	Draft Approximate Value
games	integer	Games played in career
pass_completions	integer	Number of successful completions for a given game
pass_attempts	integer	Career pass attempts
pass_yards	integer	Number of yards gained on pass plays
pass_tds	integer	Career pass touchdowns thrown
pass_ints	integer	Career pass interceptions thrown
rush_atts	integer	Career rushing attempts
rush_yards	integer	The number of rushing yards gained
rush_tds	integer	Career rushing touchdowns
receptions	integer	The number of pass receptions. Lateral receptions officially don't count as reception.
rec_yards	integer	Career receiving yards
rec_tds	integer	Career receiving touchdowns
def_solo_tackles	integer	Career solo tackles
def_ints	integer	Career interceptions
def_sacks	double	Number of sacks form this player

Example

from sportsdataverse.nfl import load_nfl_draft_picks
picks = load_nfl_draft_picks()
picks.shape

# Filter to a single year and round

import polars as pl
r1_2024 = (
    load_nfl_draft_picks()
    .filter((pl.col("season") == 2024) & (pl.col("round") == 1))
)

load_espn_qbr(seasons: 'List[int]', summary_type: 'str' = 'season', return_as_pandas: 'bool' = False, *, source: 'str' = 'nflverse') -> 'pl.DataFrame'

Load ESPN Total QBR (Quarterback Rating) data going back to 2006.

Mirrors nflreadpy / nflreadr load_espn_qbr -- the lone nflreadpy dataset that previously had no sdv-py loader. ESPN publishes Total QBR only from 2006 onward, so 2006 is the earliest available season (unlike the 1999 floor on play-by-play). nflverse republishes ESPN's QBR through the espn_data release as two combined files (one per summary_type), each covering all seasons; this loader reads the requested file once and post-filters by season (the same access pattern as load_nfl_schedule).

Parameters

Parameter	Type	Default	Description
seasons	list		Seasons to return. 2006 is the earliest available season.
summary_type	str	'season'	Aggregation level. "season" (default) returns one row per quarterback-season; "week" returns one row per quarterback-game. Any other value raises ValueError.
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe.
source	str	'nflverse'	Which QBR release to read. "nflverse" (the default, also accepts None) returns the nflverse espn_data release. "sportsdataverse" / "sdv" returns the SDV-native nfl_espn_qbr release (built by nfl-data from ESPN's QBR web endpoint -- the same source nflverse's espnscrapeR uses). Any other value raises ValueError.

Returns

Polars dataframe containing ESPN Total QBR for the requested seasons, summarized per summary_type.

Example

from sportsdataverse.nfl import load_nfl_espn_qbr
qbr = load_nfl_espn_qbr(seasons=[2024])
qbr.shape

# Week-level QBR

qbr_week = load_nfl_espn_qbr(seasons=[2024], summary_type="week")

# Multi-season range

qbr = load_nfl_espn_qbr(seasons=range(2020, 2025))

# Pandas round-trip

qbr_pd = load_nfl_espn_qbr(seasons=[2024], return_as_pandas=True)
qbr_pd[["season", "team_abb", "qbr_total"]].head()

load_ff_opportunity(seasons: 'List[int]', stat_type: 'str' = 'weekly', model_version: 'str' = 'latest', return_as_pandas=False) -> 'pl.DataFrame'

Load NFL fantasy football opportunity data from ffverse/ffopportunity

Parameters

Parameter	Type	Default	Description
seasons	list		Used to define different seasons. 2006 is the earliest available season.
stat_type	str	'weekly'	One of "weekly", "pbp_pass", "pbp_rush". Defaults to "weekly".
model_version	str	'latest'	One of "latest", "v1.0.0". Defaults to "latest".
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe.

Returns

Polars dataframe containing fantasy football opportunity data for the requested seasons.

col_name	type	description
season	character	4 digit number indicating to which season(s) the specified timeframe belongs to.
posteam	character	String abbreviation for the team with possession.
week	double	Season week.
game_id	character	Ten digit identifier for NFL game.
player_id	character	Player ID (aka GSIS ID) as defined by nflreadr::load_rosters
full_name	character	Full name as per NFL.com
position	character	Primary position as reported by NFL.com
pass_attempt	double	Binary indicator for if the play was a pass attempt (includes sacks).
rec_attempt	double	Total number of targets for a given game
rush_attempt	double	Binary indicator for if the play was a run.
pass_air_yards	double	Total air yards thrown for a given game
rec_air_yards	double	Total air yards on receiving attempts for a given game
pass_completions	double	Number of successful completions for a given game
receptions	double	The number of pass receptions. Lateral receptions officially don't count as reception.
pass_completions_exp	double	Expected number of pass_completions in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
receptions_exp	double	Expected number of receptions in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
pass_yards_gained	double	Total passing yards gained for a given game
rec_yards_gained	double	Total receiving yards gained for a given game
rush_yards_gained	double	Total rushing yards gained for a given game
pass_yards_gained_exp	double	Expected number of pass_yards_gained in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
rec_yards_gained_exp	double	Expected number of rec_yards_gained in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
rush_yards_gained_exp	double	Expected number of rush_yards_gained in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
pass_touchdown	double	Binary indicator for if the play resulted in a passing TD.
rec_touchdown	double	Total receiving touchdowns
rush_touchdown	double	Binary indicator for if the play resulted in a rushing TD.
pass_touchdown_exp	double	Expected number of pass_touchdown in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
rec_touchdown_exp	double	Expected number of rec_touchdown in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
rush_touchdown_exp	double	Expected number of rush_touchdown in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
pass_two_point_conv	double	Number of successful passing two point conversions
rec_two_point_conv	double	Number of successful receiving two point conversions
rush_two_point_conv	double	Number of successful rushing two point conversions
pass_two_point_conv_exp	double	Expected number of pass_two_point_conv in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
rec_two_point_conv_exp	double	Expected number of rec_two_point_conv in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
rush_two_point_conv_exp	double	Expected number of rush_two_point_conv in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
pass_first_down	double	Number of passing first downs
rec_first_down	double	Number of receiving first downs
rush_first_down	double	Number of rushing first downs
pass_first_down_exp	double	Expected number of pass_first_down in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
rec_first_down_exp	double	Expected number of rec_first_down in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
rush_first_down_exp	double	Expected number of rush_first_down in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
pass_interception	double	Number of interceptions thrown
rec_interception	double	Number of interceptions on targets
pass_interception_exp	double	Expected number of pass_interception in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
rec_interception_exp	double	Expected number of rec_interception in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
rec_fumble_lost	double	Number of fumbles on receiving attempts
rush_fumble_lost	double	Number of fumbles on rushing attempts
pass_fantasy_points_exp	double	Expected number of pass_fantasy_points in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
rec_fantasy_points_exp	double	Expected number of rec_fantasy_points in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
rush_fantasy_points_exp	double	Expected number of rush_fantasy_points in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
pass_fantasy_points	double	Total fantasy points from passing, assuming 0.04 points per pass yard, 4 points per pass TD, -2 points per interception
rec_fantasy_points	double	Total fantasy points from receiving, assuming PPR scoring
rush_fantasy_points	double	Total fantasy points from rushing, assuming PPR scoring
total_yards_gained	double	Total scrimmage yards (sum of pass, rush, and receiving yards)
total_yards_gained_exp	double	Expected number of total_yards_gained in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
total_touchdown	double	Total touchdowns (sum of pass, rush, and receiving touchdowns)
total_touchdown_exp	double	Expected number of total_touchdown in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
total_first_down	double	Total first downs (sum of pass, rush, and receiving first downs)
total_first_down_exp	double	Expected number of total_first_down in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
total_fantasy_points	double	Total fantasy points (sum of pass, rush, and receiving fantasy points)
total_fantasy_points_exp	double	Expected number of total_fantasy_points in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
pass_completions_diff	double	Difference between actual and expected number of pass_completions - often interpreted as efficiency for a given play/game
receptions_diff	double	Difference between actual and expected number of receptions - often interpreted as efficiency for a given play/game
pass_yards_gained_diff	double	Difference between actual and expected number of pass_yards_gained - often interpreted as efficiency for a given play/game
rec_yards_gained_diff	double	Difference between actual and expected number of rec_yards_gained - often interpreted as efficiency for a given play/game
rush_yards_gained_diff	double	Difference between actual and expected number of rush_yards_gained - often interpreted as efficiency for a given play/game
pass_touchdown_diff	double	Difference between actual and expected number of pass_touchdown - often interpreted as efficiency for a given play/game
rec_touchdown_diff	double	Difference between actual and expected number of rec_touchdown - often interpreted as efficiency for a given play/game
rush_touchdown_diff	double	Difference between actual and expected number of rush_touchdown - often interpreted as efficiency for a given play/game
pass_two_point_conv_diff	double	Difference between actual and expected number of pass_two_point_conv - often interpreted as efficiency for a given play/game
rec_two_point_conv_diff	double	Difference between actual and expected number of rec_two_point_conv - often interpreted as efficiency for a given play/game
rush_two_point_conv_diff	double	Difference between actual and expected number of rush_two_point_conv - often interpreted as efficiency for a given play/game
pass_first_down_diff	double	Difference between actual and expected number of pass_first_down - often interpreted as efficiency for a given play/game
rec_first_down_diff	double	Difference between actual and expected number of rec_first_down - often interpreted as efficiency for a given play/game
rush_first_down_diff	double	Difference between actual and expected number of rush_first_down - often interpreted as efficiency for a given play/game
pass_interception_diff	double	Difference between actual and expected number of pass_interception - often interpreted as efficiency for a given play/game
rec_interception_diff	double	Difference between actual and expected number of rec_interception - often interpreted as efficiency for a given play/game
pass_fantasy_points_diff	double	Difference between actual and expected number of pass_fantasy_points - often interpreted as efficiency for a given play/game
rec_fantasy_points_diff	double	Difference between actual and expected number of rec_fantasy_points - often interpreted as efficiency for a given play/game
rush_fantasy_points_diff	double	Difference between actual and expected number of rush_fantasy_points - often interpreted as efficiency for a given play/game
total_yards_gained_diff	double	Difference between actual and expected number of total_yards_gained - often interpreted as efficiency for a given play/game
total_touchdown_diff	double	Difference between actual and expected number of total_touchdown - often interpreted as efficiency for a given play/game
total_first_down_diff	double	Difference between actual and expected number of total_first_down - often interpreted as efficiency for a given play/game
total_fantasy_points_diff	double	Difference between actual and expected number of total_fantasy_points - often interpreted as efficiency for a given play/game
pass_attempt_team	double	Team-level total pass_attempt for a game, summed across all plays/players for that team.
rec_attempt_team	double	Team-level total rec_attempt for a game, summed across all plays/players for that team.
rush_attempt_team	double	Team-level total rush_attempt for a game, summed across all plays/players for that team.
pass_air_yards_team	double	Team-level total pass_air_yards for a game, summed across all plays/players for that team.
rec_air_yards_team	double	Team-level total rec_air_yards for a game, summed across all plays/players for that team.
pass_completions_team	double	Team-level total pass_completions for a game, summed across all plays/players for that team.
receptions_team	double	Team-level total receptions for a game, summed across all plays/players for that team.
pass_completions_exp_team	double	Team-level total expected pass_completions_exp for a game, summed across all plays & players for that team.
receptions_exp_team	double	Team-level total expected receptions_exp for a game, summed across all plays & players for that team.
pass_yards_gained_team	double	Team-level total pass_yards_gained for a game, summed across all plays/players for that team.
rec_yards_gained_team	double	Team-level total rec_yards_gained for a game, summed across all plays/players for that team.
rush_yards_gained_team	double	Team-level total rush_yards_gained for a game, summed across all plays/players for that team.
pass_yards_gained_exp_team	double	Team-level total expected pass_yards_gained_exp for a game, summed across all plays & players for that team.
rec_yards_gained_exp_team	double	Team-level total expected rec_yards_gained_exp for a game, summed across all plays & players for that team.
rush_yards_gained_exp_team	double	Team-level total expected rush_yards_gained_exp for a game, summed across all plays & players for that team.
pass_touchdown_team	double	Team-level total pass_touchdown for a game, summed across all plays/players for that team.
rec_touchdown_team	double	Team-level total rec_touchdown for a game, summed across all plays/players for that team.
rush_touchdown_team	double	Team-level total rush_touchdown for a game, summed across all plays/players for that team.
pass_touchdown_exp_team	double	Team-level total expected pass_touchdown_exp for a game, summed across all plays & players for that team.
rec_touchdown_exp_team	double	Team-level total expected rec_touchdown_exp for a game, summed across all plays & players for that team.
rush_touchdown_exp_team	double	Team-level total expected rush_touchdown_exp for a game, summed across all plays & players for that team.
pass_two_point_conv_team	double	Team-level total pass_two_point_conv for a game, summed across all plays/players for that team.
rec_two_point_conv_team	double	Team-level total rec_two_point_conv for a game, summed across all plays/players for that team.
rush_two_point_conv_team	double	Team-level total rush_two_point_conv for a game, summed across all plays/players for that team.
pass_two_point_conv_exp_team	double	Team-level total expected pass_two_point_conv_exp for a game, summed across all plays & players for that team.
rec_two_point_conv_exp_team	double	Team-level total expected rec_two_point_conv_exp for a game, summed across all plays & players for that team.
rush_two_point_conv_exp_team	double	Team-level total expected rush_two_point_conv_exp for a game, summed across all plays & players for that team.
pass_first_down_team	double	Team-level total pass_first_down for a game, summed across all plays/players for that team.
rec_first_down_team	double	Team-level total rec_first_down for a game, summed across all plays/players for that team.
rush_first_down_team	double	Team-level total rush_first_down for a game, summed across all plays/players for that team.
pass_first_down_exp_team	double	Team-level total expected pass_first_down_exp for a game, summed across all plays & players for that team.
rec_first_down_exp_team	double	Team-level total expected rec_first_down_exp for a game, summed across all plays & players for that team.
rush_first_down_exp_team	double	Team-level total expected rush_first_down_exp for a game, summed across all plays & players for that team.
pass_interception_team	double	Team-level total pass_interception for a game, summed across all plays/players for that team.
rec_interception_team	double	Team-level total rec_interception for a game, summed across all plays/players for that team.
pass_interception_exp_team	double	Team-level total expected pass_interception_exp for a game, summed across all plays & players for that team.
rec_interception_exp_team	double	Team-level total expected rec_interception_exp for a game, summed across all plays & players for that team.
rec_fumble_lost_team	double	Team-level total rec_fumble_lost for a game, summed across all plays/players for that team.
rush_fumble_lost_team	double	Team-level total rush_fumble_lost for a game, summed across all plays/players for that team.
pass_fantasy_points_exp_team	double	Team-level total expected pass_fantasy_points_exp for a game, summed across all plays & players for that team.
rec_fantasy_points_exp_team	double	Team-level total expected rec_fantasy_points_exp for a game, summed across all plays & players for that team.
rush_fantasy_points_exp_team	double	Team-level total expected rush_fantasy_points_exp for a game, summed across all plays & players for that team.
pass_fantasy_points_team	double	Team-level total pass_fantasy_points for a game, summed across all plays/players for that team.
rec_fantasy_points_team	double	Team-level total rec_fantasy_points for a game, summed across all plays/players for that team.
rush_fantasy_points_team	double	Team-level total rush_fantasy_points for a game, summed across all plays/players for that team.
pass_completions_diff_team	double	Team-level difference between actual and expected number of pass_completions_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
receptions_diff_team	double	Team-level difference between actual and expected number of receptions_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
pass_yards_gained_diff_team	double	Team-level difference between actual and expected number of pass_yards_gained_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
rec_yards_gained_diff_team	double	Team-level difference between actual and expected number of rec_yards_gained_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
rush_yards_gained_diff_team	double	Team-level difference between actual and expected number of rush_yards_gained_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
pass_touchdown_diff_team	double	Team-level difference between actual and expected number of pass_touchdown_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
rec_touchdown_diff_team	double	Team-level difference between actual and expected number of rec_touchdown_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
rush_touchdown_diff_team	double	Team-level difference between actual and expected number of rush_touchdown_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
pass_two_point_conv_diff_team	double	Team-level difference between actual and expected number of pass_two_point_conv_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
rec_two_point_conv_diff_team	double	Team-level difference between actual and expected number of rec_two_point_conv_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
rush_two_point_conv_diff_team	double	Team-level difference between actual and expected number of rush_two_point_conv_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
pass_first_down_diff_team	double	Team-level difference between actual and expected number of pass_first_down_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
rec_first_down_diff_team	double	Team-level difference between actual and expected number of rec_first_down_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
rush_first_down_diff_team	double	Team-level difference between actual and expected number of rush_first_down_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
pass_interception_diff_team	double	Team-level difference between actual and expected number of pass_interception_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
rec_interception_diff_team	double	Team-level difference between actual and expected number of rec_interception_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
pass_fantasy_points_diff_team	double	Team-level difference between actual and expected number of pass_fantasy_points_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
rec_fantasy_points_diff_team	double	Team-level difference between actual and expected number of rec_fantasy_points_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
rush_fantasy_points_diff_team	double	Team-level difference between actual and expected number of rush_fantasy_points_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
total_yards_gained_team	double	Team-level total total_yards_gained for a game, summed across all plays/players for that team.
total_yards_gained_exp_team	double	Team-level total expected total_yards_gained_exp for a game, summed across all plays & players for that team.
total_yards_gained_diff_team	double	Team-level difference between actual and expected number of total_yards_gained_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
total_touchdown_team	double	Team-level total total_touchdown for a game, summed across all plays/players for that team.
total_touchdown_exp_team	double	Team-level total expected total_touchdown_exp for a game, summed across all plays & players for that team.
total_touchdown_diff_team	double	Team-level difference between actual and expected number of total_touchdown_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
total_first_down_team	double	Team-level total total_first_down for a game, summed across all plays/players for that team.
total_first_down_exp_team	double	Team-level total expected total_first_down_exp for a game, summed across all plays & players for that team.
total_first_down_diff_team	double	Team-level difference between actual and expected number of total_first_down_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
total_fantasy_points_team	double	Team-level total total_fantasy_points for a game, summed across all plays/players for that team.
total_fantasy_points_exp_team	double	Team-level total expected total_fantasy_points_exp for a game, summed across all plays & players for that team.
total_fantasy_points_diff_team	double	Team-level difference between actual and expected number of total_fantasy_points_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.

Example

from sportsdataverse.nfl import load_nfl_ff_opportunity
weekly = load_nfl_ff_opportunity(seasons=[2024])

# Pass play-by-play opportunity stats

pbp_pass = load_nfl_ff_opportunity(seasons=[2024], stat_type="pbp_pass")

# Rush play-by-play opportunity stats with pinned model version

pbp_rush = load_nfl_ff_opportunity(
    seasons=[2024], stat_type="pbp_rush", model_version="v1.0.0"
)

load_ff_playerids(return_as_pandas=False) -> 'pl.DataFrame'

Load fantasy football player IDs from DynastyProcess.com

Parameters

Parameter	Type	Default	Description
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe.

Returns

Polars dataframe containing fantasy football player ID mappings across platforms.

col_name	type	description
mfl_id	integer	MyFantasyLeague.com ID - this is the primary key for this table and is unique and complete. Usually an integer of 5 digits.
sportradar_id	character	SportRadar ID - often also called sportsdata_id by other services. A UUID.
fantasypros_id	character	FantasyPros.com ID - usually an integer of 5 digits.
gsis_id	character	Game Stats and Info Service ID: the primary ID for play-by-play data.
pff_id	character	Pro Football Focus ID - usually an integer with between 3 and 6 digits.
sleeper_id	integer	Sleeper ID - usually an integer with ~4 digits.
nfl_id	character	NFL ID of player (this is used in Big Data Bowl Data)
espn_id	integer	ESPN ID - usual format is an integer with ~5 digits
yahoo_id	character	Yahoo ID - usual format is an integer with ~5 digits
fleaflicker_id	character	Fleaflicker ID - usual format is an integer with ~4 digits. Fleaflicker API also has sportradar and that's generally preferred.
cbs_id	integer	CBS ID - usual format is an integer with ~ 7 digits.
pfr_id	character	Pro-Football-Reference ID for player
cfbref_id	character	College Football Reference ID - usual format is firstname-lastname-integer
rotowire_id	integer	Rotowire ID - usual format is an integer with ~four digits. Not to be confused with rotowire_id.
rotoworld_id	character	Rotoworld ID - usual format is an integer with ~four digits. Not to be confused with rotowire_id.
ktc_id	integer	KeepTradeCut ID - usual format is an integer with ~four digits.
stats_id	integer	Stats ID - usual format is five digit integer
stats_global_id	integer	Stats Global ID - usual format is a six digit integer
fantasy_data_id	integer	FantasyData ID - usual format five digit integer
swish_id	character	Player ID for Swish Analytics
name	character	Name, as reported by MFL but reordered into FirstName LastName instead of Last, First
merge_name	character	Name but formatted for name joins via ffscrapr::dp_cleannames() - coerced to lowercase, stripped of punctuation and suffixes, and common substitutions performed.
position	character	Primary position as reported by NFL.com
team	character	NFL team. Uses official abbreviations as per NFL.com
birthdate	character	Birthdate
age	double	Age as of last pipeline build, rounded to one decimal. Pipeline is built on a weekly basis.
draft_year	integer	Year that player was drafted
draft_round	integer	Round that player was drafted in
draft_pick	integer	Draft pick within round, i.e. 32nd pick of second round.
draft_ovr	integer	Overall draft pick selection. This can be a little bit patchy, since MFL does not report this number.
twitter_username	character	Official twitter handle, if known
height	integer	Official height, in inches
weight	integer	Official weight, in pounds
college	character	Official college (usually the last one attended)
db_season	integer	Year of database build. Previous years may also be available via dynastyprocess.

Example

from sportsdataverse.nfl import load_nfl_ff_playerids
ids = load_nfl_ff_playerids()
ids.shape

# Filter to active QBs

import polars as pl
qbs = (
    load_nfl_ff_playerids()
    .filter((pl.col("position") == "QB") & (pl.col("status") == "ACT"))
)

load_ff_rankings(type: 'str' = 'draft', kind: 'str' = None, return_as_pandas=False) -> 'pl.DataFrame'

Load fantasy football rankings and projections

Parameters

Parameter	Type	Default	Description
type	str	'draft'	Type of rankings to load. One of "draft" (current draft rankings), "week" (weekly rankings), or "all" (full historical rankings). Defaults to "draft". Kept for nflreadpy parity since its parameter is also called type; the forward-going preferred name is kind.
kind	str	None	Preferred parameter name. Same semantics and allowed values as type. If both are supplied, kind wins. If neither is supplied, defaults to "draft" via type.
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe.

Returns

Polars dataframe containing fantasy football rankings data.

col_name	type	description
fp_page	character	The relative url that the data was scraped from (add the prefix https://www.fantasypros.com/ to visit the page)
page_type	character	Two word identifier separated by a dash identifying the type of fantasy ranking (best = bestball; dynasty; redraft) and what position it applies to
ecr_type	character	A two letter identifier combining the ranking type (b = bestball; d = dynasty; r = redraft) and position type (o = overall; p = positional; sf = superflex; rk = rookie)
player	character	Player name
id	integer	ID of the player in the 'name' column.
pos	character	Position as tracked by FP
team	character	NFL team. Uses official abbreviations as per NFL.com
ecr	double	Average (mean) expert ranking for this player
sd	double	Standard deviation of expert rankings for this player
best	integer	The highest ranking given for this player by any one expert
worst	integer	The lowest ranking given for this player by any one expert
sportsdata_id	character	ID - also known as sportradar_id (they are equivalent!)
player_filename	character	base URL for this player on fantasypros.com
yahoo_id	character	Yahoo ID - usual format is an integer with ~5 digits
cbs_id	character	CBS ID - usual format is an integer with ~ 7 digits.
player_owned_avg	double	The average percentage this player is rostered across ESPN and Yahoo
player_owned_espn	character	The percentage that this player is rostered in ESPN leagues
player_owned_yahoo	character	The percentage that this player is rostered in Yahoo leagues
player_image_url	character	An image of the player
player_square_image_url	character	An square image of the player
rank_delta	integer	Change in ranks over a recent period
bye	integer	NFL bye week
mergename	character	Player name after being cleaned by dp_cleannames - generally strips punctuation and suffixes as well as performing common name substitutions.
scrape_date	character	Date this dataframe was last updated
tm	character	Team ID as used on MyFantasyLeague.com

Example

from sportsdataverse.nfl import load_nfl_ff_rankings
draft = load_nfl_ff_rankings(kind="draft")

# Weekly rankings

weekly = load_nfl_ff_rankings(kind="week")

# Full historical rankings (parquet)

history = load_nfl_ff_rankings(kind="all")

# nflreadpy-parity ``type=`` parameter (still supported)

draft = load_nfl_ff_rankings(type="draft")

load_ftn_charting(seasons: 'List[int]', return_as_pandas=False) -> 'pl.DataFrame'

Load NFL FTN charting data going back to 2022

Parameters

Parameter	Type	Default	Description
seasons	list		Used to define different seasons. 2022 is the earliest available season.
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe.

Returns

Polars dataframe containing FTN charting data available for the requested seasons.

col_name	type	description
ftn_game_id	integer	FTN game ID
nflverse_game_id	character	nflverse identifier for games. Format is season, week, away_team, home_team
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
week	integer	Season week.
ftn_play_id	integer	FTN play ID
nflverse_play_id	integer	Play ID used by nflverse, corresponds to GSIS play ID
starting_hash	character	hash the ball was place(L = left, M = middle, R = right)
qb_location	character	pre-snap position of quarterback(U = under center, S = shotgun, P = pistol)
n_offense_backfield	integer	number of players in the backfield at the snap
n_defense_box	integer	Number of defenders positioned in the box at the snap, as charted by FTN Data.
is_no_huddle	logical	no huddle
is_motion	logical	motion occurred on the play before or at the time of the snap
is_play_action	logical	play-action pass
is_screen_pass	logical	screen pass
is_rpo	logical	play is considered run-pass option
is_trick_play	logical	trick play
is_qb_out_of_pocket	logical	quarterback moved out of pocket
is_interception_worthy	logical	interception worthy pass
is_throw_away	logical	quarterback thrown away
read_thrown	character	read the ball was thrown
is_catchable_ball	logical	catchable ball(defined by throws that are generally on target that are not defended away)
is_contested_ball	logical	contested ball(defined by whether or not the receiver is facing physical contact at the time of the catch)
is_created_reception	logical	created reception(defined by a reception that only occurs due to an exceptional play by the receiver)
is_drop	logical	receiver drop
is_qb_sneak	logical	quarterback sneak
n_blitzers	integer	number of blitzers
n_pass_rushers	integer	number of pass rushers
is_qb_fault_sack	logical	sack that is the fault of the quarterback
date_pulled	character	Date the data was retrieved from the FTN Data API by nflverse jobs

Example

from sportsdataverse.nfl import load_nfl_ftn_charting
charting = load_nfl_ftn_charting(seasons=[2024])

# Multi-season range

charting = load_nfl_ftn_charting(seasons=range(2022, 2025))

# Filter to plays with motion

import polars as pl
motion_plays = (
    load_nfl_ftn_charting(seasons=[2024])
    .filter(pl.col("is_motion") == 1)
)

load_injuries(seasons: 'List[int]', return_as_pandas=False) -> 'pl.DataFrame'

Load NFL injuries data for selected seasons

Parameters

Parameter	Type	Default	Description
seasons	list		Used to define different seasons. 2009 is the earliest available season.
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe.

Returns

Polars dataframe containing injuries data available for the requested seasons.

col_name	type	description
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
game_type	character	The most recent game type of that season that a player appeared on the roster.
team	character	NFL team. Uses official abbreviations as per NFL.com
week	integer	Season week.
gsis_id	character	Game Stats and Info Service ID: the primary ID for play-by-play data.
position	character	Primary position as reported by NFL.com
full_name	character	Full name as per NFL.com
first_name	character	First name of player
last_name	character	Last name of player
report_primary_injury	character	Primary injury listed on official injury report
report_secondary_injury	character	Secondary injury listed on official injury report
report_status	character	Player's status for game on official injury report
practice_primary_injury	character	Primary injury listed on practice injury report
practice_secondary_injury	character	Secondary injury listed on practice injury report
practice_status	character	Player's participation in practice
date_modified	character	Date and time that injury information was updated

Example

from sportsdataverse.nfl import load_nfl_injuries
injuries = load_nfl_injuries(seasons=[2024])

# Multi-season range with team filter

import polars as pl
sf_injuries = (
    load_nfl_injuries(seasons=range(2020, 2025))
    .filter(pl.col("team") == "SF")
)

load_nextgen_stats(seasons: 'List[int]', stat_type: 'str' = 'passing', return_as_pandas: 'bool' = False) -> 'pl.DataFrame'

Load NFL NextGen Stats data going back to 2016.

Unified loader that consolidates the per-stat-type NextGen Stats accessors. Mirrors the API surface of nflreadpy's load_nextgen_stats so downstream code can swap engines without changing call sites.

Parameters

Parameter	Type	Default	Description
seasons	list[int]		Seasons to filter to. The upstream parquet covers a single combined file per stat type — seasons is applied as a post-filter on the season column.
stat_type	str	'passing'	One of "passing", "rushing", "receiving". Defaults to "passing".
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe.

Returns

Polars dataframe containing NextGen Stats data for the requested stat_type and seasons.

col_name	type	description
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
season_type	character	REG or POST indicating if the timeframe belongs to regular or post season.
week	integer	Season week.
player_display_name	character	Full name of the player
player_position	character	Position of the player accordinng to NGS
team_abbr	character	Official team abbreveation
avg_time_to_throw	double	Average time elapsed from the time of snap to throw on every pass attempt for a passer (sacks excluded).
avg_completed_air_yards	double	Average air yards on completed passes
avg_intended_air_yards	double	Average air yards on all attempted passes
avg_air_yards_differential	double	Air Yards Differential is calculated by subtracting the passer's average Intended Air Yards from his average Completed Air Yards. This stat indicates if he is on average attempting deep passes than he on average completes.
aggressiveness	double	Aggressiveness tracks the amount of passing attempts a quarterback makes that are into tight coverage, where there is a defender within 1 yard or less of the receiver at the time of completion or incompletion. AGG is shown as a % of attempts into tight windows over all passing attempts.
max_completed_air_distance	double	Air Distance is the amount of yards the ball has traveled on a pass, from the point of release to the point of reception (as the crow flies). Unlike Air Yards, Air Distance measures the actual distance the passer throws the ball.
avg_air_yards_to_sticks	double	Air Yards to the Sticks shows the amount of Air Yards ahead or behind the first down marker on all attempts for a passer. The metric indicates if the passer is attempting his passes past the 1st down marker, or if he is relying on his skill position players to make yards after catch.
attempts	integer	The number of pass attempts as defined by the NFL.
pass_yards	integer	Number of yards gained on pass plays
pass_touchdowns	integer	Number of touchdowns scored on pass plays
interceptions	integer	The number of interceptions thrown.
passer_rating	double	Overall NFL passer rating
completions	integer	The number of completed passes.
completion_percentage	double	Percentage of completed passes
expected_completion_percentage	double	Using a passer's Completion Probability on every play, determine what a passer's completion percentage is expected to be.
completion_percentage_above_expectation	double	A passer's actual completion percentage compared to their Expected Completion Percentage.
avg_air_distance	double	A receiver's average depth of target
max_air_distance	double	A receiver's maximum depth of target
player_gsis_id	character	Unique identifier of the player
player_first_name	character	Player's first name
player_last_name	character	Player's last name
player_jersey_number	integer	Player's jersey number
player_short_name	character	Short version of player's name

Example

from sportsdataverse.nfl import load_nfl_nextgen_stats
ngs_pass = load_nfl_nextgen_stats(seasons=[2024], stat_type="passing")

# Rushing NextGen stats

ngs_rush = load_nfl_nextgen_stats(seasons=[2024], stat_type="rushing")

# Receiving NextGen stats with a follow-up filter

import polars as pl
ngs_rec = (
    load_nfl_nextgen_stats(seasons=[2024], stat_type="receiving")
    .filter(pl.col("week") > 0)
)

# Pandas round-trip

ngs_pd = load_nfl_nextgen_stats(
    seasons=[2024], stat_type="passing", return_as_pandas=True
)

load_nfl_combine(return_as_pandas=False) -> 'pl.DataFrame'

Load NFL Combine information

Parameters

Parameter	Type	Default	Description
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe.

Returns

Polars dataframe containing NFL combine data available.

col_name	type	description
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
draft_year	double	Year that player was drafted
draft_team	character	Team that drafted player
draft_round	double	Round that player was drafted in
draft_ovr	double	Overall draft pick selection. This can be a little bit patchy, since MFL does not report this number.
pfr_id	character	Pro-Football-Reference ID for player
cfb_id	character	Sports Reference (CFB) ID for player
player_name	character	Full name of player
pos	character	Position as tracked by FP
school	character	College of player
ht	character	Height of player (feet and inches)
wt	double	Weight of player (lbs)
forty	double	Player's 40 yard dash time at combine (seconds)
bench	double	Reps benched by player at combine
vertical	double	Player's vertical jump at combine (inches)
broad_jump	double	Player's broad jump at combine (inches)
cone	double	Player's 3 cone drill time at combine (seconds)
shuttle	double	Player's shuttle run time at combine (seconds)

Example

from sportsdataverse.nfl import load_nfl_combine
combine = load_nfl_combine()
combine.shape

# Filter by draft year and position

import polars as pl
qbs_2024 = (
    load_nfl_combine()
    .filter((pl.col("season") == 2024) & (pl.col("pos") == "QB"))
)

load_nfl_contracts(return_as_pandas=False) -> 'pl.DataFrame'

Load NFL Historical contracts information

Parameters

Parameter	Type	Default	Description
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe.

Returns

Polars dataframe containing historical contracts available.

col_name	type	description
player	character	Player name
position	character	Primary position as reported by NFL.com
team	character	NFL team. Uses official abbreviations as per NFL.com
is_active	logical	Active contract
year_signed	integer	Year the contract was signed
years	integer	Contract length
value	double	Total contract value
apy	double	Average money per contract year
guaranteed	double	Total guaranteed money
apy_cap_pct	double	Average money per contract year as percentage of the team's salary cap at signing
inflated_value	double	Total contract value inflated to account for the rise of the salary cap
inflated_apy	double	Average money per contract year inflated to account for the rise of the salary cap
inflated_guaranteed	double	Total guaranteed money inflated to account for the rise of the salary cap
player_page	character	Player's OverTheCap url
otc_id	integer	Over the Cap ID for player
gsis_id	character	Game Stats and Info Service ID: the primary ID for play-by-play data.
date_of_birth	character	Date of birth (YYYY-MM-DD).
height	character	Official height, in inches
weight	character	Official weight, in pounds
college	character	Official college (usually the last one attended)
draft_year	integer	Year that player was drafted
draft_round	integer	Round that player was drafted in
draft_overall	integer	Overall draft selection number.
draft_team	character	Team that drafted player
cols	double	Number of contract columns returned in the contracts dataset (metadata artifact from the loader).

Example

from sportsdataverse.nfl import load_nfl_contracts
contracts = load_nfl_contracts()
contracts.shape

# Pandas round-trip with sort by APY

contracts_pd = load_nfl_contracts(return_as_pandas=True)
contracts_pd.sort_values("apy", ascending=False).head()

load_nfl_draft_picks(return_as_pandas=False) -> 'pl.DataFrame'

Load NFL Draft picks information

Parameters

Parameter	Type	Default	Description
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe.

Returns

Polars dataframe containing NFL Draft picks data available.

col_name	type	description
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
round	integer	Draft round
pick	integer	Draft overall pick
team	character	NFL team. Uses official abbreviations as per NFL.com
gsis_id	character	Game Stats and Info Service ID: the primary ID for play-by-play data.
pfr_player_id	character	ID from Pro Football Reference
cfb_player_id	character	ID from College Football Reference
pfr_player_name	character	Player's name as recorded by PFR
hof	logical	Whether player has been selected to the Pro Football Hall of Fame
position	character	Primary position as reported by NFL.com
category	character	Broader category of player positions
side	character	O for offense, D for defense, S for special teams
college	character	Official college (usually the last one attended)
age	integer	Age as of last pipeline build, rounded to one decimal. Pipeline is built on a weekly basis.
to	integer	Final season played in NFL
allpro	integer	Number of AP First Team All-Pro selections as recorded by PFR
probowls	integer	Number of Pro Bowls
seasons_started	integer	Number of seasons recorded as primary starter for position
w_av	integer	Weighted Approximate Value
car_av	logical	Career Approximate Value
dr_av	integer	Draft Approximate Value
games	integer	Games played in career
pass_completions	integer	Number of successful completions for a given game
pass_attempts	integer	Career pass attempts
pass_yards	integer	Number of yards gained on pass plays
pass_tds	integer	Career pass touchdowns thrown
pass_ints	integer	Career pass interceptions thrown
rush_atts	integer	Career rushing attempts
rush_yards	integer	The number of rushing yards gained
rush_tds	integer	Career rushing touchdowns
receptions	integer	The number of pass receptions. Lateral receptions officially don't count as reception.
rec_yards	integer	Career receiving yards
rec_tds	integer	Career receiving touchdowns
def_solo_tackles	integer	Career solo tackles
def_ints	integer	Career interceptions
def_sacks	double	Number of sacks form this player

Example

from sportsdataverse.nfl import load_nfl_draft_picks
picks = load_nfl_draft_picks()
picks.shape

# Filter to a single year and round

import polars as pl
r1_2024 = (
    load_nfl_draft_picks()
    .filter((pl.col("season") == 2024) & (pl.col("round") == 1))
)

load_nfl_espn_qbr(seasons: 'List[int]', summary_type: 'str' = 'season', return_as_pandas: 'bool' = False, *, source: 'str' = 'nflverse') -> 'pl.DataFrame'

Load ESPN Total QBR (Quarterback Rating) data going back to 2006.

Mirrors nflreadpy / nflreadr load_espn_qbr -- the lone nflreadpy dataset that previously had no sdv-py loader. ESPN publishes Total QBR only from 2006 onward, so 2006 is the earliest available season (unlike the 1999 floor on play-by-play). nflverse republishes ESPN's QBR through the espn_data release as two combined files (one per summary_type), each covering all seasons; this loader reads the requested file once and post-filters by season (the same access pattern as load_nfl_schedule).

Parameters

Parameter	Type	Default	Description
seasons	list		Seasons to return. 2006 is the earliest available season.
summary_type	str	'season'	Aggregation level. "season" (default) returns one row per quarterback-season; "week" returns one row per quarterback-game. Any other value raises ValueError.
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe.
source	str	'nflverse'	Which QBR release to read. "nflverse" (the default, also accepts None) returns the nflverse espn_data release. "sportsdataverse" / "sdv" returns the SDV-native nfl_espn_qbr release (built by nfl-data from ESPN's QBR web endpoint -- the same source nflverse's espnscrapeR uses). Any other value raises ValueError.

Returns

Polars dataframe containing ESPN Total QBR for the requested seasons, summarized per summary_type.

col_name	type	description
season	integer	NFL season (year) the Total QBR record covers.
season_type	character	Season segment for the record -- regular season or postseason.
game_week	character	Week scope of the QBR aggregation; for season-level rows this is the season-summary tag.
team_abb	character	Team abbreviation for the quarterback's team during the period.
player_id	character	ESPN athlete identifier for the quarterback.
name_short	character	Abbreviated display name of the quarterback (e.g. 'P. Mahomes').
rank	double	Quarterback's rank by Total QBR among qualified passers for the period.
qbr_total	double	ESPN Total QBR on a 0-100 scale -- the headline opponent-adjusted quarterback rating.
pts_added	double	Points the quarterback added versus a league-average passer (ESPN QBR points-added component).
qb_plays	double	Count of qualifying quarterback action plays used to compute QBR.
epa_total	double	Total expected points added across the quarterback's plays (ESPN QBR EPA component).
pass	double	QBR points contribution from pass plays.
run	double	QBR points contribution from designed runs and scrambles.
exp_sack	double	QBR points contribution adjustment from expected sacks.
penalty	double	QBR points contribution from penalties attributed to the quarterback.
qbr_raw	double	Raw (non-opponent-adjusted) QBR for the period.
sack	double	QBR points contribution from sacks taken.
name_first	character	Quarterback's first name.
name_last	character	Quarterback's last name.
name_display	character	Quarterback's full display name.
headshot_href	character	URL of the quarterback's ESPN headshot image.
team	character	Full team name for the quarterback's team during the period.
qualified	logical	Whether the quarterback met ESPN's minimum action-play threshold to qualify for ranking.

Example

from sportsdataverse.nfl import load_nfl_espn_qbr
qbr = load_nfl_espn_qbr(seasons=[2024])
qbr.shape

# Week-level QBR

qbr_week = load_nfl_espn_qbr(seasons=[2024], summary_type="week")

# Multi-season range

qbr = load_nfl_espn_qbr(seasons=range(2020, 2025))

# Pandas round-trip

qbr_pd = load_nfl_espn_qbr(seasons=[2024], return_as_pandas=True)
qbr_pd[["season", "team_abb", "qbr_total"]].head()

load_nfl_ff_opportunity(seasons: 'List[int]', stat_type: 'str' = 'weekly', model_version: 'str' = 'latest', return_as_pandas=False) -> 'pl.DataFrame'

Load NFL fantasy football opportunity data from ffverse/ffopportunity

Parameters

Parameter	Type	Default	Description
seasons	list		Used to define different seasons. 2006 is the earliest available season.
stat_type	str	'weekly'	One of "weekly", "pbp_pass", "pbp_rush". Defaults to "weekly".
model_version	str	'latest'	One of "latest", "v1.0.0". Defaults to "latest".
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe.

Returns

Polars dataframe containing fantasy football opportunity data for the requested seasons.

col_name	type	description
season	character	4 digit number indicating to which season(s) the specified timeframe belongs to.
posteam	character	String abbreviation for the team with possession.
week	double	Season week.
game_id	character	Ten digit identifier for NFL game.
player_id	character	Player ID (aka GSIS ID) as defined by nflreadr::load_rosters
full_name	character	Full name as per NFL.com
position	character	Primary position as reported by NFL.com
pass_attempt	double	Binary indicator for if the play was a pass attempt (includes sacks).
rec_attempt	double	Total number of targets for a given game
rush_attempt	double	Binary indicator for if the play was a run.
pass_air_yards	double	Total air yards thrown for a given game
rec_air_yards	double	Total air yards on receiving attempts for a given game
pass_completions	double	Number of successful completions for a given game
receptions	double	The number of pass receptions. Lateral receptions officially don't count as reception.
pass_completions_exp	double	Expected number of pass_completions in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
receptions_exp	double	Expected number of receptions in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
pass_yards_gained	double	Total passing yards gained for a given game
rec_yards_gained	double	Total receiving yards gained for a given game
rush_yards_gained	double	Total rushing yards gained for a given game
pass_yards_gained_exp	double	Expected number of pass_yards_gained in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
rec_yards_gained_exp	double	Expected number of rec_yards_gained in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
rush_yards_gained_exp	double	Expected number of rush_yards_gained in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
pass_touchdown	double	Binary indicator for if the play resulted in a passing TD.
rec_touchdown	double	Total receiving touchdowns
rush_touchdown	double	Binary indicator for if the play resulted in a rushing TD.
pass_touchdown_exp	double	Expected number of pass_touchdown in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
rec_touchdown_exp	double	Expected number of rec_touchdown in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
rush_touchdown_exp	double	Expected number of rush_touchdown in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
pass_two_point_conv	double	Number of successful passing two point conversions
rec_two_point_conv	double	Number of successful receiving two point conversions
rush_two_point_conv	double	Number of successful rushing two point conversions
pass_two_point_conv_exp	double	Expected number of pass_two_point_conv in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
rec_two_point_conv_exp	double	Expected number of rec_two_point_conv in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
rush_two_point_conv_exp	double	Expected number of rush_two_point_conv in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
pass_first_down	double	Number of passing first downs
rec_first_down	double	Number of receiving first downs
rush_first_down	double	Number of rushing first downs
pass_first_down_exp	double	Expected number of pass_first_down in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
rec_first_down_exp	double	Expected number of rec_first_down in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
rush_first_down_exp	double	Expected number of rush_first_down in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
pass_interception	double	Number of interceptions thrown
rec_interception	double	Number of interceptions on targets
pass_interception_exp	double	Expected number of pass_interception in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
rec_interception_exp	double	Expected number of rec_interception in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
rec_fumble_lost	double	Number of fumbles on receiving attempts
rush_fumble_lost	double	Number of fumbles on rushing attempts
pass_fantasy_points_exp	double	Expected number of pass_fantasy_points in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
rec_fantasy_points_exp	double	Expected number of rec_fantasy_points in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
rush_fantasy_points_exp	double	Expected number of rush_fantasy_points in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
pass_fantasy_points	double	Total fantasy points from passing, assuming 0.04 points per pass yard, 4 points per pass TD, -2 points per interception
rec_fantasy_points	double	Total fantasy points from receiving, assuming PPR scoring
rush_fantasy_points	double	Total fantasy points from rushing, assuming PPR scoring
total_yards_gained	double	Total scrimmage yards (sum of pass, rush, and receiving yards)
total_yards_gained_exp	double	Expected number of total_yards_gained in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
total_touchdown	double	Total touchdowns (sum of pass, rush, and receiving touchdowns)
total_touchdown_exp	double	Expected number of total_touchdown in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
total_first_down	double	Total first downs (sum of pass, rush, and receiving first downs)
total_first_down_exp	double	Expected number of total_first_down in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
total_fantasy_points	double	Total fantasy points (sum of pass, rush, and receiving fantasy points)
total_fantasy_points_exp	double	Expected number of total_fantasy_points in this game (weekly) or on this play (pbp_rush/pbp_pass) given situation
pass_completions_diff	double	Difference between actual and expected number of pass_completions - often interpreted as efficiency for a given play/game
receptions_diff	double	Difference between actual and expected number of receptions - often interpreted as efficiency for a given play/game
pass_yards_gained_diff	double	Difference between actual and expected number of pass_yards_gained - often interpreted as efficiency for a given play/game
rec_yards_gained_diff	double	Difference between actual and expected number of rec_yards_gained - often interpreted as efficiency for a given play/game
rush_yards_gained_diff	double	Difference between actual and expected number of rush_yards_gained - often interpreted as efficiency for a given play/game
pass_touchdown_diff	double	Difference between actual and expected number of pass_touchdown - often interpreted as efficiency for a given play/game
rec_touchdown_diff	double	Difference between actual and expected number of rec_touchdown - often interpreted as efficiency for a given play/game
rush_touchdown_diff	double	Difference between actual and expected number of rush_touchdown - often interpreted as efficiency for a given play/game
pass_two_point_conv_diff	double	Difference between actual and expected number of pass_two_point_conv - often interpreted as efficiency for a given play/game
rec_two_point_conv_diff	double	Difference between actual and expected number of rec_two_point_conv - often interpreted as efficiency for a given play/game
rush_two_point_conv_diff	double	Difference between actual and expected number of rush_two_point_conv - often interpreted as efficiency for a given play/game
pass_first_down_diff	double	Difference between actual and expected number of pass_first_down - often interpreted as efficiency for a given play/game
rec_first_down_diff	double	Difference between actual and expected number of rec_first_down - often interpreted as efficiency for a given play/game
rush_first_down_diff	double	Difference between actual and expected number of rush_first_down - often interpreted as efficiency for a given play/game
pass_interception_diff	double	Difference between actual and expected number of pass_interception - often interpreted as efficiency for a given play/game
rec_interception_diff	double	Difference between actual and expected number of rec_interception - often interpreted as efficiency for a given play/game
pass_fantasy_points_diff	double	Difference between actual and expected number of pass_fantasy_points - often interpreted as efficiency for a given play/game
rec_fantasy_points_diff	double	Difference between actual and expected number of rec_fantasy_points - often interpreted as efficiency for a given play/game
rush_fantasy_points_diff	double	Difference between actual and expected number of rush_fantasy_points - often interpreted as efficiency for a given play/game
total_yards_gained_diff	double	Difference between actual and expected number of total_yards_gained - often interpreted as efficiency for a given play/game
total_touchdown_diff	double	Difference between actual and expected number of total_touchdown - often interpreted as efficiency for a given play/game
total_first_down_diff	double	Difference between actual and expected number of total_first_down - often interpreted as efficiency for a given play/game
total_fantasy_points_diff	double	Difference between actual and expected number of total_fantasy_points - often interpreted as efficiency for a given play/game
pass_attempt_team	double	Team-level total pass_attempt for a game, summed across all plays/players for that team.
rec_attempt_team	double	Team-level total rec_attempt for a game, summed across all plays/players for that team.
rush_attempt_team	double	Team-level total rush_attempt for a game, summed across all plays/players for that team.
pass_air_yards_team	double	Team-level total pass_air_yards for a game, summed across all plays/players for that team.
rec_air_yards_team	double	Team-level total rec_air_yards for a game, summed across all plays/players for that team.
pass_completions_team	double	Team-level total pass_completions for a game, summed across all plays/players for that team.
receptions_team	double	Team-level total receptions for a game, summed across all plays/players for that team.
pass_completions_exp_team	double	Team-level total expected pass_completions_exp for a game, summed across all plays & players for that team.
receptions_exp_team	double	Team-level total expected receptions_exp for a game, summed across all plays & players for that team.
pass_yards_gained_team	double	Team-level total pass_yards_gained for a game, summed across all plays/players for that team.
rec_yards_gained_team	double	Team-level total rec_yards_gained for a game, summed across all plays/players for that team.
rush_yards_gained_team	double	Team-level total rush_yards_gained for a game, summed across all plays/players for that team.
pass_yards_gained_exp_team	double	Team-level total expected pass_yards_gained_exp for a game, summed across all plays & players for that team.
rec_yards_gained_exp_team	double	Team-level total expected rec_yards_gained_exp for a game, summed across all plays & players for that team.
rush_yards_gained_exp_team	double	Team-level total expected rush_yards_gained_exp for a game, summed across all plays & players for that team.
pass_touchdown_team	double	Team-level total pass_touchdown for a game, summed across all plays/players for that team.
rec_touchdown_team	double	Team-level total rec_touchdown for a game, summed across all plays/players for that team.
rush_touchdown_team	double	Team-level total rush_touchdown for a game, summed across all plays/players for that team.
pass_touchdown_exp_team	double	Team-level total expected pass_touchdown_exp for a game, summed across all plays & players for that team.
rec_touchdown_exp_team	double	Team-level total expected rec_touchdown_exp for a game, summed across all plays & players for that team.
rush_touchdown_exp_team	double	Team-level total expected rush_touchdown_exp for a game, summed across all plays & players for that team.
pass_two_point_conv_team	double	Team-level total pass_two_point_conv for a game, summed across all plays/players for that team.
rec_two_point_conv_team	double	Team-level total rec_two_point_conv for a game, summed across all plays/players for that team.
rush_two_point_conv_team	double	Team-level total rush_two_point_conv for a game, summed across all plays/players for that team.
pass_two_point_conv_exp_team	double	Team-level total expected pass_two_point_conv_exp for a game, summed across all plays & players for that team.
rec_two_point_conv_exp_team	double	Team-level total expected rec_two_point_conv_exp for a game, summed across all plays & players for that team.
rush_two_point_conv_exp_team	double	Team-level total expected rush_two_point_conv_exp for a game, summed across all plays & players for that team.
pass_first_down_team	double	Team-level total pass_first_down for a game, summed across all plays/players for that team.
rec_first_down_team	double	Team-level total rec_first_down for a game, summed across all plays/players for that team.
rush_first_down_team	double	Team-level total rush_first_down for a game, summed across all plays/players for that team.
pass_first_down_exp_team	double	Team-level total expected pass_first_down_exp for a game, summed across all plays & players for that team.
rec_first_down_exp_team	double	Team-level total expected rec_first_down_exp for a game, summed across all plays & players for that team.
rush_first_down_exp_team	double	Team-level total expected rush_first_down_exp for a game, summed across all plays & players for that team.
pass_interception_team	double	Team-level total pass_interception for a game, summed across all plays/players for that team.
rec_interception_team	double	Team-level total rec_interception for a game, summed across all plays/players for that team.
pass_interception_exp_team	double	Team-level total expected pass_interception_exp for a game, summed across all plays & players for that team.
rec_interception_exp_team	double	Team-level total expected rec_interception_exp for a game, summed across all plays & players for that team.
rec_fumble_lost_team	double	Team-level total rec_fumble_lost for a game, summed across all plays/players for that team.
rush_fumble_lost_team	double	Team-level total rush_fumble_lost for a game, summed across all plays/players for that team.
pass_fantasy_points_exp_team	double	Team-level total expected pass_fantasy_points_exp for a game, summed across all plays & players for that team.
rec_fantasy_points_exp_team	double	Team-level total expected rec_fantasy_points_exp for a game, summed across all plays & players for that team.
rush_fantasy_points_exp_team	double	Team-level total expected rush_fantasy_points_exp for a game, summed across all plays & players for that team.
pass_fantasy_points_team	double	Team-level total pass_fantasy_points for a game, summed across all plays/players for that team.
rec_fantasy_points_team	double	Team-level total rec_fantasy_points for a game, summed across all plays/players for that team.
rush_fantasy_points_team	double	Team-level total rush_fantasy_points for a game, summed across all plays/players for that team.
pass_completions_diff_team	double	Team-level difference between actual and expected number of pass_completions_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
receptions_diff_team	double	Team-level difference between actual and expected number of receptions_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
pass_yards_gained_diff_team	double	Team-level difference between actual and expected number of pass_yards_gained_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
rec_yards_gained_diff_team	double	Team-level difference between actual and expected number of rec_yards_gained_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
rush_yards_gained_diff_team	double	Team-level difference between actual and expected number of rush_yards_gained_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
pass_touchdown_diff_team	double	Team-level difference between actual and expected number of pass_touchdown_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
rec_touchdown_diff_team	double	Team-level difference between actual and expected number of rec_touchdown_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
rush_touchdown_diff_team	double	Team-level difference between actual and expected number of rush_touchdown_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
pass_two_point_conv_diff_team	double	Team-level difference between actual and expected number of pass_two_point_conv_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
rec_two_point_conv_diff_team	double	Team-level difference between actual and expected number of rec_two_point_conv_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
rush_two_point_conv_diff_team	double	Team-level difference between actual and expected number of rush_two_point_conv_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
pass_first_down_diff_team	double	Team-level difference between actual and expected number of pass_first_down_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
rec_first_down_diff_team	double	Team-level difference between actual and expected number of rec_first_down_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
rush_first_down_diff_team	double	Team-level difference between actual and expected number of rush_first_down_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
pass_interception_diff_team	double	Team-level difference between actual and expected number of pass_interception_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
rec_interception_diff_team	double	Team-level difference between actual and expected number of rec_interception_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
pass_fantasy_points_diff_team	double	Team-level difference between actual and expected number of pass_fantasy_points_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
rec_fantasy_points_diff_team	double	Team-level difference between actual and expected number of rec_fantasy_points_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
rush_fantasy_points_diff_team	double	Team-level difference between actual and expected number of rush_fantasy_points_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
total_yards_gained_team	double	Team-level total total_yards_gained for a game, summed across all plays/players for that team.
total_yards_gained_exp_team	double	Team-level total expected total_yards_gained_exp for a game, summed across all plays & players for that team.
total_yards_gained_diff_team	double	Team-level difference between actual and expected number of total_yards_gained_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
total_touchdown_team	double	Team-level total total_touchdown for a game, summed across all plays/players for that team.
total_touchdown_exp_team	double	Team-level total expected total_touchdown_exp for a game, summed across all plays & players for that team.
total_touchdown_diff_team	double	Team-level difference between actual and expected number of total_touchdown_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
total_first_down_team	double	Team-level total total_first_down for a game, summed across all plays/players for that team.
total_first_down_exp_team	double	Team-level total expected total_first_down_exp for a game, summed across all plays & players for that team.
total_first_down_diff_team	double	Team-level difference between actual and expected number of total_first_down_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.
total_fantasy_points_team	double	Team-level total total_fantasy_points for a game, summed across all plays/players for that team.
total_fantasy_points_exp_team	double	Team-level total expected total_fantasy_points_exp for a game, summed across all plays & players for that team.
total_fantasy_points_diff_team	double	Team-level difference between actual and expected number of total_fantasy_points_diff for a game, summed across all plays/players for that team. Often interpreted as team-level efficiency.

Example

from sportsdataverse.nfl import load_nfl_ff_opportunity
weekly = load_nfl_ff_opportunity(seasons=[2024])

# Pass play-by-play opportunity stats

pbp_pass = load_nfl_ff_opportunity(seasons=[2024], stat_type="pbp_pass")

# Rush play-by-play opportunity stats with pinned model version

pbp_rush = load_nfl_ff_opportunity(
    seasons=[2024], stat_type="pbp_rush", model_version="v1.0.0"
)

load_nfl_ff_playerids(return_as_pandas=False) -> 'pl.DataFrame'

Load fantasy football player IDs from DynastyProcess.com

Parameters

Parameter	Type	Default	Description
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe.

Returns

Polars dataframe containing fantasy football player ID mappings across platforms.

col_name	type	description
mfl_id	integer	MyFantasyLeague.com ID - this is the primary key for this table and is unique and complete. Usually an integer of 5 digits.
sportradar_id	character	SportRadar ID - often also called sportsdata_id by other services. A UUID.
fantasypros_id	character	FantasyPros.com ID - usually an integer of 5 digits.
gsis_id	character	Game Stats and Info Service ID: the primary ID for play-by-play data.
pff_id	character	Pro Football Focus ID - usually an integer with between 3 and 6 digits.
sleeper_id	integer	Sleeper ID - usually an integer with ~4 digits.
nfl_id	character	NFL ID of player (this is used in Big Data Bowl Data)
espn_id	integer	ESPN ID - usual format is an integer with ~5 digits
yahoo_id	character	Yahoo ID - usual format is an integer with ~5 digits
fleaflicker_id	character	Fleaflicker ID - usual format is an integer with ~4 digits. Fleaflicker API also has sportradar and that's generally preferred.
cbs_id	integer	CBS ID - usual format is an integer with ~ 7 digits.
pfr_id	character	Pro-Football-Reference ID for player
cfbref_id	character	College Football Reference ID - usual format is firstname-lastname-integer
rotowire_id	integer	Rotowire ID - usual format is an integer with ~four digits. Not to be confused with rotowire_id.
rotoworld_id	character	Rotoworld ID - usual format is an integer with ~four digits. Not to be confused with rotowire_id.
ktc_id	integer	KeepTradeCut ID - usual format is an integer with ~four digits.
stats_id	integer	Stats ID - usual format is five digit integer
stats_global_id	integer	Stats Global ID - usual format is a six digit integer
fantasy_data_id	integer	FantasyData ID - usual format five digit integer
swish_id	character	Player ID for Swish Analytics
name	character	Name, as reported by MFL but reordered into FirstName LastName instead of Last, First
merge_name	character	Name but formatted for name joins via ffscrapr::dp_cleannames() - coerced to lowercase, stripped of punctuation and suffixes, and common substitutions performed.
position	character	Primary position as reported by NFL.com
team	character	NFL team. Uses official abbreviations as per NFL.com
birthdate	character	Birthdate
age	double	Age as of last pipeline build, rounded to one decimal. Pipeline is built on a weekly basis.
draft_year	integer	Year that player was drafted
draft_round	integer	Round that player was drafted in
draft_pick	integer	Draft pick within round, i.e. 32nd pick of second round.
draft_ovr	integer	Overall draft pick selection. This can be a little bit patchy, since MFL does not report this number.
twitter_username	character	Official twitter handle, if known
height	integer	Official height, in inches
weight	integer	Official weight, in pounds
college	character	Official college (usually the last one attended)
db_season	integer	Year of database build. Previous years may also be available via dynastyprocess.

Example

from sportsdataverse.nfl import load_nfl_ff_playerids
ids = load_nfl_ff_playerids()
ids.shape

# Filter to active QBs

import polars as pl
qbs = (
    load_nfl_ff_playerids()
    .filter((pl.col("position") == "QB") & (pl.col("status") == "ACT"))
)

load_nfl_ff_rankings(type: 'str' = 'draft', kind: 'str' = None, return_as_pandas=False) -> 'pl.DataFrame'

Load fantasy football rankings and projections

Parameters

Parameter	Type	Default	Description
type	str	'draft'	Type of rankings to load. One of "draft" (current draft rankings), "week" (weekly rankings), or "all" (full historical rankings). Defaults to "draft". Kept for nflreadpy parity since its parameter is also called type; the forward-going preferred name is kind.
kind	str	None	Preferred parameter name. Same semantics and allowed values as type. If both are supplied, kind wins. If neither is supplied, defaults to "draft" via type.
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe.

Returns

Polars dataframe containing fantasy football rankings data.

col_name	type	description
fp_page	character	The relative url that the data was scraped from (add the prefix https://www.fantasypros.com/ to visit the page)
page_type	character	Two word identifier separated by a dash identifying the type of fantasy ranking (best = bestball; dynasty; redraft) and what position it applies to
ecr_type	character	A two letter identifier combining the ranking type (b = bestball; d = dynasty; r = redraft) and position type (o = overall; p = positional; sf = superflex; rk = rookie)
player	character	Player name
id	integer	ID of the player in the 'name' column.
pos	character	Position as tracked by FP
team	character	NFL team. Uses official abbreviations as per NFL.com
ecr	double	Average (mean) expert ranking for this player
sd	double	Standard deviation of expert rankings for this player
best	integer	The highest ranking given for this player by any one expert
worst	integer	The lowest ranking given for this player by any one expert
sportsdata_id	character	ID - also known as sportradar_id (they are equivalent!)
player_filename	character	base URL for this player on fantasypros.com
yahoo_id	character	Yahoo ID - usual format is an integer with ~5 digits
cbs_id	character	CBS ID - usual format is an integer with ~ 7 digits.
player_owned_avg	double	The average percentage this player is rostered across ESPN and Yahoo
player_owned_espn	character	The percentage that this player is rostered in ESPN leagues
player_owned_yahoo	character	The percentage that this player is rostered in Yahoo leagues
player_image_url	character	An image of the player
player_square_image_url	character	An square image of the player
rank_delta	integer	Change in ranks over a recent period
bye	integer	NFL bye week
mergename	character	Player name after being cleaned by dp_cleannames - generally strips punctuation and suffixes as well as performing common name substitutions.
scrape_date	character	Date this dataframe was last updated
tm	character	Team ID as used on MyFantasyLeague.com

Example

from sportsdataverse.nfl import load_nfl_ff_rankings
draft = load_nfl_ff_rankings(kind="draft")

# Weekly rankings

weekly = load_nfl_ff_rankings(kind="week")

# Full historical rankings (parquet)

history = load_nfl_ff_rankings(kind="all")

# nflreadpy-parity ``type=`` parameter (still supported)

draft = load_nfl_ff_rankings(type="draft")

load_nfl_nextgen_stats(seasons: 'List[int]', stat_type: 'str' = 'passing', return_as_pandas: 'bool' = False) -> 'pl.DataFrame'

Load NFL NextGen Stats data going back to 2016.

Unified loader that consolidates the per-stat-type NextGen Stats accessors. Mirrors the API surface of nflreadpy's load_nextgen_stats so downstream code can swap engines without changing call sites.

Parameters

Parameter	Type	Default	Description
seasons	list[int]		Seasons to filter to. The upstream parquet covers a single combined file per stat type — seasons is applied as a post-filter on the season column.
stat_type	str	'passing'	One of "passing", "rushing", "receiving". Defaults to "passing".
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe.

Returns

Polars dataframe containing NextGen Stats data for the requested stat_type and seasons.

col_name	type	description
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
season_type	character	REG or POST indicating if the timeframe belongs to regular or post season.
week	integer	Season week.
player_display_name	character	Full name of the player
player_position	character	Position of the player accordinng to NGS
team_abbr	character	Official team abbreveation
avg_time_to_throw	double	Average time elapsed from the time of snap to throw on every pass attempt for a passer (sacks excluded).
avg_completed_air_yards	double	Average air yards on completed passes
avg_intended_air_yards	double	Average air yards on all attempted passes
avg_air_yards_differential	double	Air Yards Differential is calculated by subtracting the passer's average Intended Air Yards from his average Completed Air Yards. This stat indicates if he is on average attempting deep passes than he on average completes.
aggressiveness	double	Aggressiveness tracks the amount of passing attempts a quarterback makes that are into tight coverage, where there is a defender within 1 yard or less of the receiver at the time of completion or incompletion. AGG is shown as a % of attempts into tight windows over all passing attempts.
max_completed_air_distance	double	Air Distance is the amount of yards the ball has traveled on a pass, from the point of release to the point of reception (as the crow flies). Unlike Air Yards, Air Distance measures the actual distance the passer throws the ball.
avg_air_yards_to_sticks	double	Air Yards to the Sticks shows the amount of Air Yards ahead or behind the first down marker on all attempts for a passer. The metric indicates if the passer is attempting his passes past the 1st down marker, or if he is relying on his skill position players to make yards after catch.
attempts	integer	The number of pass attempts as defined by the NFL.
pass_yards	integer	Number of yards gained on pass plays
pass_touchdowns	integer	Number of touchdowns scored on pass plays
interceptions	integer	The number of interceptions thrown.
passer_rating	double	Overall NFL passer rating
completions	integer	The number of completed passes.
completion_percentage	double	Percentage of completed passes
expected_completion_percentage	double	Using a passer's Completion Probability on every play, determine what a passer's completion percentage is expected to be.
completion_percentage_above_expectation	double	A passer's actual completion percentage compared to their Expected Completion Percentage.
avg_air_distance	double	A receiver's average depth of target
max_air_distance	double	A receiver's maximum depth of target
player_gsis_id	character	Unique identifier of the player
player_first_name	character	Player's first name
player_last_name	character	Player's last name
player_jersey_number	integer	Player's jersey number
player_short_name	character	Short version of player's name

Example

from sportsdataverse.nfl import load_nfl_nextgen_stats
ngs_pass = load_nfl_nextgen_stats(seasons=[2024], stat_type="passing")

# Rushing NextGen stats

ngs_rush = load_nfl_nextgen_stats(seasons=[2024], stat_type="rushing")

# Receiving NextGen stats with a follow-up filter

import polars as pl
ngs_rec = (
    load_nfl_nextgen_stats(seasons=[2024], stat_type="receiving")
    .filter(pl.col("week") > 0)
)

# Pandas round-trip

ngs_pd = load_nfl_nextgen_stats(
    seasons=[2024], stat_type="passing", return_as_pandas=True
)

load_nfl_ngs_passing(seasons: 'List[int]' = None, return_as_pandas: 'bool' = False) -> 'pl.DataFrame'

Deprecated alias for load_nfl_nextgen_stats(stat_type='passing').

Will be removed in a future release. Migrate callers to the unified load_nfl_nextgen_stats function.

Parameters

Parameter	Type	Default	Description
seasons	List[int]	None
return_as_pandas	bool	False

Returns

col_name	type	description
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
season_type	character	REG or POST indicating if the timeframe belongs to regular or post season.
week	integer	Season week.
player_display_name	character	Full name of the player
player_position	character	Position of the player accordinng to NGS
team_abbr	character	Official team abbreveation
avg_time_to_throw	double	Average time elapsed from the time of snap to throw on every pass attempt for a passer (sacks excluded).
avg_completed_air_yards	double	Average air yards on completed passes
avg_intended_air_yards	double	Average air yards on all attempted passes
avg_air_yards_differential	double	Air Yards Differential is calculated by subtracting the passer's average Intended Air Yards from his average Completed Air Yards. This stat indicates if he is on average attempting deep passes than he on average completes.
aggressiveness	double	Aggressiveness tracks the amount of passing attempts a quarterback makes that are into tight coverage, where there is a defender within 1 yard or less of the receiver at the time of completion or incompletion. AGG is shown as a % of attempts into tight windows over all passing attempts.
max_completed_air_distance	double	Air Distance is the amount of yards the ball has traveled on a pass, from the point of release to the point of reception (as the crow flies). Unlike Air Yards, Air Distance measures the actual distance the passer throws the ball.
avg_air_yards_to_sticks	double	Air Yards to the Sticks shows the amount of Air Yards ahead or behind the first down marker on all attempts for a passer. The metric indicates if the passer is attempting his passes past the 1st down marker, or if he is relying on his skill position players to make yards after catch.
attempts	integer	The number of pass attempts as defined by the NFL.
pass_yards	integer	Number of yards gained on pass plays
pass_touchdowns	integer	Number of touchdowns scored on pass plays
interceptions	integer	The number of interceptions thrown.
passer_rating	double	Overall NFL passer rating
completions	integer	The number of completed passes.
completion_percentage	double	Percentage of completed passes
expected_completion_percentage	double	Using a passer's Completion Probability on every play, determine what a passer's completion percentage is expected to be.
completion_percentage_above_expectation	double	A passer's actual completion percentage compared to their Expected Completion Percentage.
avg_air_distance	double	A receiver's average depth of target
max_air_distance	double	A receiver's maximum depth of target
player_gsis_id	character	Unique identifier of the player
player_first_name	character	Player's first name
player_last_name	character	Player's last name
player_jersey_number	integer	Player's jersey number
player_short_name	character	Short version of player's name

Example

from sportsdataverse.nfl import load_nfl_nextgen_stats
ngs = load_nfl_nextgen_stats(seasons=[2024], stat_type="passing")

load_nfl_ngs_receiving(seasons: 'List[int]' = None, return_as_pandas: 'bool' = False) -> 'pl.DataFrame'

Deprecated alias for load_nfl_nextgen_stats(stat_type='receiving').

Will be removed in a future release. Migrate callers to the unified load_nfl_nextgen_stats function.

Parameters

Parameter	Type	Default	Description
seasons	List[int]	None
return_as_pandas	bool	False

Returns

col_name	type	description
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
season_type	character	REG or POST indicating if the timeframe belongs to regular or post season.
week	integer	Season week.
player_display_name	character	Full name of the player
player_position	character	Position of the player accordinng to NGS
team_abbr	character	Official team abbreveation
avg_cushion	double	The distance (in yards) measured between a WR/TE and the defender they're lined up against at the time of snap on all targets.
avg_separation	double	The distance (in yards) measured between a WR/TE and the nearest defender at the time of catch or incompletion.
avg_intended_air_yards	double	Average air yards on all attempted passes
percent_share_of_intended_air_yards	double	The sum of the receivers total intended air yards (all attempts) over the sum of his team's total intended air yards. Represented as a percentage, this statistic represents how much of a team's deep yards does the player account for.
receptions	integer	The number of pass receptions. Lateral receptions officially don't count as reception.
targets	integer	The number of pass plays where the player was the targeted receiver.
catch_percentage	double	Percentage of caught passes relative to targets
yards	integer	The number of receiving yards
rec_touchdowns	integer	The number of touchdown receptions
avg_yac	double	Average yards gained after catch by a receiver.
avg_expected_yac	double	Average expected yards after catch, based on numerous factors using tracking data such as how open the receiver is, how fast they're traveling, how many defenders/blockers are in space, etc
avg_yac_above_expectation	double	A receiver's YAC compared to their Expected YAC.
player_gsis_id	character	Unique identifier of the player
player_first_name	character	Player's first name
player_last_name	character	Player's last name
player_jersey_number	integer	Player's jersey number
player_short_name	character	Short version of player's name

Example

from sportsdataverse.nfl import load_nfl_nextgen_stats
ngs = load_nfl_nextgen_stats(seasons=[2024], stat_type="receiving")

load_nfl_ngs_rushing(seasons: 'List[int]' = None, return_as_pandas: 'bool' = False) -> 'pl.DataFrame'

Deprecated alias for load_nfl_nextgen_stats(stat_type='rushing').

Will be removed in a future release. Migrate callers to the unified load_nfl_nextgen_stats function.

Parameters

Parameter	Type	Default	Description
seasons	List[int]	None
return_as_pandas	bool	False

Returns

col_name	type	description
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
season_type	character	REG or POST indicating if the timeframe belongs to regular or post season.
week	integer	Season week.
player_display_name	character	Full name of the player
player_position	character	Position of the player accordinng to NGS
team_abbr	character	Official team abbreveation
efficiency	double	Rushing efficiency is calculated by taking the total distance a player traveled on rushing plays as a ball carrier according to Next Gen Stats (measured in yards) per rushing yards gained. The lower the number, the more of a North/South runner.
percent_attempts_gte_eight_defenders	double	On every play, Next Gen Stats calculates how many defenders are stacked in the box at snap. Using that logic, DIB% calculates how often does a rusher see 8 or more defenders in the box against them.
avg_time_to_los	double	Next Gen Stats measures the amount of time a ball carrier spends (measured to the 10th of a second) before crossing the Line of Scrimmage. TLOS is the average time behind the LOS on all rushing plays where the player is the rusher.
rush_attempts	integer	The number of rushing attempts
rush_yards	integer	The number of rushing yards gained
avg_rush_yards	double	AVerage rush yards gained
rush_touchdowns	integer	The number of scored rushing touchdowns
player_gsis_id	character	Unique identifier of the player
player_first_name	character	Player's first name
player_last_name	character	Player's last name
player_jersey_number	integer	Player's jersey number
player_short_name	character	Short version of player's name
expected_rush_yards	double	Expected rushing yards based on Nextgenstats' Big Data Bowl model
rush_yards_over_expected	double	A rusher's rush yards gained compared to the expected rush yards
rush_yards_over_expected_per_att	double	Average rush yards above expectation
rush_pct_over_expected	double	Rushing percentage above expectation

Example

from sportsdataverse.nfl import load_nfl_nextgen_stats
ngs = load_nfl_nextgen_stats(seasons=[2024], stat_type="rushing")

load_nfl_officials(return_as_pandas=False) -> 'pl.DataFrame'

Load NFL Officials information

Parameters

Parameter	Type	Default	Description
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe.

Returns

Polars dataframe containing officials available.

col_name	type	description
game_id	character	Ten digit identifier for NFL game.
game_key	character	Unique nflverse game identifier linking the officiating record to a specific NFL game.
official_name	character	Official name.
position	character	Primary position as reported by NFL.com
jersey_number	integer	Jersey number. Often useful for joins by name/team/jersey.
official_id	character	Unique official / referee identifier.
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
season_type	character	REG or POST indicating if the timeframe belongs to regular or post season.
week	integer	Season week.

Example

from sportsdataverse.nfl import load_nfl_officials
officials = load_nfl_officials()
officials.shape

# Pandas round-trip

officials_pd = load_nfl_officials(return_as_pandas=True)
officials_pd.head()

load_nfl_pfr_advstats(seasons: 'List[int]', stat_type: 'str' = 'pass', summary_level: 'str' = 'week', return_as_pandas: 'bool' = False) -> 'pl.DataFrame'

Load Pro-Football Reference advanced statistics going back to 2018.

Unified loader that consolidates the per-stat-type / per-summary-level PFR advstats accessors. Mirrors the API surface of nflreadpy's load_pfr_advstats so downstream code can swap engines without changing call sites.

Parameters

Parameter	Type	Default	Description
seasons	list[int]		Seasons to load. For summary_level='week' this drives the per-season parquet fan-out; for summary_level='season' it post-filters the combined parquet by the season column.
stat_type	str	'pass'	One of "pass", "rush", "rec", "def". Defaults to "pass".
summary_level	str	'week'	One of "week" or "season". Defaults to "week".
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe.

Returns

Polars dataframe containing PFR advanced stats data for the requested stat_type, summary_level, and seasons.

col_name	type	description
game_id	character	Ten digit identifier for NFL game.
pfr_game_id	character	PFR game ID
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
week	integer	Season week.
game_type	character	The most recent game type of that season that a player appeared on the roster.
team	character	NFL team. Uses official abbreviations as per NFL.com
opponent	character	Opposing team of player
pfr_player_name	character	Player's name as recorded by PFR
pfr_player_id	character	ID from Pro Football Reference
passing_drops	double	Total number of catchable passes thrown by the passer that were dropped by receivers per Pro Football Reference.
passing_drop_pct	double	Percentage of a passer's catchable targets that were dropped by the intended receiver.
receiving_drop	double	Number of catchable targets the receiver dropped during the game or season period per Pro Football Reference.
receiving_drop_pct	double	Percentage of the receiver's catchable targets that were dropped during the game or season period.
passing_bad_throws	double	Total number of passes thrown by the passer that were classified as inaccurate or poor-quality throws per Pro Football Reference.
passing_bad_throw_pct	double	Percentage of a passer's attempts classified as bad throws (inaccurate, off-target, or uncatchable).
times_sacked	double	Total number of times the passer was sacked during the game or season period per Pro Football Reference.
times_blitzed	double	Number of times blitzed
times_hurried	double	Number of times hurried
times_hit	double	Number of times hit
times_pressured	double	Number of times pressured
times_pressured_pct	double	Percentage of the passer's dropbacks during which they faced pressure from the opposing defense.
def_times_blitzed	double	Number of times the defensive player sent additional rushers on a blitz during the game or season period.
def_times_hurried	double	Number of times the defensive player hurried or pressured the quarterback without recording a sack.
def_times_hitqb	double	Number of times the defensive player made contact with the quarterback (hit on the QB) during pass rushes.

Example

from sportsdataverse.nfl import load_nfl_pfr_advstats
pass_week = load_nfl_pfr_advstats(
    seasons=[2024], stat_type="pass", summary_level="week"
)

# Season-level rushing summaries (one row per player per season)

rush_season = load_nfl_pfr_advstats(
    seasons=[2024], stat_type="rush", summary_level="season"
)

# Defensive stats with a follow-up filter

import polars as pl
def_week = (
    load_nfl_pfr_advstats(seasons=[2024], stat_type="def", summary_level="week")
    .filter(pl.col("week") <= 8)
)

# Pandas round-trip

rec_pd = load_nfl_pfr_advstats(
    seasons=[2024],
    stat_type="rec",
    summary_level="season",
    return_as_pandas=True,
)

load_nfl_pfr_def(return_as_pandas: 'bool' = False) -> 'pl.DataFrame'

Deprecated alias for load_nfl_pfr_advstats(stat_type='def', summary_level='season').

Will be removed in a future release. Migrate callers to the unified load_nfl_pfr_advstats function.

Parameters

Parameter	Type	Default	Description
return_as_pandas	bool	False

Returns

col_name	type	description
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
player	character	Player name
pfr_id	character	Pro-Football-Reference ID for player
tm	character	Team ID as used on MyFantasyLeague.com
age	double	Age as of last pipeline build, rounded to one decimal. Pipeline is built on a weekly basis.
pos	character	Position as tracked by FP
g	double	Goals (skaters).
gs	double	Number of games the player started during the season or period covered by this row.
int	double	Binary flag for an interception.
tgt	double	Total number of times the player was the nearest defender on a pass attempt (targets in coverage) per Pro Football Reference.
cmp	double	Number of passes completed by the opposing quarterback when targeting the player in coverage.
cmp_percent	double	Completion percentage allowed by the player in coverage (completions divided by targets).
yds	double	Total passing yards allowed by the player in coverage.
yds_cmp	double	Average yards allowed per completion when the player was in coverage.
yds_tgt	double	Average yards allowed per target thrown at the player in coverage.
td	double	Number of touchdowns allowed by the player in coverage.
rat	double	Passer rating allowed by the player in coverage — the NFL passer rating of quarterbacks when targeting this defender.
dadot	double	Depth of target air yards on defended passes — average distance downfield at the point of the throw when the player was in coverage.
air	double	Total air yards (depth of target) on passes thrown at the player in coverage, as tracked by Pro Football Reference.
yac	double	Yards after catch allowed by the player — yards gained by receivers after the catch when the player was the nearest defender.
bltz	double	Number of snaps on which the player blitzed the quarterback, as recorded by Pro Football Reference.
hrry	double	Number of times the player hurried the opposing quarterback without recording a full sack, per Pro Football Reference.
qbkd	double	Number of times the player knocked down the quarterback, making contact after or during a pass attempt.
sk	double	Number of sacks recorded by the player, bringing the quarterback down behind the line of scrimmage.
prss	double	Number of times the player pressured the quarterback (combining sacks, hits, and hurries) per Pro Football Reference.
comb	double	Total combined tackles (solo plus assisted) recorded by the player per Pro Football Reference.
m_tkl	double	Number of missed tackles attributed to the player by Pro Football Reference.
m_tkl_percent	double	Percentage of tackle attempts the player missed out of total tackle opportunities.
loaded	character	Indicator or metadata field from the Pro Football Reference data load, typically flagging the data source state or row completeness.
bats	double	Number of passes batted down at the line of scrimmage by the player.

Example

from sportsdataverse.nfl import load_nfl_pfr_advstats
df = load_nfl_pfr_advstats(
    seasons=[2024], stat_type="def", summary_level="season"
)

load_nfl_pfr_pass(return_as_pandas: 'bool' = False) -> 'pl.DataFrame'

Deprecated alias for load_nfl_pfr_advstats(stat_type='pass', summary_level='season').

Will be removed in a future release. Migrate callers to the unified load_nfl_pfr_advstats function.

Parameters

Parameter	Type	Default	Description
return_as_pandas	bool	False

Returns

col_name	type	description
player	character	Player name
team	character	NFL team. Uses official abbreviations as per NFL.com
pass_attempts	double	Career pass attempts
throwaways	double	Throwaways
spikes	double	Spikes
drops	double	Throws dropped
drop_pct	double	Percent of throws dropped
bad_throws	double	Bad throws
bad_throw_pct	double	Percent of throws that were bad
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
pfr_id	character	Pro-Football-Reference ID for player
pocket_time	double	Average time in pocket
times_blitzed	double	Number of times blitzed
times_hurried	double	Number of times hurried
times_hit	double	Number of times hit
times_pressured	double	Number of times pressured
pressure_pct	double	Percent of the time pressured
batted_balls	double	Batted balls
on_tgt_throws	double	On target throws
on_tgt_pct	double	Percent of throws on target
rpo_plays	double	Number of RPO plays
rpo_yards	double	Yards on RPOs
rpo_pass_att	double	Number of pass attempts on RPOs
rpo_pass_yards	double	Passing yards on RPOs
rpo_rush_att	double	Rush attempts on RPOs
rpo_rush_yards	double	Rushing yards on RPOs
pa_pass_att	double	Play action pass attempts
pa_pass_yards	double	Play action passing yards
intended_air_yards	double	Total air yards on all pass attempts including incompletions, measuring aggregate downfield targeting intent from Pro Football Reference.
intended_air_yards_per_pass_attempt	double	Average intended air yards per pass attempt, capturing the passer's average depth of target regardless of completion outcome.
completed_air_yards	double	Total air yards on completed passes only, measuring how far the ball traveled downfield through the air to the point of completion.
completed_air_yards_per_completion	double	Average air yards per completed pass, representing the passer's typical depth of target on successful throws.
completed_air_yards_per_pass_attempt	double	Average completed air yards per pass attempt (including incompletions), a rate measure of downfield passing efficiency.
pass_yards_after_catch	double	Total yards gained by receivers after the catch, isolating the yards generated after initial ball reception from Pro Football Reference.
pass_yards_after_catch_per_completion	double	Average yards after catch per completion, measuring how much yardage receivers generate on the ground after catching the ball.
scrambles	double	Total number of quarterback scrambles (designed dropback converted to a run) recorded by Pro Football Reference.
scramble_yards_per_attempt	double	Average yards gained per scramble attempt by the quarterback, from Pro Football Reference advanced passing stats.

Example

from sportsdataverse.nfl import load_nfl_pfr_advstats
df = load_nfl_pfr_advstats(
    seasons=[2024], stat_type="pass", summary_level="season"
)

load_nfl_pfr_rec(return_as_pandas: 'bool' = False) -> 'pl.DataFrame'

Deprecated alias for load_nfl_pfr_advstats(stat_type='rec', summary_level='season').

Will be removed in a future release. Migrate callers to the unified load_nfl_pfr_advstats function.

Parameters

Parameter	Type	Default	Description
return_as_pandas	bool	False

Returns

col_name	type	description
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
player	character	Player name
pfr_id	character	Pro-Football-Reference ID for player
tm	character	Team ID as used on MyFantasyLeague.com
age	double	Age as of last pipeline build, rounded to one decimal. Pipeline is built on a weekly basis.
pos	character	Position as tracked by FP
g	double	Goals (skaters).
gs	double	Number of games the player started at a receiver position during the period covered.
tgt	double	Total number of times the player was the intended receiver on a pass attempt.
rec	double	Total receptions made by the player during the period covered.
yds	double	Total receiving yards gained by the player on all receptions.
td	double	Total receiving touchdowns scored by the player.
x1d	double	Number of receptions by the player that resulted in a first down.
ybc	double	Total yards the ball traveled in the air (before the catch) on receptions by the player.
ybc_r	double	Average air yards before the catch per reception.
yac	double	Total yards gained by the player after the catch.
yac_r	double	Average yards after the catch per reception.
adot	double	Average depth of target — mean air yards at point of throw on pass attempts directed at the receiver, per Pro Football Reference.
brk_tkl	double	Number of broken tackles credited to the player after a reception, per Pro Football Reference.
rec_br	double	Receptions per broken tackle — number of receptions for each broken tackle the player forced after the catch, per Pro Football Reference.
drop	double	Number of catchable passes the player dropped (failed to secure after the ball reached the receiver's hands).
drop_percent	double	Percentage of catchable targets that the player dropped.
int	double	Binary flag for an interception.
rat	double	Passer rating generated on passes thrown to the player — the NFL passer rating when the receiver is targeted.
loaded	character	Indicator or metadata field from the Pro Football Reference data load, flagging the row's data source state or completeness.

Example

from sportsdataverse.nfl import load_nfl_pfr_advstats
df = load_nfl_pfr_advstats(
    seasons=[2024], stat_type="rec", summary_level="season"
)

load_nfl_pfr_rush(return_as_pandas: 'bool' = False) -> 'pl.DataFrame'

Deprecated alias for load_nfl_pfr_advstats(stat_type='rush', summary_level='season').

Will be removed in a future release. Migrate callers to the unified load_nfl_pfr_advstats function.

Parameters

Parameter	Type	Default	Description
return_as_pandas	bool	False

Returns

col_name	type	description
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
player	character	Player name
pfr_id	character	Pro-Football-Reference ID for player
tm	character	Team ID as used on MyFantasyLeague.com
age	double	Age as of last pipeline build, rounded to one decimal. Pipeline is built on a weekly basis.
pos	character	Position as tracked by FP
g	double	Goals (skaters).
gs	double	Number of games started by the player during the period covered.
att	double	Total rushing attempts by the player during the period covered.
yds	double	Total rushing yards gained during the period covered.
td	double	Total rushing touchdowns scored during the period covered.
x1d	double	Number of first downs gained via rushing during the period covered.
ybc	double	Yards before contact accumulated on rushing plays, measuring yards gained in open field before being touched.
ybc_att	double	Yards before contact per rushing attempt.
yac	double	Yards after contact accumulated on rushing plays.
yac_att	double	Yards after contact per rushing attempt.
brk_tkl	double	Number of broken tackles recorded on rushing plays.
att_br	double	Rushing attempts per broken tackle, measuring how often the player required contact to break free.
loaded	character	Source or load-batch identifier indicating which data file or release this row was pulled from.

Example

from sportsdataverse.nfl import load_nfl_pfr_advstats
df = load_nfl_pfr_advstats(
    seasons=[2024], stat_type="rush", summary_level="season"
)

load_nfl_pfr_weekly_def(seasons: 'List[int]', return_as_pandas: 'bool' = False) -> 'pl.DataFrame'

Deprecated alias for load_nfl_pfr_advstats(stat_type='def', summary_level='week').

Will be removed in a future release. Migrate callers to the unified load_nfl_pfr_advstats function.

Parameters

Parameter	Type	Default	Description
seasons	List[int]
return_as_pandas	bool	False

Returns

col_name	type	description
game_id	character	Ten digit identifier for NFL game.
pfr_game_id	character	PFR game ID
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
week	integer	Season week.
game_type	character	The most recent game type of that season that a player appeared on the roster.
team	character	NFL team. Uses official abbreviations as per NFL.com
opponent	character	Opposing team of player
pfr_player_name	character	Player's name as recorded by PFR
pfr_player_id	character	ID from Pro Football Reference
def_ints	double	Career interceptions
def_targets	double	Number of passing attempts thrown at or into the coverage area of this defender during the week.
def_completions_allowed	double	Number of completions allowed by the defender on passes thrown into their coverage during the week.
def_completion_pct	double	Completion percentage allowed by the defender on targets thrown in their coverage during the week.
def_yards_allowed	double	Total receiving yards allowed by this defender in coverage during the week per Pro Football Reference.
def_yards_allowed_per_cmp	double	Receiving yards allowed per completion by this defender in coverage during the week.
def_yards_allowed_per_tgt	double	Receiving yards allowed per target thrown at this defender in coverage during the week.
def_receiving_td_allowed	double	Number of receiving touchdowns allowed by the defender while in coverage during the week.
def_passer_rating_allowed	double	NFL passer rating of quarterbacks when targeting this defender in coverage during the week.
def_adot	double	Average depth of target (in yards) against this defender on passing plays during the week.
def_air_yards_completed	double	Total air yards on completed passes allowed by the defender during the week.
def_yards_after_catch	double	Total yards gained by receivers after the catch on completions allowed by this defender during the week.
def_times_blitzed	double	Number of times this defender was sent as a blitzer on a passing play during the week.
def_times_hurried	double	Number of times this defender hurried the quarterback on a pass rush without recording a sack during the week.
def_times_hitqb	double	Number of times this defender made contact with the quarterback as part of a pass rush during the week.
def_sacks	double	Number of sacks form this player
def_pressures	double	Total number of quarterback pressures (hurries + hits + sacks) generated by the defender during the week.
def_tackles_combined	double	Total combined tackles (solo + assisted) recorded by the defender during the week per Pro Football Reference.
def_missed_tackles	double	Number of missed tackles recorded against this defender during the week per Pro Football Reference.
def_missed_tackle_pct	double	Percentage of the defender's tackle opportunities that resulted in a missed tackle during the week.

Example

from sportsdataverse.nfl import load_nfl_pfr_advstats
df = load_nfl_pfr_advstats(
    seasons=[2024], stat_type="def", summary_level="week"
)

load_nfl_pfr_weekly_pass(seasons: 'List[int]', return_as_pandas: 'bool' = False) -> 'pl.DataFrame'

Deprecated alias for load_nfl_pfr_advstats(stat_type='pass', summary_level='week').

Will be removed in a future release. Migrate callers to the unified load_nfl_pfr_advstats function.

Parameters

Parameter	Type	Default	Description
seasons	List[int]
return_as_pandas	bool	False

Returns

col_name	type	description
game_id	character	Ten digit identifier for NFL game.
pfr_game_id	character	PFR game ID
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
week	integer	Season week.
game_type	character	The most recent game type of that season that a player appeared on the roster.
team	character	NFL team. Uses official abbreviations as per NFL.com
opponent	character	Opposing team of player
pfr_player_name	character	Player's name as recorded by PFR
pfr_player_id	character	ID from Pro Football Reference
passing_drops	double	Number of catchable passes thrown by the quarterback that were dropped by receivers.
passing_drop_pct	double	Percentage of catchable targets that were dropped by receivers on passes thrown by the quarterback.
receiving_drop	double	Number of catchable targets the player (as receiver) dropped in this weekly game.
receiving_drop_pct	double	Percentage of catchable targets the player dropped in this weekly game.
passing_bad_throws	double	Number of pass attempts charted as poor or inaccurate throws by the quarterback in this weekly log.
passing_bad_throw_pct	double	Percentage of pass attempts that were charted as poor throws (inaccurate, off-target, or otherwise below expectation), per Pro Football Reference.
times_sacked	double	Number of times the quarterback was sacked in this weekly game log.
times_blitzed	double	Number of times blitzed
times_hurried	double	Number of times hurried
times_hit	double	Number of times hit
times_pressured	double	Number of times pressured
times_pressured_pct	double	Percentage of pass plays on which the quarterback faced defensive pressure in this weekly game.
def_times_blitzed	double	Number of pass plays on which the defense sent a blitz, as experienced by the quarterback in this weekly log.
def_times_hurried	double	Number of times the quarterback was hurried (pressured into an early throw) without being sacked or hit.
def_times_hitqb	double	Number of times the quarterback was hit by a defender on a pass play in this weekly game log.

Example

from sportsdataverse.nfl import load_nfl_pfr_advstats
df = load_nfl_pfr_advstats(
    seasons=[2024], stat_type="pass", summary_level="week"
)

load_nfl_pfr_weekly_rec(seasons: 'List[int]', return_as_pandas: 'bool' = False) -> 'pl.DataFrame'

Deprecated alias for load_nfl_pfr_advstats(stat_type='rec', summary_level='week').

Will be removed in a future release. Migrate callers to the unified load_nfl_pfr_advstats function.

Parameters

Parameter	Type	Default	Description
seasons	List[int]
return_as_pandas	bool	False

Returns

col_name	type	description
game_id	character	Ten digit identifier for NFL game.
pfr_game_id	character	PFR game ID
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
week	integer	Season week.
game_type	character	The most recent game type of that season that a player appeared on the roster.
team	character	NFL team. Uses official abbreviations as per NFL.com
opponent	character	Opposing team of player
pfr_player_name	character	Player's name as recorded by PFR
pfr_player_id	character	ID from Pro Football Reference
rushing_broken_tackles	double	Number of broken tackles recorded by the player on rushing plays during the week.
receiving_broken_tackles	double	Number of broken tackles recorded by the receiver after the catch.
passing_drops	double	Number of passes dropped by intended receivers on targeted throws.
passing_drop_pct	double	Percentage of pass targets that resulted in a drop, as tracked by Pro Football Reference.
receiving_drop	double	Number of catchable passes dropped by the receiver during the week.
receiving_drop_pct	double	Percentage of catchable targets that were dropped by the receiver during the week.
receiving_int	double	Number of passes intended for the receiver that were intercepted.
receiving_rat	double	Passer rating when targeting this receiver, reflecting QB efficiency on those throws.

Example

from sportsdataverse.nfl import load_nfl_pfr_advstats
df = load_nfl_pfr_advstats(
    seasons=[2024], stat_type="rec", summary_level="week"
)

load_nfl_pfr_weekly_rush(seasons: 'List[int]', return_as_pandas: 'bool' = False) -> 'pl.DataFrame'

Deprecated alias for load_nfl_pfr_advstats(stat_type='rush', summary_level='week').

Will be removed in a future release. Migrate callers to the unified load_nfl_pfr_advstats function.

Parameters

Parameter	Type	Default	Description
seasons	List[int]
return_as_pandas	bool	False

Returns

col_name	type	description
game_id	character	Ten digit identifier for NFL game.
pfr_game_id	character	PFR game ID
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
week	integer	Season week.
game_type	character	The most recent game type of that season that a player appeared on the roster.
team	character	NFL team. Uses official abbreviations as per NFL.com
opponent	character	Opposing team of player
pfr_player_name	character	Player's name as recorded by PFR
pfr_player_id	character	ID from Pro Football Reference
carries	double	The number of official rush attempts (incl. scrambles and kneel downs). Rushes after a lateral reception don't count as carry.
rushing_yards_before_contact	double	Total rushing yards gained by the player before making contact with a defender in this weekly game log.
rushing_yards_before_contact_avg	double	Average rushing yards gained before first contact per rushing attempt in this weekly game log.
rushing_yards_after_contact	double	Total rushing yards gained by the player after initial contact with a defender in this weekly game log.
rushing_yards_after_contact_avg	double	Average rushing yards gained after initial contact per rushing attempt in this weekly game log.
rushing_broken_tackles	double	Number of broken tackles the player recorded on rushing plays in this weekly game log.
receiving_broken_tackles	double	Number of broken tackles the player recorded after a reception in this weekly game log.

Example

from sportsdataverse.nfl import load_nfl_pfr_advstats
df = load_nfl_pfr_advstats(
    seasons=[2024], stat_type="rush", summary_level="week"
)

load_nfl_player_stats(kicking=False, return_as_pandas=False, *, source: 'str' = 'nflverse') -> 'pl.DataFrame'

Load NFL player stats data

One combined week-level parquet (all seasons, offense) mirroring nflverse's player_stats.

Parameters

Parameter	Type	Default	Description
kicking	bool	False	If True, load kicking stats. If False, load all other stats.
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe.
source	str	'nflverse'	Which player-stats release to read. "nflverse" (the default, also accepts None) returns the nflverse published player_stats.parquet. "sportsdataverse" / "sdv" returns the SDV-native nfl_player_stats release built by sportsdataverse.nfl.build_nfl_player_stats from SDV-native play-by-play (1999-present, week-level, REG+POST). Any other value raises ValueError.

Returns

Polars dataframe containing player stats.

col_name	type	description
player_id	character	Player ID (aka GSIS ID) as defined by nflreadr::load_rosters
player_name	character	Full name of player
player_display_name	character	Full name of the player
position	character	Primary position as reported by NFL.com
position_group	character	Postion group of player as listed by NFL
headshot_url	character	A URL string that points to player photos used by NFL.com (or sometimes ESPN)
recent_team	character	Most recent team player appears in pbp with.
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
week	integer	Season week.
season_type	character	REG or POST indicating if the timeframe belongs to regular or post season.
opponent_team	character	Abbreviation or name of the opposing team faced by the player in a given game or week.
completions	integer	The number of completed passes.
attempts	integer	The number of pass attempts as defined by the NFL.
passing_yards	double	Numeric yards by the passer_player_name, including yards gained in pass plays with laterals. This should equal official passing statistics.
passing_tds	integer	The number of passing touchdowns.
interceptions	double	The number of interceptions thrown.
sacks	double	The Number of times sacked.
sack_yards	double	Yards lost on sack plays.
sack_fumbles	integer	The number of sacks with a fumble.
sack_fumbles_lost	integer	The number of sacks with a lost fumble.
passing_air_yards	double	Passing air yards (includes incomplete passes).
passing_yards_after_catch	double	Yards after the catch gained on plays in which player was the passer (this is an unofficial stat and may differ slightly between different sources).
passing_first_downs	double	First downs on pass attempts.
passing_epa	double	Total expected points added on pass attempts and sacks. NOTE: this uses the variable qb_epa, which gives QB credit for EPA for up to the point where a receiver lost a fumble after a completed catch and makes EPA work more like passing yards on plays with fumbles.
passing_2pt_conversions	integer	Two-point conversion passes.
pacr	double	Passing (yards) Air (yards) Conversion Ratio - the number of passing yards per air yards thrown per game
dakota	double	Adjusted EPA + CPOE composite based on coefficients which best predict adjusted EPA/play in the following year.
carries	integer	The number of official rush attempts (incl. scrambles and kneel downs). Rushes after a lateral reception don't count as carry.
rushing_yards	double	Numeric yards by the rusher_player_name, excluding yards gained in rush plays with laterals. This should equal official rushing statistics but could miss yards gained in rush plays with laterals. Please see the description of lateral_rusher_player_name for further information.
rushing_tds	integer	The number of rushing touchdowns (incl. scrambles). Also includes touchdowns after obtaining a lateral on a play that started with a rushing attempt.
rushing_fumbles	double	The number of rushes with a fumble.
rushing_fumbles_lost	double	The number of rushes with a lost fumble.
rushing_first_downs	double	First downs on rush attempts (incl. scrambles).
rushing_epa	double	Expected points added on rush attempts (incl. scrambles and kneel downs).
rushing_2pt_conversions	integer	Two-point conversion rushes
receptions	integer	The number of pass receptions. Lateral receptions officially don't count as reception.
targets	integer	The number of pass plays where the player was the targeted receiver.
receiving_yards	double	Numeric yards by the receiver_player_name, excluding yards gained in pass plays with laterals. This should equal official receiving statistics but could miss yards gained in pass plays with laterals. Please see the description of lateral_receiver_player_name for further information.
receiving_tds	integer	The number of touchdowns following a pass reception. Also includes touchdowns after receiving a lateral on a play that started as a pass play.
receiving_fumbles	double	The number of fumbles after a pass reception.
receiving_fumbles_lost	double	The number of fumbles lost after a pass reception.
receiving_air_yards	double	Receiving air yards (incl. incomplete passes).
receiving_yards_after_catch	double	Yards after the catch gained on plays in which player was receiver (this is an unofficial stat and may differ slightly between different sources).
receiving_first_downs	double	Total number of first downs gained on receptions
receiving_epa	double	Total EPA on plays where this receiver was targeted
receiving_2pt_conversions	integer	Two-point conversion receptions
racr	double	Receiving (yards) Air (yards) Conversion Ratio - the number of receiving yards per air yards targeted per game
target_share	double	"Player's share of team receiving targets in this game"
air_yards_share	double	Player's share of the team's air yards in this game
wopr	double	Weighted OPportunity Rating - 1.5 x target_share + 0.7 x air_yards_share - a weighted average that contextualizes total fantasy usage.
special_teams_tds	double	Total number of kick/punt return touchdowns
fantasy_points	double	Standard fantasy points.
fantasy_points_ppr	double	PPR fantasy points.

Example

from sportsdataverse.nfl import load_nfl_player_stats
stats = load_nfl_player_stats()
stats.shape

# SDV-native player stats (week-level, built from SDV play-by-play)

stats_sdv = load_nfl_player_stats(source="sdv")
stats_sdv.select(["season", "week", "player_id", "attempts"]).head()

# Kicking-only stats (nflverse source only)

kicking = load_nfl_player_stats(kicking=True)

# Filter to a single season after load

import polars as pl
stats_2024 = load_nfl_player_stats().filter(pl.col("season") == 2024)

load_nfl_players(return_as_pandas=False, *, source: 'str' = 'nflverse') -> 'pl.DataFrame'

Load the nflverse NFL player-identity master.

Reads nflverse's published players.parquet — a one-row-per-player identity master that is the union of seven upstream systems (GSIS, ESPN, NGS roster, Pro-Football-Reference, OverTheCap, PFF, and the Sleeper / Yahoo cross-walk). It is the canonical source for cross-system identifier columns (gsis_id, espn_id, pfr_id, pff_id, otc_id, smart_id, esb_id, nfl_id) plus name, position, physical, draft, and status fields.

This is the full identity master. For an SDV-native, public-source-only alternative that does not depend on the nflverse release, see sportsdataverse.nfl.build_nfl_players (ESPN-athletes tier only) and sportsdataverse.nfl.nfl_players_crosswalk (a thin ID-only slice of this same parquet).

Parameters

Parameter	Type	Default	Description
return_as_pandas	bool	False	If True, return a pandas.DataFrame; otherwise a polars.DataFrame (default).
source	str	'nflverse'	Which player-master release to read. "nflverse" (the default, also accepts None) returns the nflverse seven-system players.parquet identity master described above. "sportsdataverse" / "sdv" returns the SDV-native nfl_players release built by sportsdataverse.nfl.build_nfl_players from the public NFL Shield / ESPN-athletes surface only. The SDV tier is a partial build: its columns are a subset of nflverse's and cross-system IDs are sparser (notably pre-2016), though espn_id is populated. The default stays "nflverse". Any other value raises ValueError.

Returns

One-row-per-player identity master. return_as_pandas narrows the return to a pandas.DataFrame.

col_name	type	description
gsis_id	character	NFL Game Statistics & Information System player identifier, the canonical nflverse player key.
display_name	character	Player's full display name as published by nflverse.
common_first_name	character	Player's commonly used first name (the name they go by, which may differ from their legal first name).
first_name	character	Player's legal first name.
last_name	character	Player's last name.
short_name	character	Abbreviated name (typically first initial plus last name).
football_name	character	Player's preferred on-field name as used in broadcast and box-score contexts.
suffix	character	Generational or honorific name suffix (e.g., Jr., Sr., III), when present.
esb_id	character	Elias Sports Bureau player identifier.
nfl_id	character	NFL.com / Shield player identifier.
pfr_id	character	Pro-Football-Reference player identifier.
pff_id	character	Pro Football Focus player identifier.
otc_id	character	OverTheCap player identifier (salary-cap data source).
espn_id	character	ESPN athlete identifier.
smart_id	character	NFL SMART (Standard Media and Reference Table) globally unique player identifier.
birth_date	character	Player's date of birth (ISO YYYY-MM-DD).
position_group	character	Broad positional grouping the player belongs to (e.g., QB, RB, WR, DL).
position	character	Player's specific listed position abbreviation.
ngs_position_group	character	Positional grouping as classified by NFL Next Gen Stats.
ngs_position	character	Specific position as classified by NFL Next Gen Stats.
height	integer	Player's height in inches.
weight	integer	Player's listed weight in pounds.
headshot	character	URL to the player's official headshot image.
college_name	character	Name of the college the player attended.
college_conference	character	Athletic conference of the player's college.
jersey_number	character	Player's uniform / jersey number.
rookie_season	integer	Season (year) the player entered the league as a rookie.
last_season	integer	Most recent season (year) the player appeared on an NFL roster.
latest_team	character	Abbreviation of the most recent team the player was rostered on.
status	character	Player's current roster status (e.g., active, retired, free agent).
ngs_status	character	Player status as reported by NFL Next Gen Stats.
ngs_status_short_description	character	Short human-readable description of the NFL Next Gen Stats status.
years_of_experience	integer	Number of accrued NFL seasons of experience.
pff_position	character	Player's position as classified by Pro Football Focus.
pff_status	character	Player's status as classified by Pro Football Focus.
draft_year	integer	Year the player was selected in the NFL Draft (null if undrafted).
draft_round	integer	Round in which the player was drafted (null if undrafted).
draft_pick	integer	Overall pick number at which the player was drafted (null if undrafted).
draft_team	character	Abbreviation of the team that drafted the player (null if undrafted).

Example

from sportsdataverse.nfl import load_nfl_players
players = load_nfl_players()
print(players.shape)

# Pandas round-trip

players_pd = load_nfl_players(return_as_pandas=True)
players_pd.head()

# SDV-native player master (public Shield/ESPN-athletes build; subset of nflverse columns, sparser cross-IDs)

players_sdv = load_nfl_players(source="sdv")
players_sdv.select(["display_name", "position", "espn_id"]).head()

# Pipeline next step (one line)

import polars as pl
load_nfl_players().select(["gsis_id", "display_name", "position"]).head()

load_nfl_schedule(seasons: 'List[int]', return_as_pandas=False) -> 'pl.DataFrame'

Load NFL schedule data

Parameters

Parameter	Type	Default	Description
seasons	list		Used to define different seasons. 1999 is the earliest available season.
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe.

Returns

Polars dataframe containing the schedule for the requested seasons.

col_name	type	description
game_id	character	Ten digit identifier for NFL game.
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
game_type	character	The most recent game type of that season that a player appeared on the roster.
week	integer	Season week.
gameday	character	The date on which the game occurred.
weekday	character	The day of the week on which the game occcured.
gametime	character	The kickoff time of the game. This is represented in 24-hour time and the Eastern time zone, regardless of what time zone the game was being played in.
away_team	character	String abbreviation for the away team.
away_score	integer	The number of points the away team scored. Is NA for games which haven't yet been played.
home_team	character	The home team. Note that this contains the designated home team for games which no team is playing at home such as Super Bowls or NFL International games.
home_score	integer	The number of points the home team scored. Is NA for games which haven't yet been played.
location	character	Either Home if the home team is playing in their home stadium, or Neutral if the game is being played at a neutral location. This still shows as Home for games between the Giants and Jets even though they share the same home stadium.
result	integer	The number of points the home team scored minus the number of points the visiting team scored. Equals h_score - v_score. Is NA for games which haven't yet been played. Convenient for evaluating against the spread bets.
total	integer	The sum of each team's score in the game. Equals h_score + v_score. Is NA for games which haven't yet been played. Convenient for evaluating over/under total bets.
overtime	integer	Binary indicator of whether or not game went to overtime.
old_game_id	character	Legacy NFL game ID.
gsis	integer	The id of the game issued by the NFL Game Statistics & Information System.
nfl_detail_id	character	The id of the game issued by NFL Detail.
pfr	character	The id of the game issued by Pro-Football-Reference
pff	integer	The id of the game issued by Pro Football Focus
espn	character	The id of the game issued by ESPN
ftn	integer	FTN Data game identifier corresponding to this scheduled game.
away_rest	integer	Days of rest that the away team is coming off of.
home_rest	integer	Days of rest that the home team is coming off of.
away_moneyline	integer	Odds for away team to win the game.
home_moneyline	integer	Odds for home team to win the game.
spread_line	double	The closing spread line for the game. A positive number means the home team was favored by that many points, a negative number means the away team was favored by that many points. (Source: Pro-Football-Reference)
away_spread_odds	integer	Odds for away team to cover the spread.
home_spread_odds	integer	Odds for home team to cover the spread.
total_line	double	The closing total line for the game. (Source: Pro-Football-Reference)
under_odds	integer	Odds that total score of game would be under the total_line.
over_odds	integer	Odds that total score of game would be over the total_ine.
div_game	integer	Binary indicator of whether or not game was played by 2 teams in the same division.
roof	character	One of 'dome', 'outdoors', 'closed', 'open' indicating indicating the roof status of the stadium the game was played in. (Source: Pro-Football-Reference)
surface	character	What type of ground the game was played on. (Source: Pro-Football-Reference)
temp	integer	The temperature at the stadium only for 'roof' = 'outdoors' or 'open'.(Source: Pro-Football-Reference)
wind	integer	The speed of the wind in miles/hour only for 'roof' = 'outdoors' or 'open'. (Source: Pro-Football-Reference)
away_qb_id	character	GSIS Player ID for away team starting quarterback.
home_qb_id	character	GSIS Player ID for home team starting quarterback.
away_qb_name	character	Name of away team starting QB.
home_qb_name	character	Name of home team starting QB.
away_coach	character	First and last name of the away team coach. (Source: Pro-Football-Reference)
home_coach	character	First and last name of the home team coach. (Source: Pro-Football-Reference)
referee	character	Name of the game's referee (head official)
stadium_id	character	ID of the stadium the game was played in. (Source: Pro-Football-Reference)
stadium	character	Name of the stadium

Example

from sportsdataverse.nfl import load_nfl_schedule
schedule = load_nfl_schedule(seasons=[2024])
schedule.shape

# Multi-season range

schedule = load_nfl_schedule(seasons=range(2020, 2025))

# Filter to a single week

import polars as pl
week_one = load_nfl_schedule(seasons=[2024]).filter(pl.col("week") == 1)

# Pandas round-trip

schedule_pd = load_nfl_schedule(seasons=[2024], return_as_pandas=True)
schedule_pd[["game_id", "home_team", "away_team", "week"]].head()

load_nfl_team_stats(seasons: 'List[int]', summary_level: 'str' = 'week', return_as_pandas=False, *, source: 'str' = 'nflverse') -> 'pl.DataFrame'

Load NFL team stats data going back to 1999

Parameters

Parameter	Type	Default	Description
seasons	list		Used to define different seasons. 1999 is the earliest available season.
summary_level	str	'week'	Aggregation level. One of "week", "reg", "post", "reg+post". Defaults to "week". Ignored when source is the SDV-native release (a single week-level parquet covering all seasons; filter post-load).
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe.
source	str	'nflverse'	Which team-stats release to read. "nflverse" (the default) reads the per-season nflverse stats_team releases. "sportsdataverse" / "sdv" reads the SDV-native nfl_team_stats release (a single combined week-level parquet, built by sportsdataverse.nfl.build_nfl_team_stats from the SDV play-by-play and filtered to the requested seasons post-load).

Returns

Polars dataframe containing team stats available for the requested seasons.

col_name	type	description
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
week	integer	Season week.
team	character	NFL team. Uses official abbreviations as per NFL.com
season_type	character	REG or POST indicating if the timeframe belongs to regular or post season.
opponent_team	character	Team abbreviation or identifier of the opposing team faced during the game or period.
completions	integer	The number of completed passes.
attempts	integer	The number of pass attempts as defined by the NFL.
passing_yards	integer	Numeric yards by the passer_player_name, including yards gained in pass plays with laterals. This should equal official passing statistics.
passing_tds	integer	The number of passing touchdowns.
passing_interceptions	integer	Total number of interceptions thrown by the team's passers during the game or season period.
sacks_suffered	integer	Total number of times the team's quarterback was sacked by the opposing defense during the period.
sack_yards_lost	integer	Total offensive yards lost by the team on plays where the quarterback was sacked during the period.
sack_fumbles	integer	The number of sacks with a fumble.
sack_fumbles_lost	integer	The number of sacks with a lost fumble.
passing_air_yards	integer	Passing air yards (includes incomplete passes).
passing_yards_after_catch	integer	Yards after the catch gained on plays in which player was the passer (this is an unofficial stat and may differ slightly between different sources).
passing_first_downs	integer	First downs on pass attempts.
passing_epa	double	Total expected points added on pass attempts and sacks. NOTE: this uses the variable qb_epa, which gives QB credit for EPA for up to the point where a receiver lost a fumble after a completed catch and makes EPA work more like passing yards on plays with fumbles.
passing_cpoe	double	Completion percentage over expectation (CPOE) for the team's passing attack during the period, relative to a model-based baseline.
passing_2pt_conversions	integer	Two-point conversion passes.
carries	integer	The number of official rush attempts (incl. scrambles and kneel downs). Rushes after a lateral reception don't count as carry.
rushing_yards	integer	Numeric yards by the rusher_player_name, excluding yards gained in rush plays with laterals. This should equal official rushing statistics but could miss yards gained in rush plays with laterals. Please see the description of lateral_rusher_player_name for further information.
rushing_tds	integer	The number of rushing touchdowns (incl. scrambles). Also includes touchdowns after obtaining a lateral on a play that started with a rushing attempt.
rushing_fumbles	integer	The number of rushes with a fumble.
rushing_fumbles_lost	integer	The number of rushes with a lost fumble.
rushing_first_downs	integer	First downs on rush attempts (incl. scrambles).
rushing_epa	double	Expected points added on rush attempts (incl. scrambles and kneel downs).
rushing_2pt_conversions	integer	Two-point conversion rushes
receptions	integer	The number of pass receptions. Lateral receptions officially don't count as reception.
targets	integer	The number of pass plays where the player was the targeted receiver.
receiving_yards	integer	Numeric yards by the receiver_player_name, excluding yards gained in pass plays with laterals. This should equal official receiving statistics but could miss yards gained in pass plays with laterals. Please see the description of lateral_receiver_player_name for further information.
receiving_tds	integer	The number of touchdowns following a pass reception. Also includes touchdowns after receiving a lateral on a play that started as a pass play.
receiving_fumbles	integer	The number of fumbles after a pass reception.
receiving_fumbles_lost	integer	The number of fumbles lost after a pass reception.
receiving_air_yards	integer	Receiving air yards (incl. incomplete passes).
receiving_yards_after_catch	integer	Yards after the catch gained on plays in which player was receiver (this is an unofficial stat and may differ slightly between different sources).
receiving_first_downs	integer	Total number of first downs gained on receptions
receiving_epa	double	Total EPA on plays where this receiver was targeted
receiving_2pt_conversions	integer	Two-point conversion receptions
special_teams_tds	integer	Total number of kick/punt return touchdowns
def_tackles_solo	integer	Total number of solo tackles for this player
def_tackles_with_assist	integer	Number of tackles this player had with an assisted tackle
def_tackle_assists	integer	Number of assisted tackles for this player
def_tackles_for_loss	integer	Number of tackles for loss (TFL) for this player
def_tackles_for_loss_yards	integer	Yards lost from TFLs involving this player
def_fumbles_forced	integer	Number of times a fumble was forced from this player
def_sacks	double	Number of sacks form this player
def_sack_yards	double	Yards lost from sacks forced by this player
def_qb_hits	integer	Number of QB hits from this player (should not include plays where the QB was sacked)
def_interceptions	integer	Number of interceptions forced by this player
def_interception_yards	integer	yards gained/lost by interception returns from this player
def_pass_defended	integer	Number of passes defended/broken up by this player
def_tds	integer	Number of defensive touchdowns scored by this player
def_fumbles	integer	Number of fumbles by this player
def_safeties	integer	Number of safeties scored by the defense (opponent tackled in their own end zone) during the period.
misc_yards	integer	Miscellaneous yards not attributed to passing, rushing, or standard return categories during the period.
fumble_recovery_own	integer	Number of fumbles recovered by the team that were originally fumbled by their own players.
fumble_recovery_yards_own	integer	Total yards gained (or lost) on recoveries of the team's own fumbles during the period.
fumble_recovery_opp	integer	Number of fumbles recovered by the team that were originally lost by the opposing team.
fumble_recovery_yards_opp	integer	Total yards gained on returns of fumbles recovered from the opposing team during the period.
fumble_recovery_tds	integer	Number of touchdowns scored by the team on fumble recoveries during the game or season period.
penalties	integer	Total number of penalties.
penalty_yards	integer	Yards gained (or lost) by the posteam from the penalty.
timeouts	integer	Number of timeouts remaining or used by the team during the game or period.
punt_returns	integer	Number of punt returns.
punt_return_yards	integer	Team punt return yards.
kickoff_returns	integer	Total number of kickoff returns recorded by the team during the game or season period.
kickoff_return_yards	integer	Total yards gained by the team on kickoff returns during the game or season period.
fg_made	integer	TRUE when the field goal attempt was successful.
fg_att	integer	Total number of field goal attempts by the team's kicker during the game or season period.
fg_missed	integer	Total number of field goal attempts missed (not blocked) by the team's kicker during the period.
fg_blocked	integer	Total number of field goal attempts that were blocked by the opposing defense during the period.
fg_long	integer	Distance in yards of the longest successful field goal made by the team's kicker during the period.
fg_pct	double	Field goal percentage (0-1).
fg_made_0_19	integer	Number of successful field goals made from 0–19 yards during the game or season period.
fg_made_20_29	integer	Number of successful field goals made from 20–29 yards during the game or season period.
fg_made_30_39	integer	Number of successful field goals made from 30–39 yards during the game or season period.
fg_made_40_49	integer	Number of successful field goals made from 40–49 yards during the game or season period.
fg_made_50_59	integer	Number of successful field goals made from 50–59 yards during the game or season period.
fg_made_60_	integer	Number of successful field goals made from 60 yards or longer during the game or season period.
fg_missed_0_19	integer	Number of field goal attempts from 0–19 yards that were missed (not blocked) during the period.
fg_missed_20_29	integer	Number of field goal attempts from 20–29 yards that were missed (not blocked) during the period.
fg_missed_30_39	integer	Number of field goal attempts from 30–39 yards that were missed (not blocked) during the period.
fg_missed_40_49	integer	Number of field goal attempts from 40–49 yards that were missed (not blocked) during the period.
fg_missed_50_59	integer	Number of field goal attempts from 50–59 yards that were missed (not blocked) during the period.
fg_missed_60_	integer	Number of field goal attempts from 60 yards or longer that were missed (not blocked) during the period.
fg_made_list	character	List of distances (in yards) of each successful field goal made during the game or season period.
fg_missed_list	character	List of distances (in yards) of each missed field goal attempt during the game or season period.
fg_blocked_list	character	List of distances (in yards) of each blocked field goal attempt during the game or season period.
fg_made_distance	integer	Cumulative distance in yards of all successful field goals made during the game or season period.
fg_missed_distance	integer	Cumulative distance in yards of all missed field goal attempts during the game or season period.
fg_blocked_distance	integer	Distance (in yards) of field goal attempts that were blocked by the opposing defense during the period.
pat_made	integer	Total number of successful point-after-touchdown kicks made by the team's kicker during the period.
pat_att	integer	Total number of point-after-touchdown (PAT / extra point) attempts by the team's kicker during the period.
pat_missed	integer	Number of point-after-touchdown attempts that were missed (not blocked) during the period.
pat_blocked	integer	Number of point-after-touchdown attempts that were blocked by the opposing defense during the period.
pat_pct	double	Percentage of point-after-touchdown attempts that were successfully converted during the period.
gwfg_made	integer	Number of successful game-winning field goals made to secure a victory in the final moments.
gwfg_att	integer	Number of game-winning field goal attempts made in the final moments to win the game.
gwfg_missed	integer	Number of game-winning field goal attempts that were missed (no good) in the final moments.
gwfg_blocked	integer	Number of game-winning field goal attempts that were blocked by the opposing defense.
gwfg_distance	integer	Distance in yards of the game-winning field goal attempt (or attempts) during the period.

Example

from sportsdataverse.nfl import load_nfl_team_stats
weekly = load_nfl_team_stats(seasons=[2024])

# Regular-season-only team stats

reg = load_nfl_team_stats(seasons=[2024], summary_level="reg")

# SDV-native team stats (built from SDV play-by-play)

sdv = load_nfl_team_stats(seasons=[2024], source="sdv")

load_nfl_teams(return_as_pandas=False) -> 'pl.DataFrame'

Load NFL team ID information and logos

Parameters

Parameter	Type	Default	Description
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe.

Returns

Polars dataframe containing teams available.

col_name	type	description
team_abbr	character	Official team abbreveation
team_name	character	Full team display name (e.g. 'Las Vegas Aces').
team_id	integer	Unique team identifier.
team_nick	character	Team nickname (e.g., 'Chiefs', 'Eagles', 'Patriots') without the city or state prefix.
team_conf	character	Conference affiliation of the team (e.g., 'AFC' or 'NFC').
team_division	character	Division affiliation of the team (e.g., 'AFC North', 'NFC West').
team_color	character	Team primary color (hex without leading '#').
team_color2	character	Secondary brand color for the team in hexadecimal format (e.g., '#FFB612').
team_color3	character	Tertiary brand color for the team in hexadecimal format, used in alternate uniforms or accents.
team_color4	character	Quaternary brand color for the team in hexadecimal format, part of the team's full brand palette.
team_logo_wikipedia	character	URL to the team's primary logo image as hosted on Wikimedia Commons / Wikipedia.
team_logo_espn	character	URL to the team's primary logo image as hosted by ESPN.
team_wordmark	character	URL to the team's wordmark image (team name rendered in official typography without the primary logo mark).
team_conference_logo	character	URL to the logo image for the team's conference (AFC or NFC).
team_league_logo	character	URL to the NFL league logo image.
team_logo_squared	character	URL to a square-cropped version of the team's logo suitable for thumbnails and grid layouts.

Example

from sportsdataverse.nfl import load_nfl_teams
teams = load_nfl_teams()
teams.shape

# Pandas round-trip

teams_pd = load_nfl_teams(return_as_pandas=True)
teams_pd[["team_abbr", "team_name", "team_conf", "team_division"]].head()

load_nfl_trades(return_as_pandas=False) -> 'pl.DataFrame'

Load NFL trades data

Parameters

Parameter	Type	Default	Description
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe.

Returns

Polars dataframe containing NFL trade information.

col_name	type	description
trade_id	integer	ID of Trade
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
trade_date	character	Exact date that trade occurred
gave	character	Team that gave pick/player in row
received	character	Team that received pick/player in row
pick_season	integer	Draft in which traded pick was in
pick_round	integer	Round in which traded pick was in
pick_number	integer	Pick number of traded pick
conditional	integer	Binary indicator of whether or not traded pick was conditional
pfr_id	character	Pro-Football-Reference ID for player
pfr_name	character	Full name of traded player

Example

from sportsdataverse.nfl import load_nfl_trades
trades = load_nfl_trades()
trades.shape

# Filter to a single season

import polars as pl
trades_2024 = load_nfl_trades().filter(pl.col("season") == 2024)

load_officials(return_as_pandas=False) -> 'pl.DataFrame'

Load NFL Officials information

Parameters

Parameter	Type	Default	Description
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe.

Returns

Polars dataframe containing officials available.

col_name	type	description
game_id	character	Ten digit identifier for NFL game.
game_key	character	Unique numeric key assigned by the NFL to identify the specific game in official records.
official_name	character	Official name.
position	character	Primary position as reported by NFL.com
jersey_number	integer	Jersey number. Often useful for joins by name/team/jersey.
official_id	character	Unique official / referee identifier.
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
season_type	character	REG or POST indicating if the timeframe belongs to regular or post season.
week	integer	Season week.

Example

from sportsdataverse.nfl import load_nfl_officials
officials = load_nfl_officials()
officials.shape

# Pandas round-trip

officials_pd = load_nfl_officials(return_as_pandas=True)
officials_pd.head()

load_participation(seasons: 'List[int]', return_as_pandas=False) -> 'pl.DataFrame'

Load NFL play-by-play participation data for selected seasons

Parameters

Parameter	Type	Default	Description
seasons	list		Used to define different seasons. 2016 is the earliest available season.
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe.

Returns

Polars dataframe containing play-by-play participation data available for the requested seasons.

col_name	type	description
nflverse_game_id	character	nflverse identifier for games. Format is season, week, away_team, home_team
old_game_id	character	Legacy NFL game ID.
play_id	double	Numeric play id that when used with game_id and drive provides the unique identifier for a single play.
possession_team	character	String abbreviation for the team with possession.
offense_formation	character	Formation the offense lines up in to snap the ball.
offense_personnel	character	The positions of the offensive personnel lined up on the field for a play.
defenders_in_box	integer	Number of defensive players lined up in the box at the snap.
defense_personnel	character	The positions of the defensive personnel lined up on the field for a play.
number_of_pass_rushers	integer	Number of defensive player who rushed the passer.
players_on_play	character	A list of every player on the field for the play, by gsis_id
offense_players	character	A list of every offensive player on the field for the play, by gsis_id
defense_players	character	A list of every defensive player on the field for the play, by gsis_id
n_offense	integer	Number of offensive players on the field for the play
n_defense	integer	Number of defensive players on the field for the play
ngs_air_yards	double	Legacy column. For 2023 and prior years, reflects the distance (in yards) that the ball traveled in the air on a given passing play as tracked by NGS. Is NA for 2024 on--we advise instead using the air_yards column from nflreadr::load_pbp() moving forward.
time_to_throw	double	Duration (in seconds) between the time of the ball being snapped and the time of release of a pass attempt
was_pressure	logical	A boolean indicating whether or not the QB was pressured on a play
route	character	A string indicating the route the primary receiver on a play took. Has the following possible values: "CORNER", "DEEP OUT", "GO", "HITCH/CURL", "IN/DIG", "POST", "QUICK OUT", "SCREEN", "SHALLOW CROSS/DRAG", "SLANT", "SWING", "TEXAS/ANGLE", "WHEEL".
defense_man_zone_type	character	A string indicating whether the defense was in man or zone coverage on a play
defense_coverage_type	character	A string indicating what type of cover the defense was in on a play. Has one of the following values: "COVER_0", "COVER_1", "COVER_2", "2_MAN", "COVER_3", "COVER_4", "COVER_6", "COVER_9", "COMBO", "BLOWN".
offense_names	character	A string listing all of the names of offensive players in the order of their gsis_ids in offense_players.
defense_names	character	A string listing all of the names of defensive players in the order of their gsis_ids in defense_players.
offense_positions	character	A string listing all of the positions of offensive players in the order of their gsis_ids in offense_players.
defense_positions	character	A string listing all of the positions of defensive players in the order of their gsis_ids in defense_players.
offense_numbers	character	A string listing all of the numbers of offensive players in the order of their gsis_ids in offense_players.
defense_numbers	character	A string listing all of the numbers of defensive players in the order of their gsis_ids in defense_players.

Example

from sportsdataverse.nfl import load_nfl_pbp_participation
participation = load_nfl_pbp_participation(seasons=[2022])

# Multi-season range

participation = load_nfl_pbp_participation(seasons=range(2018, 2023))

load_pbp(seasons: 'List[int]', return_as_pandas=False, *, source: 'str' = 'nflverse') -> 'pl.DataFrame'

Load NFL play by play data going back to 1999

Parameters

Parameter	Type	Default	Description
seasons	list		Used to define different seasons. 1999 is the earliest available season.
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe.
source	str	'nflverse'	Which enriched play-by-play release to read. "nflverse" (the default, also accepts None) returns the nflverse/nflfastR play_by_play_{season}.parquet releases — full history from 1999, unchanged behavior. "sportsdataverse" / "sdv" returns the SDV-native nfl_model_pbp release: a Python-built, nflfastR-faithful enriched frame (ep/epa, wp/wpa/vegas_wp, cp/cpoe, xyac_*/air_epa) that covers the published seasons (2023+) and drops administrative / timeout rows for a clean modeling subset. Any other value raises ValueError.

Returns

Polars dataframe containing the play-by-plays available for the requested seasons.

col_name	type	description
play_id	double	Numeric play id that when used with game_id and drive provides the unique identifier for a single play.
game_id	character	Ten digit identifier for NFL game.
old_game_id	character	Legacy NFL game ID.
home_team	character	The home team. Note that this contains the designated home team for games which no team is playing at home such as Super Bowls or NFL International games.
away_team	character	String abbreviation for the away team.
season_type	character	REG or POST indicating if the timeframe belongs to regular or post season.
week	integer	Season week.
posteam	character	String abbreviation for the team with possession.
posteam_type	character	String indicating whether the posteam team is home or away.
defteam	character	String abbreviation for the team on defense.
side_of_field	character	String abbreviation for which team's side of the field the team with possession is currently on.
yardline_100	double	Numeric distance in the number of yards from the opponent's endzone for the posteam.
game_date	character	Date of the game.
quarter_seconds_remaining	double	Numeric seconds remaining in the quarter.
half_seconds_remaining	double	Numeric seconds remaining in the half.
game_seconds_remaining	double	Numeric seconds remaining in the game.
game_half	character	String indicating which half the play is in, either Half1, Half2, or Overtime.
quarter_end	double	Binary indicator for whether or not the row of the data is marking the end of a quarter.
drive	double	Numeric drive number in the game.
sp	double	Binary indicator for whether or not a score occurred on the play.
qtr	double	Quarter of the game (5 is overtime).
down	double	The down for the given play.
goal_to_go	integer	Binary indicator for whether or not the posteam is in a goal down situation.
time	character	Time at start of play provided in string format as minutes:seconds remaining in the quarter.
yrdln	character	String indicating the current field position for a given play.
ydstogo	double	Numeric yards in distance from either the first down marker or the endzone in goal down situations.
ydsnet	double	Numeric value for total yards gained on the given drive.
desc	character	Detailed string description for the given play.
play_type	character	String indicating the type of play: pass (includes sacks), run (includes scrambles), punt, field_goal, kickoff, extra_point, qb_kneel, qb_spike, no_play (timeouts and penalties), and missing for rows indicating end of play.
yards_gained	double	Numeric yards gained (or lost) by the possessing team, excluding yards gained via fumble recoveries and laterals.
shotgun	double	Binary indicator for whether or not the play was in shotgun formation.
no_huddle	double	Binary indicator for whether or not the play was in no_huddle formation.
qb_dropback	double	Binary indicator for whether or not the QB dropped back on the play (pass attempt, sack, or scrambled).
qb_kneel	double	Binary indicator for whether or not the QB took a knee.
qb_spike	double	Binary indicator for whether or not the QB spiked the ball.
qb_scramble	double	Binary indicator for whether or not the QB scrambled.
pass_length	character	String indicator for pass length: short or deep.
pass_location	character	String indicator for pass location: left, middle, or right.
air_yards	double	Numeric value for distance in yards perpendicular to the line of scrimmage at where the targeted receiver either caught or didn't catch the ball.
yards_after_catch	double	Numeric value for distance in yards perpendicular to the yard line where the receiver made the reception to where the play ended.
run_location	character	String indicator for location of run: left, middle, or right.
run_gap	character	String indicator for line gap of run: end, guard, or tackle
field_goal_result	character	String indicator for result of field goal attempt: made, missed, or blocked.
kick_distance	double	Numeric distance in yards for kickoffs, field goals, and punts.
extra_point_result	character	String indicator for the result of the extra point attempt: good, failed, blocked, safety (touchback in defensive endzone is 1 point apparently), or aborted.
two_point_conv_result	character	String indicator for result of two point conversion attempt: success, failure, safety (touchback in defensive endzone is 1 point apparently), or return.
home_timeouts_remaining	double	Numeric timeouts remaining in the half for the home team.
away_timeouts_remaining	double	Numeric timeouts remaining in the half for the away team.
timeout	double	Binary indicator for whether or not a timeout was called by either team.
timeout_team	character	String abbreviation for which team called the timeout.
td_team	character	String abbreviation for which team scored the touchdown.
td_player_name	character	String name of the player who scored a touchdown.
td_player_id	character	Unique identifier of the player who scored a touchdown.
posteam_timeouts_remaining	double	Number of timeouts remaining for the possession team.
defteam_timeouts_remaining	double	Number of timeouts remaining for the team on defense.
total_home_score	double	Score for the home team at the start of the play.
total_away_score	double	Score for the away team at the start of the play.
posteam_score	double	Score the posteam at the start of the play.
defteam_score	double	Score the defteam at the start of the play.
score_differential	double	Score differential between the posteam and defteam at the start of the play.
posteam_score_post	double	Score for the posteam at the end of the play.
defteam_score_post	double	Score for the defteam at the end of the play.
score_differential_post	double	Score differential between the posteam and defteam at the end of the play.
no_score_prob	double	Predicted probability of no score occurring for the rest of the half based on the expected points model.
opp_fg_prob	double	Predicted probability of the defteam scoring a FG next. 'Next' in this context means the next score in the same game half.
opp_safety_prob	double	Predicted probability of the defteam scoring a safety next. 'Next' in this context means the next score in the same game half.
opp_td_prob	double	Predicted probability of the defteam scoring a TD next. 'Next' in this context means the next score in the same game half.
fg_prob	double	Predicted probability of the posteam scoring a FG next. 'Next' in this context means the next score in the same game half.
safety_prob	double	Predicted probability of the posteam scoring a safety next. 'Next' in this context means the next score in the same game half.
td_prob	double	Predicted probability of the posteam scoring a TD next. 'Next' in this context means the next score in the same game half.
extra_point_prob	double	Predicted probability of the posteam scoring an extra point.
two_point_conversion_prob	double	Predicted probability of the posteam scoring the two point conversion.
ep	double	Using the scoring event probabilities, the estimated expected points with respect to the possession team for the given play.
epa	double	Expected points added (EPA) by the posteam for the given play.
total_home_epa	double	Cumulative total EPA for the home team in the game so far.
total_away_epa	double	Cumulative total EPA for the away team in the game so far.
total_home_rush_epa	double	Cumulative total rushing EPA for the home team in the game so far.
total_away_rush_epa	double	Cumulative total rushing EPA for the away team in the game so far.
total_home_pass_epa	double	Cumulative total passing EPA for the home team in the game so far.
total_away_pass_epa	double	Cumulative total passing EPA for the away team in the game so far.
air_epa	double	EPA from the air yards alone. For completions this represents the actual value provided through the air. For incompletions this represents the hypothetical value that could've been added through the air if the pass was completed.
yac_epa	double	EPA from the yards after catch alone. For completions this represents the actual value provided after the catch. For incompletions this represents the difference between the hypothetical air_epa and the play's raw observed EPA (how much the incomplete pass cost the posteam).
comp_air_epa	double	EPA from the air yards alone only for completions.
comp_yac_epa	double	EPA from the yards after catch alone only for completions.
total_home_comp_air_epa	double	Cumulative total completions air EPA for the home team in the game so far.
total_away_comp_air_epa	double	Cumulative total completions air EPA for the away team in the game so far.
total_home_comp_yac_epa	double	Cumulative total completions yac EPA for the home team in the game so far.
total_away_comp_yac_epa	double	Cumulative total completions yac EPA for the away team in the game so far.
total_home_raw_air_epa	double	Cumulative total raw air EPA for the home team in the game so far.
total_away_raw_air_epa	double	Cumulative total raw air EPA for the away team in the game so far.
total_home_raw_yac_epa	double	Cumulative total raw yac EPA for the home team in the game so far.
total_away_raw_yac_epa	double	Cumulative total raw yac EPA for the away team in the game so far.
wp	double	Estimated win probabiity for the posteam given the current situation at the start of the given play.
def_wp	double	Estimated win probability for the defteam.
home_wp	double	Estimated win probability for the home team.
away_wp	double	Estimated win probability for the away team.
wpa	double	Win probability added (WPA) for the posteam.
vegas_wpa	double	Win probability added (WPA) for the posteam: spread_adjusted model.
vegas_home_wpa	double	Win probability added (WPA) for the home team: spread_adjusted model.
home_wp_post	double	Estimated win probability for the home team at the end of the play.
away_wp_post	double	Estimated win probability for the away team at the end of the play.
vegas_wp	double	Estimated win probabiity for the posteam given the current situation at the start of the given play, incorporating pre-game Vegas line.
vegas_home_wp	double	Estimated win probability for the home team incorporating pre-game Vegas line.
total_home_rush_wpa	double	Cumulative total rushing WPA for the home team in the game so far.
total_away_rush_wpa	double	Cumulative total rushing WPA for the away team in the game so far.
total_home_pass_wpa	double	Cumulative total passing WPA for the home team in the game so far.
total_away_pass_wpa	double	Cumulative total passing WPA for the away team in the game so far.
air_wpa	double	WPA through the air (same logic as air_epa).
yac_wpa	double	WPA from yards after the catch (same logic as yac_epa).
comp_air_wpa	double	The air_wpa for completions only.
comp_yac_wpa	double	The yac_wpa for completions only.
total_home_comp_air_wpa	double	Cumulative total completions air WPA for the home team in the game so far.
total_away_comp_air_wpa	double	Cumulative total completions air WPA for the away team in the game so far.
total_home_comp_yac_wpa	double	Cumulative total completions yac WPA for the home team in the game so far.
total_away_comp_yac_wpa	double	Cumulative total completions yac WPA for the away team in the game so far.
total_home_raw_air_wpa	double	Cumulative total raw air WPA for the home team in the game so far.
total_away_raw_air_wpa	double	Cumulative total raw air WPA for the away team in the game so far.
total_home_raw_yac_wpa	double	Cumulative total raw yac WPA for the home team in the game so far.
total_away_raw_yac_wpa	double	Cumulative total raw yac WPA for the away team in the game so far.
punt_blocked	double	Binary indicator for if the punt was blocked.
first_down_rush	double	Binary indicator for if a running play converted the first down.
first_down_pass	double	Binary indicator for if a passing play converted the first down.
first_down_penalty	double	Binary indicator for if a penalty converted the first down.
third_down_converted	double	Binary indicator for if the first down was converted on third down.
third_down_failed	double	Binary indicator for if the posteam failed to convert first down on third down.
fourth_down_converted	double	Binary indicator for if the first down was converted on fourth down.
fourth_down_failed	double	Binary indicator for if the posteam failed to convert first down on fourth down.
incomplete_pass	double	Binary indicator for if the pass was incomplete.
touchback	double	Binary indicator for if a touchback occurred on the play.
interception	double	Binary indicator for if the pass was intercepted.
punt_inside_twenty	double	Binary indicator for if the punt ended inside the twenty yard line.
punt_in_endzone	double	Binary indicator for if the punt was in the endzone.
punt_out_of_bounds	double	Binary indicator for if the punt went out of bounds.
punt_downed	double	Binary indicator for if the punt was downed.
punt_fair_catch	double	Binary indicator for if the punt was caught with a fair catch.
kickoff_inside_twenty	double	Binary indicator for if the kickoff ended inside the twenty yard line.
kickoff_in_endzone	double	Binary indicator for if the kickoff was in the endzone.
kickoff_out_of_bounds	double	Binary indicator for if the kickoff went out of bounds.
kickoff_downed	double	Binary indicator for if the kickoff was downed.
kickoff_fair_catch	double	Binary indicator for if the kickoff was caught with a fair catch.
fumble_forced	double	Binary indicator for if the fumble was forced.
fumble_not_forced	double	Binary indicator for if the fumble was not forced.
fumble_out_of_bounds	double	Binary indicator for if the fumble went out of bounds.
solo_tackle	double	Binary indicator if the play had a solo tackle (could be multiple due to fumbles).
safety	double	Binary indicator for whether or not a safety occurred.
penalty	double	Binary indicator for whether or not a penalty occurred.
tackled_for_loss	double	Binary indicator for whether or not a tackle for loss on a run play occurred.
fumble_lost	double	Binary indicator for if the fumble was lost.
own_kickoff_recovery	double	Binary indicator for if the kicking team recovered the kickoff.
own_kickoff_recovery_td	double	Binary indicator for if the kicking team recovered the kickoff and scored a TD.
qb_hit	double	Binary indicator if the QB was hit on the play.
rush_attempt	double	Binary indicator for if the play was a run.
pass_attempt	double	Binary indicator for if the play was a pass attempt (includes sacks).
sack	double	Binary indicator for if the play ended in a sack.
touchdown	double	Binary indicator for if the play resulted in a TD.
pass_touchdown	double	Binary indicator for if the play resulted in a passing TD.
rush_touchdown	double	Binary indicator for if the play resulted in a rushing TD.
return_touchdown	double	Binary indicator for if the play resulted in a return TD. Returns may occur on any of: interception, fumble, kickoff, punt, or blocked kicks.
extra_point_attempt	double	Binary indicator for extra point attempt.
two_point_attempt	double	Binary indicator for two point conversion attempt.
field_goal_attempt	double	Binary indicator for field goal attempt.
kickoff_attempt	double	Binary indicator for kickoff.
punt_attempt	double	Binary indicator for punts.
fumble	double	Binary indicator for if a fumble occurred.
complete_pass	double	Binary indicator for if the pass was completed.
assist_tackle	double	Binary indicator for if an assist tackle occurred.
lateral_reception	double	Binary indicator for if a lateral occurred on the reception.
lateral_rush	double	Binary indicator for if a lateral occurred on a run.
lateral_return	double	Binary indicator for if a lateral occurred on a return. Returns may occur on any of: interception, fumble, kickoff, punt, or blocked kicks.
lateral_recovery	double	Binary indicator for if a lateral occurred on a fumble recovery.
passer_player_id	character	Unique identifier for the player that attempted the pass.
passer_player_name	character	String name for the player that attempted the pass.
passing_yards	double	Numeric yards by the passer_player_name, including yards gained in pass plays with laterals. This should equal official passing statistics.
receiver_player_id	character	Unique identifier for the receiver that was targeted on the pass.
receiver_player_name	character	String name for the targeted receiver.
receiving_yards	double	Numeric yards by the receiver_player_name, excluding yards gained in pass plays with laterals. This should equal official receiving statistics but could miss yards gained in pass plays with laterals. Please see the description of lateral_receiver_player_name for further information.
rusher_player_id	character	Unique identifier for the player that attempted the run.
rusher_player_name	character	String name for the player that attempted the run.
rushing_yards	double	Numeric yards by the rusher_player_name, excluding yards gained in rush plays with laterals. This should equal official rushing statistics but could miss yards gained in rush plays with laterals. Please see the description of lateral_rusher_player_name for further information.
lateral_receiver_player_id	character	Unique identifier for the player that received the last(!) lateral on a pass play.
lateral_receiver_player_name	character	String name for the player that received the last(!) lateral on a pass play. If there were multiple laterals in the same play, this will only be the last player who received a lateral. Please see https://github.com/mrcaseb/nfl-data/tree/master/data/lateral_yards for a list of plays where multiple players recorded lateral receiving yards.
lateral_receiving_yards	double	Numeric yards by the lateral_receiver_player_name in pass plays with laterals. Please see the description of lateral_receiver_player_name for further information.
lateral_rusher_player_id	character	Unique identifier for the player that received the last(!) lateral on a run play.
lateral_rusher_player_name	character	String name for the player that received the last(!) lateral on a run play. If there were multiple laterals in the same play, this will only be the last player who received a lateral. Please see https://github.com/mrcaseb/nfl-data/tree/master/data/lateral_yards for a list of plays where multiple players recorded lateral rushing yards.
lateral_rushing_yards	double	Numeric yards by the lateral_rusher_player_name in run plays with laterals. Please see the description of lateral_rusher_player_name for further information.
lateral_sack_player_id	character	Unique identifier for the player that received the lateral on a sack.
lateral_sack_player_name	character	String name for the player that received the lateral on a sack.
interception_player_id	character	Unique identifier for the player that intercepted the pass.
interception_player_name	character	String name for the player that intercepted the pass.
lateral_interception_player_id	character	Unique indentifier for the player that received the lateral on an interception.
lateral_interception_player_name	character	String name for the player that received the lateral on an interception.
punt_returner_player_id	character	Unique identifier for the punt returner.
punt_returner_player_name	character	String name for the punt returner.
lateral_punt_returner_player_id	character	Unique identifier for the player that received the lateral on a punt return.
lateral_punt_returner_player_name	character	String name for the player that received the lateral on a punt return.
kickoff_returner_player_name	character	String name for the kickoff returner.
kickoff_returner_player_id	character	Unique identifier for the kickoff returner.
lateral_kickoff_returner_player_id	character	Unique identifier for the player that received the lateral on a kickoff return.
lateral_kickoff_returner_player_name	character	String name for the player that received the lateral on a kickoff return.
punter_player_id	character	Unique identifier for the punter.
punter_player_name	character	String name for the punter.
kicker_player_name	character	String name for the kicker on FG or kickoff.
kicker_player_id	character	Unique identifier for the kicker on FG or kickoff.
own_kickoff_recovery_player_id	character	Unique identifier for the player that recovered their own kickoff.
own_kickoff_recovery_player_name	character	String name for the player that recovered their own kickoff.
blocked_player_id	character	Unique identifier for the player that blocked the punt or FG.
blocked_player_name	character	String name for the player that blocked the punt or FG.
tackle_for_loss_1_player_id	character	Unique identifier for one of the potential players with the tackle for loss.
tackle_for_loss_1_player_name	character	String name for one of the potential players with the tackle for loss.
tackle_for_loss_2_player_id	character	Unique identifier for one of the potential players with the tackle for loss.
tackle_for_loss_2_player_name	character	String name for one of the potential players with the tackle for loss.
qb_hit_1_player_id	character	Unique identifier for one of the potential players that hit the QB. No sack as the QB was not the ball carrier. For sacks please see sack_player or half_sack_*_player.
qb_hit_1_player_name	character	String name for one of the potential players that hit the QB. No sack as the QB was not the ball carrier. For sacks please see sack_player or half_sack_*_player.
qb_hit_2_player_id	character	Unique identifier for one of the potential players that hit the QB. No sack as the QB was not the ball carrier. For sacks please see sack_player or half_sack_*_player.
qb_hit_2_player_name	character	String name for one of the potential players that hit the QB. No sack as the QB was not the ball carrier. For sacks please see sack_player or half_sack_*_player.
forced_fumble_player_1_team	character	Team of one of the players with a forced fumble.
forced_fumble_player_1_player_id	character	Unique identifier of one of the players with a forced fumble.
forced_fumble_player_1_player_name	character	String name of one of the players with a forced fumble.
forced_fumble_player_2_team	character	Team of one of the players with a forced fumble.
forced_fumble_player_2_player_id	character	Unique identifier of one of the players with a forced fumble.
forced_fumble_player_2_player_name	character	String name of one of the players with a forced fumble.
solo_tackle_1_team	character	Team of one of the players with a solo tackle.
solo_tackle_2_team	character	Team of one of the players with a solo tackle.
solo_tackle_1_player_id	character	Unique identifier of one of the players with a solo tackle.
solo_tackle_2_player_id	character	Unique identifier of one of the players with a solo tackle.
solo_tackle_1_player_name	character	String name of one of the players with a solo tackle.
solo_tackle_2_player_name	character	String name of one of the players with a solo tackle.
assist_tackle_1_player_id	character	Unique identifier of one of the players with a tackle assist.
assist_tackle_1_player_name	character	String name of one of the players with a tackle assist.
assist_tackle_1_team	character	Team of one of the players with a tackle assist.
assist_tackle_2_player_id	character	Unique identifier of one of the players with a tackle assist.
assist_tackle_2_player_name	character	String name of one of the players with a tackle assist.
assist_tackle_2_team	character	Team of one of the players with a tackle assist.
assist_tackle_3_player_id	character	Unique identifier of one of the players with a tackle assist.
assist_tackle_3_player_name	character	String name of one of the players with a tackle assist.
assist_tackle_3_team	character	Team of one of the players with a tackle assist.
assist_tackle_4_player_id	character	Unique identifier of one of the players with a tackle assist.
assist_tackle_4_player_name	character	String name of one of the players with a tackle assist.
assist_tackle_4_team	character	Team of one of the players with a tackle assist.
tackle_with_assist	double	Binary indicator for if there has been a tackle with assist.
tackle_with_assist_1_player_id	character	Unique identifier of one of the players with a tackle with assist.
tackle_with_assist_1_player_name	character	String name of one of the players with a tackle with assist.
tackle_with_assist_1_team	character	Team of one of the players with a tackle with assist.
tackle_with_assist_2_player_id	character	Unique identifier of one of the players with a tackle with assist.
tackle_with_assist_2_player_name	character	String name of one of the players with a tackle with assist.
tackle_with_assist_2_team	character	Team of one of the players with a tackle with assist.
pass_defense_1_player_id	character	Unique identifier of one of the players with a pass defense.
pass_defense_1_player_name	character	String name of one of the players with a pass defense.
pass_defense_2_player_id	character	Unique identifier of one of the players with a pass defense.
pass_defense_2_player_name	character	String name of one of the players with a pass defense.
fumbled_1_team	character	Team of one of the first player with a fumble.
fumbled_1_player_id	character	Unique identifier of the first player who fumbled on the play.
fumbled_1_player_name	character	String name of one of the first player who fumbled on the play.
fumbled_2_player_id	character	Unique identifier of the second player who fumbled on the play.
fumbled_2_player_name	character	String name of one of the second player who fumbled on the play.
fumbled_2_team	character	Team of one of the second player with a fumble.
fumble_recovery_1_team	character	Team of one of the players with a fumble recovery.
fumble_recovery_1_yards	double	Yards gained by one of the players with a fumble recovery.
fumble_recovery_1_player_id	character	Unique identifier of one of the players with a fumble recovery.
fumble_recovery_1_player_name	character	String name of one of the players with a fumble recovery.
fumble_recovery_2_team	character	Team of one of the players with a fumble recovery.
fumble_recovery_2_yards	double	Yards gained by one of the players with a fumble recovery.
fumble_recovery_2_player_id	character	Unique identifier of one of the players with a fumble recovery.
fumble_recovery_2_player_name	character	String name of one of the players with a fumble recovery.
sack_player_id	character	Unique identifier of the player who recorded a solo sack.
sack_player_name	character	String name of the player who recorded a solo sack.
half_sack_1_player_id	character	Unique identifier of the first player who recorded half a sack.
half_sack_1_player_name	character	String name of the first player who recorded half a sack.
half_sack_2_player_id	character	Unique identifier of the second player who recorded half a sack.
half_sack_2_player_name	character	String name of the second player who recorded half a sack.
return_team	character	String abbreviation of the return team. Returns may occur on any of: interception, fumble, kickoff, punt, or blocked kicks.
return_yards	double	Yards gained by the return team. Returns may occur on any of: interception, fumble, kickoff, punt, or blocked kicks.
penalty_team	character	String abbreviation of the team with the penalty.
penalty_player_id	character	Unique identifier for the player with the penalty.
penalty_player_name	character	String name for the player with the penalty.
penalty_yards	double	Yards gained (or lost) by the posteam from the penalty.
replay_or_challenge	double	Binary indicator for whether or not a replay or challenge.
replay_or_challenge_result	character	String indicating the result of the replay or challenge.
penalty_type	character	String indicating the penalty type of the first penalty in the given play. Will be NA if desc is missing the type.
defensive_two_point_attempt	double	Binary indicator whether or not the defense was able to have an attempt on a two point conversion, this results following a turnover.
defensive_two_point_conv	double	Binary indicator whether or not the defense successfully scored on the two point conversion.
defensive_extra_point_attempt	double	Binary indicator whether or not the defense was able to have an attempt on an extra point attempt, this results following a blocked attempt that the defense recovers the ball.
defensive_extra_point_conv	double	Binary indicator whether or not the defense successfully scored on an extra point attempt.
safety_player_name	character	String name for the player who scored a safety.
safety_player_id	character	Unique identifier for the player who scored a safety.
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
cp	double	Numeric value indicating the probability for a complete pass based on comparable game situations.
cpoe	double	For a single pass play this is 1 - cp when the pass was completed or 0 - cp when the pass was incomplete. Analyzed for a whole game or season an indicator for the passer how much over or under expectation his completion percentage was.
series	double	Starts at 1, each new first down increments, numbers shared across both teams NA: kickoffs, extra point/two point conversion attempts, non-plays, no posteam
series_success	double	1: scored touchdown, gained enough yards for first down.
series_result	character	Possible values: First down, Touchdown, Opp touchdown, Field goal, Missed field goal, Safety, Turnover, Punt, Turnover on downs, QB kneel, End of half
order_sequence	double	Column provided by NFL to fix out-of-order plays. Available 2011 and beyond with source "nfl".
start_time	character	Kickoff time in eastern time zone.
time_of_day	character	Time of day of play in UTC "HH:MM:SS" format. Available 2011 and beyond with source "nfl".
stadium	character	Name of the stadium
weather	character	String describing the weather including temperature, humidity and wind (direction and speed). Doesn't change during the game!
nfl_api_id	character	UUID of the game in the new NFL API.
play_clock	character	Time on the playclock when the ball was snapped.
play_deleted	double	Binary indicator for deleted plays.
play_type_nfl	character	Play type as listed in the NFL source. Slightly different to the regular play_type variable.
special_teams_play	double	Binary indicator for whether play is special teams play from NFL source. Available 2011 and beyond with source "nfl".
st_play_type	character	Type of special teams play from NFL source. Available 2011 and beyond with source "nfl".
end_clock_time	character	Game time at the end of a given play.
end_yard_line	character	String indicating the yardline at the end of the given play consisting of team half and yard line number.
fixed_drive	double	Manually created drive number in a game.
fixed_drive_result	character	Manually created drive result.
drive_real_start_time	character	Local day time when the drive started (currently not used by the NFL and therefore mostly 'NA').
drive_play_count	double	Numeric value of how many regular plays happened in a given drive.
drive_time_of_possession	character	Time of possession in a given drive.
drive_first_downs	double	Number of first downs in a given drive.
drive_inside20	double	Binary indicator if the offense was able to get inside the opponents 20 yard line.
drive_ended_with_score	double	Binary indicator the drive ended with a score.
drive_quarter_start	double	Numeric value indicating in which quarter the given drive has started.
drive_quarter_end	double	Numeric value indicating in which quarter the given drive has ended.
drive_yards_penalized	double	Numeric value of how many yards the offense gained or lost through penalties in the given drive.
drive_start_transition	character	String indicating how the offense got the ball.
drive_end_transition	character	String indicating how the offense lost the ball.
drive_game_clock_start	character	Game time at the beginning of a given drive.
drive_game_clock_end	character	Game time at the end of a given drive.
drive_start_yard_line	character	String indicating where a given drive started consisting of team half and yard line number.
drive_end_yard_line	character	String indicating where a given drive ended consisting of team half and yard line number.
drive_play_id_started	double	Play_id of the first play in the given drive.
drive_play_id_ended	double	Play_id of the last play in the given drive.
away_score	integer	The number of points the away team scored. Is NA for games which haven't yet been played.
home_score	integer	The number of points the home team scored. Is NA for games which haven't yet been played.
location	character	Either Home if the home team is playing in their home stadium, or Neutral if the game is being played at a neutral location. This still shows as Home for games between the Giants and Jets even though they share the same home stadium.
result	integer	The number of points the home team scored minus the number of points the visiting team scored. Equals h_score - v_score. Is NA for games which haven't yet been played. Convenient for evaluating against the spread bets.
total	integer	The sum of each team's score in the game. Equals h_score + v_score. Is NA for games which haven't yet been played. Convenient for evaluating over/under total bets.
spread_line	double	The closing spread line for the game. A positive number means the home team was favored by that many points, a negative number means the away team was favored by that many points. (Source: Pro-Football-Reference)
total_line	double	The closing total line for the game. (Source: Pro-Football-Reference)
div_game	integer	Binary indicator of whether or not game was played by 2 teams in the same division.
roof	character	One of 'dome', 'outdoors', 'closed', 'open' indicating indicating the roof status of the stadium the game was played in. (Source: Pro-Football-Reference)
surface	character	What type of ground the game was played on. (Source: Pro-Football-Reference)
temp	integer	The temperature at the stadium only for 'roof' = 'outdoors' or 'open'.(Source: Pro-Football-Reference)
wind	integer	The speed of the wind in miles/hour only for 'roof' = 'outdoors' or 'open'. (Source: Pro-Football-Reference)
home_coach	character	First and last name of the home team coach. (Source: Pro-Football-Reference)
away_coach	character	First and last name of the away team coach. (Source: Pro-Football-Reference)
stadium_id	character	ID of the stadium the game was played in. (Source: Pro-Football-Reference)
game_stadium	character	Name of the stadium the game was played in. (Source: Pro-Football-Reference)
aborted_play	double	Binary indicator if the play description indicates "Aborted".
success	double	Binary indicator wheter epa > 0 in the given play.
passer	character	Name of the dropback player (scrambles included) including plays with penalties.
passer_jersey_number	integer	Jersey number of the passer.
rusher	character	Name of the rusher (no scrambles) including plays with penalties.
rusher_jersey_number	integer	Jersey number of the rusher.
receiver	character	Name of the receiver including plays with penalties.
receiver_jersey_number	integer	Jersey number of the receiver.
pass	double	Binary indicator if the play was a pass play (sacks and scrambles included).
rush	double	Binary indicator if the play was a rushing play.
first_down	double	Binary indicator if the play ended in a first down.
special	double	Binary indicator if "play_type" is one of "extra_point", "field_goal", "kickoff", or "punt".
play	double	Binary indicator: 1 if the play was a 'normal' play (including penalties), 0 otherwise.
passer_id	character	ID of the player in the 'passer' column.
rusher_id	character	ID of the player in the 'rusher' column.
receiver_id	character	ID of the player in the 'receiver' column.
name	character	Name, as reported by MFL but reordered into FirstName LastName instead of Last, First
jersey_number	integer	Jersey number. Often useful for joins by name/team/jersey.
id	character	ID of the player in the 'name' column.
fantasy_player_name	character	Name of the rusher on rush plays or receiver on pass plays (from official stats).
fantasy_player_id	character	ID of the rusher on rush plays or receiver on pass plays (from official stats).
fantasy	character	Name of the rusher on rush plays or receiver on pass plays.
fantasy_id	character	ID of the rusher on rush plays or receiver on pass plays.
out_of_bounds	double	1 if play description contains ran ob, pushed ob, or sacked ob; 0 otherwise.
home_opening_kickoff	double	1 if the home team received the opening kickoff, 0 otherwise.
qb_epa	double	Gives QB credit for EPA for up to the point where a receiver lost a fumble after a completed catch and makes EPA work more like passing yards on plays with fumbles.
xyac_epa	double	Expected value of EPA gained after the catch, starting from where the catch was made. Zero yards after the catch would be listed as zero EPA.
xyac_mean_yardage	double	Average expected yards after the catch based on where the ball was caught.
xyac_median_yardage	integer	Median expected yards after the catch based on where the ball was caught.
xyac_success	double	Probability play earns positive EPA (relative to where play started) based on where ball was caught.
xyac_fd	double	Probability play earns a first down based on where the ball was caught.
xpass	double	Probability of dropback scaled from 0 to 1.
pass_oe	double	Dropback percent over expected on a given play scaled from 0 to 100.

Example

from sportsdataverse.nfl import load_nfl_pbp
pbp = load_nfl_pbp(seasons=[2024])
print(pbp.shape)

# Multi-season range

pbp = load_nfl_pbp(seasons=range(2020, 2025))

# SDV-native enriched play-by-play (Python-built, nflfastR-faithful EP/WP/CP/xYAC; published seasons only; admin/timeout rows dropped)

pbp_sdv = load_nfl_pbp(seasons=[2024], source="sdv")
pbp_sdv.select(["ep", "epa", "wp", "wpa", "cp", "cpoe", "xyac_epa"]).head()

# With cache off (development workflow)

from sportsdataverse.nfl import load_nfl_pbp, update_config
update_config(cache_mode="off")
pbp = load_nfl_pbp(seasons=[2024])

# Pandas round-trip

pbp_pd = load_nfl_pbp(seasons=[2024], return_as_pandas=True)
pbp_pd.head()

load_pfr_advstats(seasons: 'List[int]', stat_type: 'str' = 'pass', summary_level: 'str' = 'week', return_as_pandas: 'bool' = False) -> 'pl.DataFrame'

Load Pro-Football Reference advanced statistics going back to 2018.

Unified loader that consolidates the per-stat-type / per-summary-level PFR advstats accessors. Mirrors the API surface of nflreadpy's load_pfr_advstats so downstream code can swap engines without changing call sites.

Parameters

Parameter	Type	Default	Description
seasons	list[int]		Seasons to load. For summary_level='week' this drives the per-season parquet fan-out; for summary_level='season' it post-filters the combined parquet by the season column.
stat_type	str	'pass'	One of "pass", "rush", "rec", "def". Defaults to "pass".
summary_level	str	'week'	One of "week" or "season". Defaults to "week".
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe.

Returns

Polars dataframe containing PFR advanced stats data for the requested stat_type, summary_level, and seasons.

col_name	type	description
game_id	character	Ten digit identifier for NFL game.
pfr_game_id	character	PFR game ID
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
week	integer	Season week.
game_type	character	The most recent game type of that season that a player appeared on the roster.
team	character	NFL team. Uses official abbreviations as per NFL.com
opponent	character	Opposing team of player
pfr_player_name	character	Player's name as recorded by PFR
pfr_player_id	character	ID from Pro Football Reference
passing_drops	double	Raw count of catchable passes dropped by the intended receiver, from Pro Football Reference charting.
passing_drop_pct	double	Percentage of pass attempts dropped by the receiver, isolating receiver-side incompletions from passer error.
receiving_drop	double	Number of catchable targets dropped by the receiver in the given game or season, from Pro Football Reference advanced receiving data.
receiving_drop_pct	double	Percentage of targets that resulted in a drop by the receiver, from Pro Football Reference advanced receiving data.
passing_bad_throws	double	Raw count of bad throws by the passer, as charted and defined by Pro Football Reference advanced passing data.
passing_bad_throw_pct	double	Percentage of pass attempts classified as bad throws by Pro Football Reference (passes the passer should not have attempted or severely underthreww/overthrew).
times_sacked	double	Total number of times the defensive player recorded a sack of the quarterback, from Pro Football Reference.
times_blitzed	double	Number of times blitzed
times_hurried	double	Number of times hurried
times_hit	double	Number of times hit
times_pressured	double	Number of times pressured
times_pressured_pct	double	Percentage of pass-blocking snaps on which the lineman or back allowed the quarterback to be pressured, from Pro Football Reference.
def_times_blitzed	double	Number of plays on which the defensive player sent five or more pass rushers, from Pro Football Reference advanced defensive stats.
def_times_hurried	double	Number of times the defensive player hurried (pressured but did not sack or hit) the quarterback, from Pro Football Reference.
def_times_hitqb	double	Number of times the defensive player hit the quarterback on a pass play without recording a sack, from Pro Football Reference.

Example

from sportsdataverse.nfl import load_nfl_pfr_advstats
pass_week = load_nfl_pfr_advstats(
    seasons=[2024], stat_type="pass", summary_level="week"
)

# Season-level rushing summaries (one row per player per season)

rush_season = load_nfl_pfr_advstats(
    seasons=[2024], stat_type="rush", summary_level="season"
)

# Defensive stats with a follow-up filter

import polars as pl
def_week = (
    load_nfl_pfr_advstats(seasons=[2024], stat_type="def", summary_level="week")
    .filter(pl.col("week") <= 8)
)

# Pandas round-trip

rec_pd = load_nfl_pfr_advstats(
    seasons=[2024],
    stat_type="rec",
    summary_level="season",
    return_as_pandas=True,
)

load_player_stats(kicking=False, return_as_pandas=False, *, source: 'str' = 'nflverse') -> 'pl.DataFrame'

Load NFL player stats data

One combined week-level parquet (all seasons, offense) mirroring nflverse's player_stats.

Parameters

Parameter	Type	Default	Description
kicking	bool	False	If True, load kicking stats. If False, load all other stats.
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe.
source	str	'nflverse'	Which player-stats release to read. "nflverse" (the default, also accepts None) returns the nflverse published player_stats.parquet. "sportsdataverse" / "sdv" returns the SDV-native nfl_player_stats release built by sportsdataverse.nfl.build_nfl_player_stats from SDV-native play-by-play (1999-present, week-level, REG+POST). Any other value raises ValueError.

Returns

Polars dataframe containing player stats.

col_name	type	description
player_id	character	Player ID (aka GSIS ID) as defined by nflreadr::load_rosters
player_name	character	Full name of player
player_display_name	character	Full name of the player
position	character	Primary position as reported by NFL.com
position_group	character	Postion group of player as listed by NFL
headshot_url	character	A URL string that points to player photos used by NFL.com (or sometimes ESPN)
recent_team	character	Most recent team player appears in pbp with.
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
week	integer	Season week.
season_type	character	REG or POST indicating if the timeframe belongs to regular or post season.
opponent_team	character	Abbreviation of the opposing team the player faced in the game or week represented by this row.
completions	integer	The number of completed passes.
attempts	integer	The number of pass attempts as defined by the NFL.
passing_yards	double	Numeric yards by the passer_player_name, including yards gained in pass plays with laterals. This should equal official passing statistics.
passing_tds	integer	The number of passing touchdowns.
interceptions	double	The number of interceptions thrown.
sacks	double	The Number of times sacked.
sack_yards	double	Yards lost on sack plays.
sack_fumbles	integer	The number of sacks with a fumble.
sack_fumbles_lost	integer	The number of sacks with a lost fumble.
passing_air_yards	double	Passing air yards (includes incomplete passes).
passing_yards_after_catch	double	Yards after the catch gained on plays in which player was the passer (this is an unofficial stat and may differ slightly between different sources).
passing_first_downs	double	First downs on pass attempts.
passing_epa	double	Total expected points added on pass attempts and sacks. NOTE: this uses the variable qb_epa, which gives QB credit for EPA for up to the point where a receiver lost a fumble after a completed catch and makes EPA work more like passing yards on plays with fumbles.
passing_2pt_conversions	integer	Two-point conversion passes.
pacr	double	Passing (yards) Air (yards) Conversion Ratio - the number of passing yards per air yards thrown per game
dakota	double	Adjusted EPA + CPOE composite based on coefficients which best predict adjusted EPA/play in the following year.
carries	integer	The number of official rush attempts (incl. scrambles and kneel downs). Rushes after a lateral reception don't count as carry.
rushing_yards	double	Numeric yards by the rusher_player_name, excluding yards gained in rush plays with laterals. This should equal official rushing statistics but could miss yards gained in rush plays with laterals. Please see the description of lateral_rusher_player_name for further information.
rushing_tds	integer	The number of rushing touchdowns (incl. scrambles). Also includes touchdowns after obtaining a lateral on a play that started with a rushing attempt.
rushing_fumbles	double	The number of rushes with a fumble.
rushing_fumbles_lost	double	The number of rushes with a lost fumble.
rushing_first_downs	double	First downs on rush attempts (incl. scrambles).
rushing_epa	double	Expected points added on rush attempts (incl. scrambles and kneel downs).
rushing_2pt_conversions	integer	Two-point conversion rushes
receptions	integer	The number of pass receptions. Lateral receptions officially don't count as reception.
targets	integer	The number of pass plays where the player was the targeted receiver.
receiving_yards	double	Numeric yards by the receiver_player_name, excluding yards gained in pass plays with laterals. This should equal official receiving statistics but could miss yards gained in pass plays with laterals. Please see the description of lateral_receiver_player_name for further information.
receiving_tds	integer	The number of touchdowns following a pass reception. Also includes touchdowns after receiving a lateral on a play that started as a pass play.
receiving_fumbles	double	The number of fumbles after a pass reception.
receiving_fumbles_lost	double	The number of fumbles lost after a pass reception.
receiving_air_yards	double	Receiving air yards (incl. incomplete passes).
receiving_yards_after_catch	double	Yards after the catch gained on plays in which player was receiver (this is an unofficial stat and may differ slightly between different sources).
receiving_first_downs	double	Total number of first downs gained on receptions
receiving_epa	double	Total EPA on plays where this receiver was targeted
receiving_2pt_conversions	integer	Two-point conversion receptions
racr	double	Receiving (yards) Air (yards) Conversion Ratio - the number of receiving yards per air yards targeted per game
target_share	double	"Player's share of team receiving targets in this game"
air_yards_share	double	Player's share of the team's air yards in this game
wopr	double	Weighted OPportunity Rating - 1.5 x target_share + 0.7 x air_yards_share - a weighted average that contextualizes total fantasy usage.
special_teams_tds	double	Total number of kick/punt return touchdowns
fantasy_points	double	Standard fantasy points.
fantasy_points_ppr	double	PPR fantasy points.

Example

from sportsdataverse.nfl import load_nfl_player_stats
stats = load_nfl_player_stats()
stats.shape

# SDV-native player stats (week-level, built from SDV play-by-play)

stats_sdv = load_nfl_player_stats(source="sdv")
stats_sdv.select(["season", "week", "player_id", "attempts"]).head()

# Kicking-only stats (nflverse source only)

kicking = load_nfl_player_stats(kicking=True)

# Filter to a single season after load

import polars as pl
stats_2024 = load_nfl_player_stats().filter(pl.col("season") == 2024)

load_players(return_as_pandas=False, *, source: 'str' = 'nflverse') -> 'pl.DataFrame'

Load the nflverse NFL player-identity master.

Reads nflverse's published players.parquet — a one-row-per-player identity master that is the union of seven upstream systems (GSIS, ESPN, NGS roster, Pro-Football-Reference, OverTheCap, PFF, and the Sleeper / Yahoo cross-walk). It is the canonical source for cross-system identifier columns (gsis_id, espn_id, pfr_id, pff_id, otc_id, smart_id, esb_id, nfl_id) plus name, position, physical, draft, and status fields.

This is the full identity master. For an SDV-native, public-source-only alternative that does not depend on the nflverse release, see sportsdataverse.nfl.build_nfl_players (ESPN-athletes tier only) and sportsdataverse.nfl.nfl_players_crosswalk (a thin ID-only slice of this same parquet).

Parameters

Parameter	Type	Default	Description
return_as_pandas	bool	False	If True, return a pandas.DataFrame; otherwise a polars.DataFrame (default).
source	str	'nflverse'	Which player-master release to read. "nflverse" (the default, also accepts None) returns the nflverse seven-system players.parquet identity master described above. "sportsdataverse" / "sdv" returns the SDV-native nfl_players release built by sportsdataverse.nfl.build_nfl_players from the public NFL Shield / ESPN-athletes surface only. The SDV tier is a partial build: its columns are a subset of nflverse's and cross-system IDs are sparser (notably pre-2016), though espn_id is populated. The default stays "nflverse". Any other value raises ValueError.

Returns

One-row-per-player identity master. return_as_pandas narrows the return to a pandas.DataFrame.

col_name	type	description
gsis_id	character	NFL Game Statistics & Information System player identifier, the canonical nflverse player key.
display_name	character	Player's full display name as published by nflverse.
common_first_name	character	Player's commonly used first name (the name they go by, which may differ from their legal first name).
first_name	character	Player's legal first name.
last_name	character	Player's last name.
short_name	character	Abbreviated name (typically first initial plus last name).
football_name	character	Player's preferred on-field name as used in broadcast and box-score contexts.
suffix	character	Generational or honorific name suffix (e.g., Jr., Sr., III), when present.
esb_id	character	Elias Sports Bureau player identifier.
nfl_id	character	NFL.com / Shield player identifier.
pfr_id	character	Pro-Football-Reference player identifier.
pff_id	character	Pro Football Focus player identifier.
otc_id	character	OverTheCap player identifier (salary-cap data source).
espn_id	character	ESPN athlete identifier.
smart_id	character	NFL SMART (Standard Media and Reference Table) globally unique player identifier.
birth_date	character	Player's date of birth (ISO YYYY-MM-DD).
position_group	character	Broad positional grouping the player belongs to (e.g., QB, RB, WR, DL).
position	character	Player's specific listed position abbreviation.
ngs_position_group	character	Positional grouping as classified by NFL Next Gen Stats.
ngs_position	character	Specific position as classified by NFL Next Gen Stats.
height	integer	Player's height in inches.
weight	integer	Player's listed weight in pounds.
headshot	character	URL to the player's official headshot image.
college_name	character	Name of the college the player attended.
college_conference	character	Athletic conference of the player's college.
jersey_number	character	Player's uniform / jersey number.
rookie_season	integer	Season (year) the player entered the league as a rookie.
last_season	integer	Most recent season (year) the player appeared on an NFL roster.
latest_team	character	Abbreviation of the most recent team the player was rostered on.
status	character	Player's current roster status (e.g., active, retired, free agent).
ngs_status	character	Player status as reported by NFL Next Gen Stats.
ngs_status_short_description	character	Short human-readable description of the NFL Next Gen Stats status.
years_of_experience	integer	Number of accrued NFL seasons of experience.
pff_position	character	Player's position as classified by Pro Football Focus.
pff_status	character	Player's status as classified by Pro Football Focus.
draft_year	integer	Year the player was selected in the NFL Draft (null if undrafted).
draft_round	integer	Round in which the player was drafted (null if undrafted).
draft_pick	integer	Overall pick number at which the player was drafted (null if undrafted).
draft_team	character	Abbreviation of the team that drafted the player (null if undrafted).

Example

from sportsdataverse.nfl import load_nfl_players
players = load_nfl_players()
print(players.shape)

# Pandas round-trip

players_pd = load_nfl_players(return_as_pandas=True)
players_pd.head()

# SDV-native player master (public Shield/ESPN-athletes build; subset of nflverse columns, sparser cross-IDs)

players_sdv = load_nfl_players(source="sdv")
players_sdv.select(["display_name", "position", "espn_id"]).head()

# Pipeline next step (one line)

import polars as pl
load_nfl_players().select(["gsis_id", "display_name", "position"]).head()

load_rosters(seasons: 'List[int]', return_as_pandas=False, *, source: 'str' = 'nflverse') -> 'pl.DataFrame'

Load NFL season roster data for the requested seasons.

Reads nflverse's published season-roster parquet (one row per player per season). nflverse's roster product is the union of three upstream tiers -- NFL Next Gen Stats (2016+), the credentialed NFL Data Exchange (2002-2015), and the public NFL Shield endpoint (all seasons) -- so it carries densely populated cross-system identifier columns (espn_id, sportradar_id, yahoo_id, pff_id, pfr_id, ...) alongside biographical and depth-chart fields. This is the richest roster surface; prefer it whenever a network round trip to nflverse is acceptable.

Parameters

Parameter	Type	Default	Description
seasons	list		Seasons to load (e.g. [2024] or range(2020, 2025)). A single int is accepted and wrapped. 1920 is the earliest available season.
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe (default).
source	str	'nflverse'	Which roster release to read. "nflverse" (the default, also accepts None) returns the nflverse season-roster releases described above -- the full multi-source product (1920+, densely populated cross-system IDs). "sportsdataverse" / "sdv" returns the SDV-native nfl_rosters release built by sportsdataverse.nfl.build_nfl_rosters from the public NFL Shield / ESPN surface only. The SDV tier is a partial build: its 30 columns are a subset of nflverse's 36, and cross-system IDs are sparser pre-2016. It covers only the published seasons (rosters 2022+). The default stays "nflverse". Any other value raises ValueError.

Returns

Polars dataframe of season rosters for the requested seasons (pandas.DataFrame when return_as_pandas=True).

col_name	type	description
season	integer	NFL season (year) the roster entry applies to.
team	character	Team abbreviation in the nflverse standard (relocations folded, e.g. 'OAK' -> 'LV', 'SD' -> 'LAC', 'STL' -> 'LA').
position	character	Position the player is listed at on the roster (e.g. 'QB', 'WR', 'CB').
depth_chart_position	character	Fine-grained depth-chart position label, which may differ from the broader position group.
jersey_number	integer	Uniform (jersey) number the player wears.
status	character	Roster status code for the player (e.g. 'ACT' active, 'INA' inactive, 'RES' reserve/injured).
full_name	character	Player's full display name.
first_name	character	Player's first (given) name.
last_name	character	Player's last (family) name.
birth_date	character	Player's date of birth (YYYY-MM-DD).
height	double	Player's height in inches.
weight	integer	Player's listed weight in pounds.
college	character	College or university the player attended.
gsis_id	character	NFL GSIS player identifier — the canonical nflverse player key used to join across datasets.
espn_id	character	ESPN player identifier for cross-system joins.
sportradar_id	character	Sportradar player identifier for cross-system joins.
yahoo_id	character	Yahoo Sports player identifier for cross-system joins.
rotowire_id	character	RotoWire player identifier for cross-system joins.
pff_id	character	Pro Football Focus (PFF) player identifier for cross-system joins.
pfr_id	character	Pro Football Reference (PFR) player identifier for cross-system joins.
fantasy_data_id	character	FantasyData player identifier for cross-system joins.
sleeper_id	character	Sleeper player identifier for cross-system joins.
years_exp	integer	Number of accrued NFL seasons of experience for the player.
headshot_url	character	URL of the player's headshot image.
ngs_position	character	Player's position as classified by NFL Next Gen Stats.
week	integer	Week of the season the roster snapshot applies to (weekly rosters only).
game_type	character	Type of game the roster snapshot applies to (e.g. 'REG', 'POST').
status_description_abbr	character	Abbreviated roster status description code from the source feed.
football_name	character	Player's preferred football (commonly used) first name.
esb_id	character	Elias Sports Bureau (ESB) player identifier used for official NFL record-keeping.
gsis_it_id	character	NFL GSIS internal tracking identifier for the player.
smart_id	character	NFL SMART player identifier (GUID) used across modern NFL data feeds.
entry_year	integer	Calendar year the player first entered the NFL.
rookie_year	integer	Calendar year of the player's rookie season.
draft_club	character	Team abbreviation of the club that drafted the player.
draft_number	integer	Overall pick number at which the player was selected in the NFL draft.

Example

from sportsdataverse.nfl import load_nfl_rosters
rosters = load_nfl_rosters(seasons=[2024])

# Multi-season range

rosters = load_nfl_rosters(seasons=range(2020, 2025))

# Filter to a single team

import polars as pl
kc = load_nfl_rosters(seasons=[2024]).filter(pl.col("team") == "KC")

# SDV-native rosters (public Shield/ESPN build; published seasons 2022+; 30-column subset of nflverse, sparser cross-IDs pre-2016)

rosters_sdv = load_nfl_rosters(seasons=[2023], source="sdv")
rosters_sdv.select(["season", "team", "full_name", "gsis_id"]).head()

load_rosters_weekly(seasons: 'List[int]', return_as_pandas=False) -> 'pl.DataFrame'

Load NFL weekly roster data for the requested seasons.

Reads nflverse's published weekly-roster parquet (one row per player per team per week), so the roster snapshot reflects mid-season transactions (signings, releases, IR moves) rather than a single season-end view. Like load_nfl_rosters it is sourced from nflverse's full multi-tier roster product and carries densely populated cross-system identifier columns plus a week / game_type pair identifying each snapshot.

Unlike load_nfl_rosters and load_nfl_players, this loader has no SDV-native (source="sdv") tier: the SDV roster build (build_nfl_rosters) is season-only, and weekly snapshots require the credential-gated NFL Data Exchange that the public build cannot reach.

Parameters

Parameter	Type	Default	Description
seasons	list		Seasons to load (e.g. [2024] or range(2022, 2025)). A single int is accepted and wrapped. 2002 is the earliest available season.
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe (default).

Returns

Polars dataframe of weekly rosters for the requested seasons (pandas.DataFrame when return_as_pandas=True).

col_name	type	description
season	integer	NFL season (year) the weekly roster snapshot applies to.
team	character	Team abbreviation in the nflverse standard (relocations folded, e.g. 'OAK' -> 'LV', 'SD' -> 'LAC', 'STL' -> 'LA').
position	character	Position the player is listed at on the roster (e.g. 'QB', 'WR', 'CB').
depth_chart_position	character	Fine-grained depth-chart position label, which may differ from the broader position group.
jersey_number	integer	Uniform (jersey) number the player wears.
status	character	Roster status code for the player (e.g. 'ACT' active, 'INA' inactive, 'RES' reserve/injured).
full_name	character	Player's full display name.
first_name	character	Player's first (given) name.
last_name	character	Player's last (family) name.
birth_date	character	Player's date of birth (YYYY-MM-DD).
height	double	Player's height in inches.
weight	integer	Player's listed weight in pounds.
college	character	College or university the player attended.
gsis_id	character	NFL GSIS player identifier — the canonical nflverse player key used to join across datasets.
espn_id	character	ESPN player identifier for cross-system joins.
sportradar_id	character	Sportradar player identifier for cross-system joins.
yahoo_id	character	Yahoo Sports player identifier for cross-system joins.
rotowire_id	character	RotoWire player identifier for cross-system joins.
pff_id	character	Pro Football Focus (PFF) player identifier for cross-system joins.
pfr_id	character	Pro Football Reference (PFR) player identifier for cross-system joins.
fantasy_data_id	character	FantasyData player identifier for cross-system joins.
sleeper_id	character	Sleeper player identifier for cross-system joins.
years_exp	integer	Number of accrued NFL seasons of experience for the player.
headshot_url	character	URL of the player's headshot image.
ngs_position	character	Player's position as classified by NFL Next Gen Stats.
week	integer	Week of the season the weekly roster snapshot applies to.
game_type	character	Type of game the weekly roster snapshot applies to (e.g. 'REG', 'POST').
status_description_abbr	character	Abbreviated roster status description code from the source feed.
football_name	character	Player's preferred football (commonly used) first name.
esb_id	character	Elias Sports Bureau (ESB) player identifier used for official NFL record-keeping.
gsis_it_id	character	NFL GSIS internal tracking identifier for the player.
smart_id	character	NFL SMART player identifier (GUID) used across modern NFL data feeds.
entry_year	integer	Calendar year the player first entered the NFL.
rookie_year	integer	Calendar year of the player's rookie season.
draft_club	character	Team abbreviation of the club that drafted the player.
draft_number	integer	Overall pick number at which the player was selected in the NFL draft.

Example

from sportsdataverse.nfl import load_nfl_weekly_rosters
weekly = load_nfl_weekly_rosters(seasons=[2024])

# Multi-season range with a follow-up week filter

import polars as pl
wk1 = (
    load_nfl_weekly_rosters(seasons=range(2022, 2025))
    .filter(pl.col("week") == 1)
)

load_schedules(seasons: 'List[int]', return_as_pandas=False) -> 'pl.DataFrame'

Load NFL schedule data

Parameters

Parameter	Type	Default	Description
seasons	list		Used to define different seasons. 1999 is the earliest available season.
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe.

Returns

Polars dataframe containing the schedule for the requested seasons.

col_name	type	description
game_id	character	Ten digit identifier for NFL game.
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
game_type	character	The most recent game type of that season that a player appeared on the roster.
week	integer	Season week.
gameday	character	The date on which the game occurred.
weekday	character	The day of the week on which the game occcured.
gametime	character	The kickoff time of the game. This is represented in 24-hour time and the Eastern time zone, regardless of what time zone the game was being played in.
away_team	character	String abbreviation for the away team.
away_score	integer	The number of points the away team scored. Is NA for games which haven't yet been played.
home_team	character	The home team. Note that this contains the designated home team for games which no team is playing at home such as Super Bowls or NFL International games.
home_score	integer	The number of points the home team scored. Is NA for games which haven't yet been played.
location	character	Either Home if the home team is playing in their home stadium, or Neutral if the game is being played at a neutral location. This still shows as Home for games between the Giants and Jets even though they share the same home stadium.
result	integer	The number of points the home team scored minus the number of points the visiting team scored. Equals h_score - v_score. Is NA for games which haven't yet been played. Convenient for evaluating against the spread bets.
total	integer	The sum of each team's score in the game. Equals h_score + v_score. Is NA for games which haven't yet been played. Convenient for evaluating over/under total bets.
overtime	integer	Binary indicator of whether or not game went to overtime.
old_game_id	character	Legacy NFL game ID.
gsis	integer	The id of the game issued by the NFL Game Statistics & Information System.
nfl_detail_id	character	The id of the game issued by NFL Detail.
pfr	character	The id of the game issued by Pro-Football-Reference
pff	integer	The id of the game issued by Pro Football Focus
espn	character	The id of the game issued by ESPN
ftn	integer	FTN Data game identifier used to join schedule records with FTN charting and tracking data.
away_rest	integer	Days of rest that the away team is coming off of.
home_rest	integer	Days of rest that the home team is coming off of.
away_moneyline	integer	Odds for away team to win the game.
home_moneyline	integer	Odds for home team to win the game.
spread_line	double	The closing spread line for the game. A positive number means the home team was favored by that many points, a negative number means the away team was favored by that many points. (Source: Pro-Football-Reference)
away_spread_odds	integer	Odds for away team to cover the spread.
home_spread_odds	integer	Odds for home team to cover the spread.
total_line	double	The closing total line for the game. (Source: Pro-Football-Reference)
under_odds	integer	Odds that total score of game would be under the total_line.
over_odds	integer	Odds that total score of game would be over the total_ine.
div_game	integer	Binary indicator of whether or not game was played by 2 teams in the same division.
roof	character	One of 'dome', 'outdoors', 'closed', 'open' indicating indicating the roof status of the stadium the game was played in. (Source: Pro-Football-Reference)
surface	character	What type of ground the game was played on. (Source: Pro-Football-Reference)
temp	integer	The temperature at the stadium only for 'roof' = 'outdoors' or 'open'.(Source: Pro-Football-Reference)
wind	integer	The speed of the wind in miles/hour only for 'roof' = 'outdoors' or 'open'. (Source: Pro-Football-Reference)
away_qb_id	character	GSIS Player ID for away team starting quarterback.
home_qb_id	character	GSIS Player ID for home team starting quarterback.
away_qb_name	character	Name of away team starting QB.
home_qb_name	character	Name of home team starting QB.
away_coach	character	First and last name of the away team coach. (Source: Pro-Football-Reference)
home_coach	character	First and last name of the home team coach. (Source: Pro-Football-Reference)
referee	character	Name of the game's referee (head official)
stadium_id	character	ID of the stadium the game was played in. (Source: Pro-Football-Reference)
stadium	character	Name of the stadium

Example

from sportsdataverse.nfl import load_nfl_schedule
schedule = load_nfl_schedule(seasons=[2024])
schedule.shape

# Multi-season range

schedule = load_nfl_schedule(seasons=range(2020, 2025))

# Filter to a single week

import polars as pl
week_one = load_nfl_schedule(seasons=[2024]).filter(pl.col("week") == 1)

# Pandas round-trip

schedule_pd = load_nfl_schedule(seasons=[2024], return_as_pandas=True)
schedule_pd[["game_id", "home_team", "away_team", "week"]].head()

load_snap_counts(seasons: 'List[int]', return_as_pandas=False) -> 'pl.DataFrame'

Load NFL snap counts data for selected seasons

Parameters

Parameter	Type	Default	Description
seasons	list		Used to define different seasons. 2012 is the earliest available season.
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe.

Returns

Polars dataframe containing snap counts available for the requested seasons.

col_name	type	description
game_id	character	Ten digit identifier for NFL game.
pfr_game_id	character	PFR game ID
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
game_type	character	The most recent game type of that season that a player appeared on the roster.
week	integer	Season week.
player	character	Player name
pfr_player_id	character	ID from Pro Football Reference
position	character	Primary position as reported by NFL.com
team	character	NFL team. Uses official abbreviations as per NFL.com
opponent	character	Opposing team of player
offense_snaps	double	Number of snaps on offense
offense_pct	double	Percent of offensive snaps taken
defense_snaps	double	Number of snaps on defense
defense_pct	double	Percent of defensive snaps taken
st_snaps	double	Number of snaps on special teams
st_pct	double	Percent of special teams snaps taken

Example

from sportsdataverse.nfl import load_nfl_snap_counts
snaps = load_nfl_snap_counts(seasons=[2024])

# Multi-season range with offense-only filter

import polars as pl
offense = (
    load_nfl_snap_counts(seasons=range(2022, 2025))
    .filter(pl.col("offense_snaps") > 0)
)

load_team_stats(seasons: 'List[int]', summary_level: 'str' = 'week', return_as_pandas=False, *, source: 'str' = 'nflverse') -> 'pl.DataFrame'

Load NFL team stats data going back to 1999

Parameters

Parameter	Type	Default	Description
seasons	list		Used to define different seasons. 1999 is the earliest available season.
summary_level	str	'week'	Aggregation level. One of "week", "reg", "post", "reg+post". Defaults to "week". Ignored when source is the SDV-native release (a single week-level parquet covering all seasons; filter post-load).
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe.
source	str	'nflverse'	Which team-stats release to read. "nflverse" (the default) reads the per-season nflverse stats_team releases. "sportsdataverse" / "sdv" reads the SDV-native nfl_team_stats release (a single combined week-level parquet, built by sportsdataverse.nfl.build_nfl_team_stats from the SDV play-by-play and filtered to the requested seasons post-load).

Returns

Polars dataframe containing team stats available for the requested seasons.

col_name	type	description
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
week	integer	Season week.
team	character	NFL team. Uses official abbreviations as per NFL.com
season_type	character	REG or POST indicating if the timeframe belongs to regular or post season.
opponent_team	character	Abbreviation of the opposing team the team faced in the game or week represented by this row.
completions	integer	The number of completed passes.
attempts	integer	The number of pass attempts as defined by the NFL.
passing_yards	integer	Numeric yards by the passer_player_name, including yards gained in pass plays with laterals. This should equal official passing statistics.
passing_tds	integer	The number of passing touchdowns.
passing_interceptions	integer	Total number of interceptions thrown by the team's quarterbacks during the period covered.
sacks_suffered	integer	Total number of times the team's quarterback was sacked during the period covered.
sack_yards_lost	integer	Total yards lost by the team's offense as a result of being sacked.
sack_fumbles	integer	The number of sacks with a fumble.
sack_fumbles_lost	integer	The number of sacks with a lost fumble.
passing_air_yards	integer	Passing air yards (includes incomplete passes).
passing_yards_after_catch	integer	Yards after the catch gained on plays in which player was the passer (this is an unofficial stat and may differ slightly between different sources).
passing_first_downs	integer	First downs on pass attempts.
passing_epa	double	Total expected points added on pass attempts and sacks. NOTE: this uses the variable qb_epa, which gives QB credit for EPA for up to the point where a receiver lost a fumble after a completed catch and makes EPA work more like passing yards on plays with fumbles.
passing_cpoe	double	Completion percentage over expectation for the team's passing game — how much better or worse actual completion rate was versus the model-predicted rate.
passing_2pt_conversions	integer	Two-point conversion passes.
carries	integer	The number of official rush attempts (incl. scrambles and kneel downs). Rushes after a lateral reception don't count as carry.
rushing_yards	integer	Numeric yards by the rusher_player_name, excluding yards gained in rush plays with laterals. This should equal official rushing statistics but could miss yards gained in rush plays with laterals. Please see the description of lateral_rusher_player_name for further information.
rushing_tds	integer	The number of rushing touchdowns (incl. scrambles). Also includes touchdowns after obtaining a lateral on a play that started with a rushing attempt.
rushing_fumbles	integer	The number of rushes with a fumble.
rushing_fumbles_lost	integer	The number of rushes with a lost fumble.
rushing_first_downs	integer	First downs on rush attempts (incl. scrambles).
rushing_epa	double	Expected points added on rush attempts (incl. scrambles and kneel downs).
rushing_2pt_conversions	integer	Two-point conversion rushes
receptions	integer	The number of pass receptions. Lateral receptions officially don't count as reception.
targets	integer	The number of pass plays where the player was the targeted receiver.
receiving_yards	integer	Numeric yards by the receiver_player_name, excluding yards gained in pass plays with laterals. This should equal official receiving statistics but could miss yards gained in pass plays with laterals. Please see the description of lateral_receiver_player_name for further information.
receiving_tds	integer	The number of touchdowns following a pass reception. Also includes touchdowns after receiving a lateral on a play that started as a pass play.
receiving_fumbles	integer	The number of fumbles after a pass reception.
receiving_fumbles_lost	integer	The number of fumbles lost after a pass reception.
receiving_air_yards	integer	Receiving air yards (incl. incomplete passes).
receiving_yards_after_catch	integer	Yards after the catch gained on plays in which player was receiver (this is an unofficial stat and may differ slightly between different sources).
receiving_first_downs	integer	Total number of first downs gained on receptions
receiving_epa	double	Total EPA on plays where this receiver was targeted
receiving_2pt_conversions	integer	Two-point conversion receptions
special_teams_tds	integer	Total number of kick/punt return touchdowns
def_tackles_solo	integer	Total number of solo tackles for this player
def_tackles_with_assist	integer	Number of tackles this player had with an assisted tackle
def_tackle_assists	integer	Number of assisted tackles for this player
def_tackles_for_loss	integer	Number of tackles for loss (TFL) for this player
def_tackles_for_loss_yards	integer	Yards lost from TFLs involving this player
def_fumbles_forced	integer	Number of times a fumble was forced from this player
def_sacks	double	Number of sacks form this player
def_sack_yards	double	Yards lost from sacks forced by this player
def_qb_hits	integer	Number of QB hits from this player (should not include plays where the QB was sacked)
def_interceptions	integer	Number of interceptions forced by this player
def_interception_yards	integer	yards gained/lost by interception returns from this player
def_pass_defended	integer	Number of passes defended/broken up by this player
def_tds	integer	Number of defensive touchdowns scored by this player
def_fumbles	integer	Number of fumbles by this player
def_safeties	integer	Number of safeties recorded by the team's defense (tackling an opponent in their own end zone).
misc_yards	integer	Yards gained by the team through miscellaneous means not captured in standard rushing, passing, or return categories.
fumble_recovery_own	integer	Number of the team's own fumbles that were recovered by the team itself.
fumble_recovery_yards_own	integer	Total yards gained after recovering their own fumbles.
fumble_recovery_opp	integer	Number of fumbles recovered by the team from the opposing offense (defensive fumble recoveries).
fumble_recovery_yards_opp	integer	Total yards gained by the team on returns of opponent fumble recoveries.
fumble_recovery_tds	integer	Number of touchdowns scored by the team on fumble recoveries (own or opponent).
penalties	integer	Total number of penalties.
penalty_yards	integer	Yards gained (or lost) by the posteam from the penalty.
timeouts	integer	Number of timeouts remaining or used by the team during the game or period covered.
punt_returns	integer	Number of punt returns.
punt_return_yards	integer	Team punt return yards.
kickoff_returns	integer	Total number of kickoff return attempts by the team.
kickoff_return_yards	integer	Total yards gained by the team on kickoff returns during the period covered.
fg_made	integer	TRUE when the field goal attempt was successful.
fg_att	integer	Total field goal attempts by the team's kicker during the period covered.
fg_missed	integer	Total number of field goal attempts that were missed (not blocked, not made) by the team's kicker.
fg_blocked	integer	Total number of field goal attempts that were blocked by the opposing defense.
fg_long	integer	Distance in yards of the team's longest successful field goal during the period covered.
fg_pct	double	Field goal percentage (0-1).
fg_made_0_19	integer	Number of field goals made by the team from 0–19 yards.
fg_made_20_29	integer	Number of field goals made by the team from 20–29 yards.
fg_made_30_39	integer	Number of field goals made by the team from 30–39 yards.
fg_made_40_49	integer	Number of field goals made by the team from 40–49 yards.
fg_made_50_59	integer	Number of field goals made by the team from 50–59 yards.
fg_made_60_	integer	Number of field goals made by the team from 60 yards or longer.
fg_missed_0_19	integer	Number of field goal attempts missed from 0–19 yards.
fg_missed_20_29	integer	Number of field goal attempts missed from 20–29 yards.
fg_missed_30_39	integer	Number of field goal attempts missed from 30–39 yards.
fg_missed_40_49	integer	Number of field goal attempts missed from 40–49 yards.
fg_missed_50_59	integer	Number of field goal attempts missed from 50–59 yards.
fg_missed_60_	integer	Number of field goal attempts missed from 60 yards or longer.
fg_made_list	character	Comma-separated list of distances (in yards) for each successful field goal made by the team.
fg_missed_list	character	Comma-separated list of distances (in yards) for each missed field goal attempt by the team.
fg_blocked_list	character	Comma-separated list of distances (in yards) for field goal attempts blocked by or against the team.
fg_made_distance	integer	Total cumulative distance in yards of all successful field goals made by the team.
fg_missed_distance	integer	Total cumulative distance in yards of all missed field goal attempts by the team.
fg_blocked_distance	integer	Distance in yards of the most recent or representative blocked field goal attempt.
pat_made	integer	Total number of extra points successfully kicked by the team.
pat_att	integer	Total number of extra point (PAT) kick attempts by the team.
pat_missed	integer	Number of extra point kick attempts that were missed (neither made nor blocked).
pat_blocked	integer	Number of extra point attempts that were blocked by the opposing defense.
pat_pct	double	Extra point conversion percentage (pat_made divided by pat_att) for the team's kicker.
gwfg_made	integer	Number of game-winning field goals successfully converted by the team's kicker.
gwfg_att	integer	Number of game-winning field goal attempts (potential go-ahead kicks in the final moments).
gwfg_missed	integer	Number of game-winning field goal attempts that were missed by the team's kicker.
gwfg_blocked	integer	Number of game-winning field goal attempts that were blocked by the opposing defense.
gwfg_distance	integer	Distance in yards of the game-winning field goal attempt(s) during the period covered.

Example

from sportsdataverse.nfl import load_nfl_team_stats
weekly = load_nfl_team_stats(seasons=[2024])

# Regular-season-only team stats

reg = load_nfl_team_stats(seasons=[2024], summary_level="reg")

# SDV-native team stats (built from SDV play-by-play)

sdv = load_nfl_team_stats(seasons=[2024], source="sdv")

load_teams(return_as_pandas=False) -> 'pl.DataFrame'

Load NFL team ID information and logos

Parameters

Parameter	Type	Default	Description
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe.

Returns

Polars dataframe containing teams available.

col_name	type	description
team_abbr	character	Official team abbreveation
team_name	character	Full team display name (e.g. 'Las Vegas Aces').
team_id	integer	Unique team identifier.
team_nick	character	Team nickname or mascot name (e.g., 'Chiefs', 'Patriots').
team_conf	character	Conference the team belongs to (e.g., 'AFC', 'NFC').
team_division	character	Division within the conference the team belongs to (e.g., 'AFC East').
team_color	character	Team primary color (hex without leading '#').
team_color2	character	Secondary brand color for the team, expressed as a hex color code.
team_color3	character	Tertiary brand color for the team, expressed as a hex color code.
team_color4	character	Quaternary brand color for the team, expressed as a hex color code.
team_logo_wikipedia	character	URL of the team's logo image as hosted on Wikipedia.
team_logo_espn	character	URL of the team's primary logo as hosted on ESPN.
team_wordmark	character	URL of the team's wordmark (text-based logo) image.
team_conference_logo	character	URL of the conference logo image associated with the team.
team_league_logo	character	URL of the NFL league logo image.
team_logo_squared	character	URL of a square-format version of the team's logo.

Example

from sportsdataverse.nfl import load_nfl_teams
teams = load_nfl_teams()
teams.shape

# Pandas round-trip

teams_pd = load_nfl_teams(return_as_pandas=True)
teams_pd[["team_abbr", "team_name", "team_conf", "team_division"]].head()

load_trades(return_as_pandas=False) -> 'pl.DataFrame'

Load NFL trades data

Parameters

Parameter	Type	Default	Description
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe.

Returns

Polars dataframe containing NFL trade information.

col_name	type	description
trade_id	integer	ID of Trade
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
trade_date	character	Exact date that trade occurred
gave	character	Team that gave pick/player in row
received	character	Team that received pick/player in row
pick_season	integer	Draft in which traded pick was in
pick_round	integer	Round in which traded pick was in
pick_number	integer	Pick number of traded pick
conditional	integer	Binary indicator of whether or not traded pick was conditional
pfr_id	character	Pro-Football-Reference ID for player
pfr_name	character	Full name of traded player

Example

from sportsdataverse.nfl import load_nfl_trades
trades = load_nfl_trades()
trades.shape

# Filter to a single season

import polars as pl
trades_2024 = load_nfl_trades().filter(pl.col("season") == 2024)

Utilities & helpers

NFLPlayProcess(gameId=0, raw=False, path_to_json='/', return_keys=None, **kwargs)

Process ESPN NFL play-by-play feeds into a tidy game-level dictionary.

Wraps the ESPN summary endpoint (or a local JSON dump) and pipes the result through a chain of feature-engineering steps -- down/distance, play-type flags, EPA, WPA, QBR, drive aggregation, and an advanced box score. Use run_processing_pipeline() for the full feature set or run_cleaning_pipeline() for a lighter clean.

Parameters

Parameter	Type	Default	Description
gameId	int	0	ESPN event id (e.g. 401671801).
raw	bool	False	If True, espn_nfl_pbp() returns the ESPN payload untouched. If False (default), it normalizes keys.
path_to_json	str	'/'	Directory containing {gameId}.json for the nfl_pbp_disk() flow (offline replay).
return_keys	list[str] | None	None	If supplied, run_processing_pipeline returns only the listed keys from the result dict.

Example

from sportsdataverse.nfl import NFLPlayProcess
proc = NFLPlayProcess(gameId=401671801)
proc.espn_nfl_pbp()
result = proc.run_processing_pipeline()
len(result["plays"])

# Offline replay from a JSON dump

proc = NFLPlayProcess(gameId=401671801, path_to_json="./pbp_dump")
proc.nfl_pbp_disk()
cleaned = proc.run_cleaning_pipeline()

# Subset the return payload

proc = NFLPlayProcess(gameId=401671801, return_keys=["plays", "boxscore"])
proc.espn_nfl_pbp()
slim = proc.run_processing_pipeline()
sorted(slim.keys())  # ['boxscore', 'plays']

Methods

NFLPlayProcess.corrupt_pbp_check()

Detect ESPN payloads that look corrupt or partial.

Returns True when one of three guard conditions trips:

• No plays at all.
• Fewer than 50 plays for a game ESPN reports as completed.
• More than 500 plays for a game ESPN reports as completed.

run_processing_pipeline() and run_cleaning_pipeline() use this to skip feature engineering on obviously broken payloads.

Returns

True if the payload looks corrupt; False otherwise.

Example

from sportsdataverse.nfl import NFLPlayProcess
proc = NFLPlayProcess(gameId=401671801)
proc.espn_nfl_pbp()
if not proc.corrupt_pbp_check():
    result = proc.run_processing_pipeline()

NFLPlayProcess.create_box_score(play_df)

Build the advanced box score (passer / rusher / receiver / team / situational / defensive / turnover / drives)

from a feature-engineered plays DataFrame.

This is normally called by run_processing_pipeline() -- it auto-runs the pipeline first if it hasn't been triggered yet, so callers can also invoke it directly on a freshly-instantiated processor.

Parameters

Parameter	Type	Default	Description
play_df	pl.DataFrame		The plays frame produced after the full feature-engineering chain (downs, play-type flags, EPA, WPA, drive aggregation).

Returns

Box score keyed by "pass", "rush", "receiver", "team", "situational", "defensive", "turnover", "drives" -- each value a list of dicts ready to be serialized.

Example

from sportsdataverse.nfl import NFLPlayProcess
proc = NFLPlayProcess(gameId=401671801)
proc.espn_nfl_pbp()
result = proc.run_processing_pipeline()
box = result["advBoxScore"]
sorted(box.keys())

NFLPlayProcess.espn_nfl_pbp(**kwargs)

espn_nfl_pbp() - Pull the game by id. Data from API endpoints: nfl/playbyplay, nfl/summary

Returns

Dictionary of game data with keys - "gameId", "plays", "boxscore", "header", "broadcasts", "videos", "playByPlaySource", "standings", "leaders", "timeouts", "homeTeamSpread", "overUnder", "pickcenter", "againstTheSpread", "odds", "predictor", "winprobability", "espnWP", "gameInfo", "season"

Example

from sportsdataverse.nfl import NFLPlayProcess
proc = NFLPlayProcess(gameId=401220403)
payload = proc.espn_nfl_pbp()
sorted(payload.keys())[:5]

# Raw ESPN passthrough (no key normalization)

proc_raw = NFLPlayProcess(gameId=401220403, raw=True)
espn_dump = proc_raw.espn_nfl_pbp()

# Chain into the full processing pipeline

proc = NFLPlayProcess(gameId=401220403)
proc.espn_nfl_pbp()
result = proc.run_processing_pipeline()

NFLPlayProcess.nfl_pbp_disk()

Load a previously-saved ESPN payload from {path_to_json}/{gameId}.json.

Use this to replay an old game offline without hitting the ESPN endpoint -- handy for snapshot-driven tests and reproducible feature engineering.

Returns

The parsed JSON content; also stored on self.json.

Example

from sportsdataverse.nfl import NFLPlayProcess
proc = NFLPlayProcess(gameId=401220403, path_to_json="./pbp_dump")
proc.nfl_pbp_disk()
result = proc.run_processing_pipeline()

NFLPlayProcess.nfl_pbp_json(**kwargs)

Set self.json to the imported json module reference (legacy stub).

Retained for API compatibility. Prefer espn_nfl_pbp() (live) or nfl_pbp_disk() (offline) to populate self.json with an actual ESPN payload.

Returns

The Python json module reference (mirrors legacy behavior).

Example

from sportsdataverse.nfl import NFLPlayProcess
proc = NFLPlayProcess(gameId=401220403)
proc.nfl_pbp_json()  # populates `self.json` with the json module

NFLPlayProcess.run_cleaning_pipeline()

Run the lighter cleaning pipeline against self.json.

Identical to run_processing_pipeline() up through the add_spread_time` step but stops short of EPA / WPA / QBR / drive aggregation and the advanced box score. Use this when you want clean play structure without the modeled features.

Returns

The cleaned game dict (or the subset specified by return_keys at construction).

Example

from sportsdataverse.nfl import NFLPlayProcess
proc = NFLPlayProcess(gameId=401671801)
proc.espn_nfl_pbp()
cleaned = proc.run_cleaning_pipeline()
"plays" in cleaned and "advBoxScore" not in cleaned

NFLPlayProcess.run_processing_pipeline()

Run the full feature-engineering pipeline against self.json.

Pipes the plays frame through the chain of helpers: downs, play-type flags, rush/pass flags, team-score variables, new play types, penalties, play-category flags, yardage cols, player cols, post-play cols, spread time, EPA, WPA, drive data, and QBR -- followed by the advanced box score build.

Returns

Dict | None: The full processed game dict (or the subset specified by return_keys at construction). Returns the partial result when corrupt_pbp_check() short-circuits.

Example

from sportsdataverse.nfl import NFLPlayProcess
proc = NFLPlayProcess(gameId=401671801)
proc.espn_nfl_pbp()
result = proc.run_processing_pipeline()
len(result["plays"]), len(result["drives"])

# Subset returned keys for downstream serialization

proc = NFLPlayProcess(
    gameId=401671801,
    return_keys=["plays", "advBoxScore", "winprobability"],
)
proc.espn_nfl_pbp()
slim = proc.run_processing_pipeline()
sorted(slim.keys())

get_current_nfl_season(roster: 'bool' = False) -> 'int'

Return the current NFL season year.

Parameters

Parameter	Type	Default	Description
roster	bool	False	If True, use roster-year logic (current calendar year on/after March 15, otherwise previous year). If False, use season logic (current calendar year on/after the Thursday following Labor Day, otherwise previous year).

Returns

The current season (or roster) year.

Example

from sportsdataverse.nfl import get_current_nfl_season
season = get_current_nfl_season()
print(season)

# Roster-year semantics (March 15 cutover)

roster_year = get_current_nfl_season(roster=True)

# Pair with a loader to fetch only the active season

from sportsdataverse.nfl import load_nfl_schedule
schedule = load_nfl_schedule(seasons=[get_current_nfl_season()])

get_current_nfl_week(use_date: 'bool' = True, roster: 'bool' = False) -> 'int'

Return the current NFL week (1-22).

Parameters

Parameter	Type	Default	Description
use_date	bool	True	If True (default), compute the week purely from the calendar (number of weeks since the first Thursday of September of the current season). If False, hit the live schedule via load_nfl_schedule() and return the week of the next unplayed game (matches nflreadpy's use_date=False path).
roster	bool	False	Forwarded to get_current_nfl_season() for season inference.

Returns

The current week, capped at 22.

Example

from sportsdataverse.nfl import get_current_nfl_week
week = get_current_nfl_week()

# Schedule-driven week (hits the live schedule parquet)

week_live = get_current_nfl_week(use_date=False)

# Roster-year season inference

week_roster = get_current_nfl_week(roster=True)

# Pair with a PBP fetch to grab only the most recent season+week

import polars as pl
from sportsdataverse.nfl import (
    get_current_nfl_season, get_current_nfl_week, load_nfl_pbp,
)
current_pbp = (
    load_nfl_pbp(seasons=[get_current_nfl_season()])
    .filter(pl.col("week") == get_current_nfl_week())
)

get_current_season(roster: 'bool' = False) -> 'int'

Return the current NFL season year.

Parameters

Parameter	Type	Default	Description
roster	bool	False	If True, use roster-year logic (current calendar year on/after March 15, otherwise previous year). If False, use season logic (current calendar year on/after the Thursday following Labor Day, otherwise previous year).

Returns

The current season (or roster) year.

Example

from sportsdataverse.nfl import get_current_nfl_season
season = get_current_nfl_season()
print(season)

# Roster-year semantics (March 15 cutover)

roster_year = get_current_nfl_season(roster=True)

# Pair with a loader to fetch only the active season

from sportsdataverse.nfl import load_nfl_schedule
schedule = load_nfl_schedule(seasons=[get_current_nfl_season()])

get_current_week(use_date: 'bool' = True, roster: 'bool' = False) -> 'int'

Return the current NFL week (1-22).

Parameters

Parameter	Type	Default	Description
use_date	bool	True	If True (default), compute the week purely from the calendar (number of weeks since the first Thursday of September of the current season). If False, hit the live schedule via load_nfl_schedule() and return the week of the next unplayed game (matches nflreadpy's use_date=False path).
roster	bool	False	Forwarded to get_current_nfl_season() for season inference.

Returns

The current week, capped at 22.

Example

from sportsdataverse.nfl import get_current_nfl_week
week = get_current_nfl_week()

# Schedule-driven week (hits the live schedule parquet)

week_live = get_current_nfl_week(use_date=False)

# Roster-year season inference

week_roster = get_current_nfl_week(roster=True)

# Pair with a PBP fetch to grab only the most recent season+week

import polars as pl
from sportsdataverse.nfl import (
    get_current_nfl_season, get_current_nfl_week, load_nfl_pbp,
)
current_pbp = (
    load_nfl_pbp(seasons=[get_current_nfl_season()])
    .filter(pl.col("week") == get_current_nfl_week())
)

most_recent_nfl_season(roster: 'bool' = False) -> 'int'

Alias for get_current_nfl_season() mirroring nflreadr's

most_recent_season().

Parameters

Parameter	Type	Default	Description
roster	bool	False

Example

from sportsdataverse.nfl.utils_date import most_recent_nfl_season
season = most_recent_nfl_season()

# Roster-year flavor

roster_year = most_recent_nfl_season(roster=True)

Other

NflConfig(cache_mode: 'CacheMode' = 'memory', cache_dir: 'Optional[Path]' = None, cache_duration: 'int' = 86400, verbose: 'bool' = True, timeout: 'int' = 30, user_agent: 'str' = 'sportsdataverse-py-nfl') -> None

Runtime configuration for sdv-py NFL loaders.

Fields mirror nflreadpy's NflreadpyConfig so users can swap engines without changing call sites. The defaults are conservative: in-memory caching with a 24-hour TTL, verbose progress bars on, 30-second HTTP timeout.

Parameters

Parameter	Type	Default	Description
cache_mode	CacheMode	'memory'
cache_dir	Optional[Path]	None
cache_duration	int	86400
verbose	bool	True
timeout	int	30
user_agent	str	'sportsdataverse-py-nfl'

Example

from sportsdataverse.nfl import get_config
cfg = get_config()  # NflConfig instance
cfg.cache_mode      # "memory"
cfg.cache_duration  # 86400 (24h)
cfg.timeout         # 30 (seconds)

# Construct a fresh instance directly (rarely needed -- prefer ``update_config``)

from sportsdataverse.nfl import NflConfig
cfg = NflConfig(cache_mode="off", timeout=10)

build_nfl_player_stats(seasons: 'List[int]', *, summary_level: 'str' = 'week', season_type: 'str' = 'REG', source: 'str' = 'sdv', return_as_pandas: 'bool' = False) -> "pl.DataFrame | 'pd.DataFrame'"

Build nflverse player_stats by aggregating SDV-native play-by-play.

A faithful polars port of nflfastR's calculate_player_stats (aggregate_game_stats.R): per-player passing / rushing / receiving frames are full-outer-joined on the group keys, special-teams touchdowns and fantasy points are added, and player metadata is joined from sportsdataverse.nfl.load_nfl_players. See the module docstring for the SDV-PBP column-gap handling (passing_epa uses the exact qb_epa; rushing_epa / receiving_epa use plain epa per nflfastR).

Parameters

Parameter	Type	Default	Description
seasons	List[int]		Four-digit NFL seasons to aggregate (e.g. [2023]).
summary_level	str	'week'	"week" (group on season + week + player_id, with opponent_team) or "season" (group on season + player_id, with recent_team = last team and games = distinct game count).
season_type	str	'REG'	"REG", "POST", or "REG+POST". Pre-filters the play-by-play before aggregation.
source	str	'sdv'	Play-by-play release passed to load_nfl_pbp. Defaults to "sdv" (the SDV-native enriched release).
return_as_pandas	bool	False	If True return a pandas DataFrame; else polars.

Returns

A polars (or pandas) DataFrame in the published load_nfl_player_stats schema. At summary_level="season" the week / season_type / opponent_team columns are replaced by a games column.

Example

from sportsdataverse.nfl import build_nfl_player_stats
wk = build_nfl_player_stats([2023], summary_level="week")
print(wk.shape)

# Season totals as pandas

df_pd = build_nfl_player_stats([2023], summary_level="season",
                               return_as_pandas=True)

# Pipeline next step (one line)

wk.filter(pl.col("attempts") >= 5).sort("passing_epa", descending=True).head()

build_nfl_players(*, return_as_pandas: 'bool' = False) -> "Union[pl.DataFrame, 'pd.DataFrame']"

Build an SDV-native NFL players frame from ESPN's public athletes endpoint.

Walks ESPN's public NFL athletes index (sports.core.api.espn.com/v2/sports/football/leagues/nfl/athletes), resolves each athlete's detail resource, flattens it onto the SDV-native players schema, dedups to the highest numeric espn_id per (full_name, birth_date) (the ESPN ~2007 4-digit -> 7-digit id migration left some players with two ids), and enriches gsis_id + other cross-IDs by a best-effort join against sportsdataverse.nfl.load_nfl_players.

This is the public ESPN-athletes tier only — a partial mirror of nflverse's full seven-source players.parquet (three of those sources, PFR / OTC / PFF, require private credentials). ESPN-native rows with no nflverse match keep only their ESPN fields (cross-IDs left null). For the full identity master prefer sportsdataverse.nfl.load_nfl_players; use build_nfl_players when you need an SDV-native frame that depends only on the live public ESPN API.

Parameters

Parameter	Type	Default	Description
return_as_pandas	bool	False	If True, return a pandas.DataFrame; otherwise a polars.DataFrame (default).

Returns

A one-row-per-player DataFrame with the documented schema (espn_id, full_name, first_name, last_name, position, team, jersey, height, weight, birth_date, status, headshot_url, gsis_id, esb_id, pfr_id, pff_id, smart_id, college). An empty / failed fetch yields a zero-row frame carrying the same column set (never a raise).

Example

from sportsdataverse.nfl import build_nfl_players
players = build_nfl_players()
print(players.shape)

# Pandas output

df = build_nfl_players(return_as_pandas=True)

# Pipeline next step (one line)

import polars as pl
build_nfl_players().filter(pl.col("position") == "QB").head()

build_nfl_rosters(seasons: 'List[int]', *, return_as_pandas: 'bool' = False) -> "Union[pl.DataFrame, 'pd.DataFrame']"

Build SDV-native NFL season rosters from the public Shield API.

For each (season, team) the public NFL Shield endpoint /football/v2/rosters returns (reached through sportsdataverse.nfl.nfl_rosters), every player in the persons[] array is flattened onto the SDV-native season-roster schema, team abbreviations are folded to the nflverse standard (season-aware relocations), and cross-system IDs + college are enriched by a best-effort left join against sportsdataverse.nfl.load_nfl_players on gsis_id.

This is the public Shield tier only — a partial mirror of nflverse's full three-tier roster product. Shield supplies gsis_id densely across all seasons, but the cross-system IDs (espn_id, sportradar_id, yahoo_id, rotowire_id, pff_id, pfr_id, fantasy_data_id, sleeper_id) and college are only as dense as the players-table cross-walk, which is sparse for pre-2016 seasons. For the richest roster data prefer sportsdataverse.nfl.load_nfl_rosters (reads nflverse's published parquet); use build_nfl_rosters when you need an SDV-native frame that depends only on the live NFL Shield API.

Parameters

Parameter	Type	Default	Description
seasons	List[int]		Seasons to build (e.g. [2023] or range(2020, 2025)). A single int is accepted and wrapped. A season Shield returns no data for contributes no rows rather than raising.
return_as_pandas	bool	False	If True, return a pandas.DataFrame; otherwise a polars.DataFrame (default).

Returns

A one-row-per-player season-roster DataFrame with the documented schema. An empty / missing season yields a zero-row frame carrying the same column set (never a raise).

Example

from sportsdataverse.nfl import build_nfl_rosters
rosters = build_nfl_rosters([2023])
print(rosters.shape)

# Multi-season build, pandas output

df = build_nfl_rosters(range(2021, 2024), return_as_pandas=True)

# Pipeline next step (one line)

import polars as pl
build_nfl_rosters([2023]).filter(pl.col("team") == "KC").head()

build_nfl_season(game_ids: 'list[int] | None' = None, *, seasons: 'list[int] | None' = None, source: 'str' = 'espn', return_as_pandas: 'bool' = False) -> "'pl.DataFrame | pd.DataFrame'"

Compile play-by-play for multiple NFL games into one tidy frame.

The source parameter determines which input parameter is required:

• source="espn" — requires game_ids; seasons must be None.
• source="nflverse" — requires seasons; game_ids must be None.

For ESPN games the function either loads a previously cached plays frame or processes the game fresh via NFLPlayProcess. Individual game failures are logged and skipped so a single bad game does not abort the whole season build. The per-game frames are concatenated with how="diagonal_relaxed" (schema union, missing columns filled with null) so games with slightly different column sets merge cleanly.

Parameters

Parameter	Type	Default	Description
game_ids	list[int] | None	None	ESPN event IDs to compile (e.g. [401671801, 401671802]). Required when source="espn"; must be None for other sources.
seasons	list[int] | None	None	Season years to compile (e.g. [2023, 2024]). Required when source="nflverse"; must be None for other sources.
source	str	'espn'	Data source. - "espn" (default): each game is processed via NFLPlayProcess(gameId=gid).espn_nfl_pbp() + run_processing_pipeline(). Pass game_ids. - "nflverse": delegates to sportsdataverse.nfl.load_nfl_pbp for the requested seasons. Pass seasons. Returns the full pre-enriched season frame as-is. - "shield": raises NotImplementedError — Shield (api.nfl.com) play-by-play lives in the native-pipeline (nfl-data) repository, not sdv-py.
return_as_pandas	bool	False	If True, return a pandas.DataFrame instead of polars.

Returns

All plays from the requested games/seasons, concatenated with schema-union semantics (missing columns are null). Returns a zero-row frame if every game failed (ESPN source only). When return_as_pandas is True, returns a pandas.DataFrame instead.

Example

from sportsdataverse.nfl import build_nfl_season
df = build_nfl_season(game_ids=[401671801, 401671802])
print(df.shape)

# nflverse season compile (pass season years)

from sportsdataverse.nfl import build_nfl_season
df = build_nfl_season(seasons=[2023], source="nflverse")
print(df.shape)

# With filesystem cache enabled (ESPN)

from sportsdataverse.nfl import build_nfl_season, update_config
update_config(cache_mode="filesystem")
df = build_nfl_season(game_ids=[401671801, 401671802])  # processes + caches
df2 = build_nfl_season(game_ids=[401671801, 401671802]) # served from cache

# Pandas output

from sportsdataverse.nfl import build_nfl_season
df_pd = build_nfl_season(game_ids=[401671801], return_as_pandas=True)
print(df_pd.shape)

build_nfl_team_stats(seasons: 'List[int]', *, summary_level: 'str' = 'week', season_type: 'str' = 'REG', source: 'str' = 'sdv', return_as_pandas: 'bool' = False) -> "pl.DataFrame | 'pd.DataFrame'"

Build nflverse team_stats by aggregating SDV-native play-by-play.

A faithful polars port of nflfastR's calculate_stats(stat_type = "team") (the aggregate_game_stats* family). Offense is keyed on posteam, defense on the tackler's team (per-play *_team slot tags -- NOT defteam, which double-counts on return plays), kicking on posteam, and returns / penalties / timeouts on the relevant play team tag. See the module docstring for the full grouping + SDV-PBP gap notes (passing_epa uses the exact qb_epa; gwfg_* derive from fixed_drive).

Parameters

Parameter	Type	Default	Description
seasons	List[int]		Four-digit NFL seasons to aggregate (e.g. [2023]).
summary_level	str	'week'	"week" (group on season + week + team, with opponent_team) or "season" (group on season + team, with a games distinct-game count replacing week / season_type / opponent_team).
season_type	str	'REG'	"REG", "POST", or "REG+POST". Pre-filters the play-by-play before aggregation.
source	str	'sdv'	Play-by-play release passed to load_nfl_pbp. Defaults to "sdv" (the SDV-native enriched release).
return_as_pandas	bool	False	If True return a pandas DataFrame; else polars.

Returns

A polars (or pandas) DataFrame in the published load_nfl_team_stats schema (~102 columns). At summary_level="season" the week / season_type / opponent_team columns are replaced by a games column.

Example

from sportsdataverse.nfl import build_nfl_team_stats
wk = build_nfl_team_stats([2023], summary_level="week")
print(wk.shape)

# Season totals as pandas

df_pd = build_nfl_team_stats([2023], summary_level="season",
                             return_as_pandas=True)

# Pipeline next step (one line)

wk.sort("def_sacks", descending=True).head()

cached_loader(func: 'F') -> 'F'

Decorator that adds caching to a load_nfl_* function.

Honors the active NflConfig.cache_mode:

• memory: dict-based per-process cache.
• filesystem: parquet-based cross-process cache under cache_dir.
• off: no caching, function runs every time.

The cache key is the hash of (qualified_name, args, kwargs) with return_as_pandas excluded so memory / disk hits work regardless of which return shape the caller asked for. The cache always stores the polars frame internally and converts to pandas on read when requested.

Parameters

Parameter	Type	Default	Description
func	F

Example

import polars as pl
from sportsdataverse.nfl.cache import cached_loader

@cached_loader
def load_my_thing(season: int, return_as_pandas: bool = False):
    # ... fetch parquet, build a polars frame ...
    return pl.DataFrame({"season": [season]})

df1 = load_my_thing(2024)            # network hit, populates cache
df2 = load_my_thing(2024)            # served from cache
df_pd = load_my_thing(2024, return_as_pandas=True)
# `return_as_pandas` is excluded from the cache key, so the
# polars hit is reused and converted to pandas on the way out.

# Switch caching modes at runtime

from sportsdataverse.nfl import clear_cache, update_config

update_config(cache_mode="filesystem")  # parquet-on-disk reuse
df3 = load_my_thing(2024)               # writes parquet under cache_dir
clear_cache()                           # wipe both memory + filesystem
update_config(cache_mode="off")         # bypass cache entirely

calculate_completion_probability(pbp_data: 'pl.DataFrame', *, return_as_pandas: 'bool' = False) -> "Union[pl.DataFrame, 'pd.DataFrame']"

Compute completion probability (CP) and CPOE for pass plays.

Mirrors nflfastR's helper_add_cp_cpoe.R. Scores only intended pass plays (where air_yards is not null); non-pass plays receive null in the cp column. When complete_pass is present, cpoe = 100 * (complete_pass - cp) is also added — on nflfastR's percentage-point scale (add_cp in helper_add_cp_cpoe.R).

Drops and recomputes any existing cp / cpoe columns.

Parameters

Parameter	Type	Default	Description
pbp_data	DataFrame		nflverse-format play-by-play DataFrame. Required: air_yards, season, ydstogo, down, posteam, home_team. Optional: roof, pass_location (for pass_middle), qb_hit, complete_pass (to derive cpoe).
return_as_pandas	bool	False	When True, return a pandas.DataFrame.

Returns

DataFrame with the original columns plus cp (null for non-pass plays) and cpoe (null when complete_pass absent).

Example

from sportsdataverse.nfl import load_nfl_pbp
from sportsdataverse.nfl.ep_wp import calculate_completion_probability

pbp = load_nfl_pbp([2023])
pbp_cp = calculate_completion_probability(pbp)
print(pbp_cp.select("cp", "cpoe").head())

calculate_epa(df: 'pl.DataFrame') -> 'pl.DataFrame'

Derive expected points added (EPA) from pre-scored EP point estimates.

This is the derivation half of NFLPlayProcess.__process_epa lifted into a shared, model-free function so the same nflfastR-faithful EPA logic can be reused by the streaming enrich_nfl_pbp pipeline and by process_epa` itself. It performs no model inference — the caller must already have scored the per-play EP point estimates.

Derivation rules (mirror nflfastR / the original process_epa`):

• Scoring overlays rewrite EP_end to the realized point value (offense TD +7 / +6.92 / 2pt variants, made FG +3, defensive scores, extra points, etc.) using the same type.text / text classification as process_epa`.
• Turnovers (end_change_vec / downs_turnover), kickoff turnovers and recovered onside kicks flip EP_end to the opponent's perspective (EP_end * -1).
• lag_EP_end is the previous play's EP_end; EP_between flips its sign on a prior-play possession change.
• Kickoffs use EP_start_touchback as EP_start.
• EPA = EP_end - EP_start normally; -EP_start on a non-scoring end-of-half play; 0 on a timeout; EP_end - EP_start + EP_between on a (non-kickoff, non-Penalty) penalty-in-text play.

Every shift is grouped .over("game_id") so a concatenated multi-game frame never leaks EP across game boundaries — this differs from process_epa` (which runs one game per instance and therefore needs no grouping).

Parameters

Parameter	Type	Default	Description
df	DataFrame		Play-by-play DataFrame that already carries the EP point estimates under the ESPN-internal names EP_start, EP_end and EP_start_touchback (e.g. as produced by the EP-scoring half of process_epa), plus the play-classification / flag columns: game_id, type.text, text, change_of_pos_team, downs_turnover, kickoff_onside, scoring_play, end_of_halfandpenalty_in_text. See EPA_REQUIRED_COLUMNS. This function does not score EP itself — score it first via the EP feature pipeline (the EP_* triple is the ESPN-internal naming, distinct from calculate_expected_points's lowercase ep).

Returns

The input frame with the EPA derivation applied. EP_start is rewritten to 0.92 for scoring-attempt play types (Extra Point Good, Extra Point Missed, Two-Point Conversion Good, Two-Point Conversion Missed, Two Point Pass, Two Point Rush, Blocked PAT) before any other overlays fire. EP_start / EP_end are then rewritten in place (overlays, sign flips, touchback), EP_between, lag_EP_end and lag_change_of_pos_team are added, EPA is added, and lowercase nflverse aliases ep (= EP_end), epa (= EPA), ep_start (= EP_start) and ep_end (= EP_end) are added for downstream contract parity.

Example

# For most use cases, call the high-level entry point instead. ``enrich_nfl_pbp`` scores EP, derives EPA, and adds WP/WPA/CP/CPOE in one shot on any nflverse-shape frame

    from sportsdataverse.nfl import load_nfl_pbp
    from sportsdataverse.nfl.ep_wp import enrich_nfl_pbp

    pbp = load_nfl_pbp([2023])
    enriched = enrich_nfl_pbp(pbp)
    print(enriched.select("game_id", "ep", "epa").head())

``calculate_epa`` directly requires ESPN-internal columns
(``EP_start``, ``EP_end``, ``EP_start_touchback``, ``type.text``,
etc.) produced by ``NFLPlayProcess``.  It is called internally by
``NFLPlayProcess.__process_epa`` and by the ``enrich_nfl_pbp``
orchestrator — a naked ``calculate_epa(load_nfl_pbp([2023]))``
will raise ``KeyError`` because those columns are absent from a
nflverse frame.

calculate_expected_points(pbp_data: 'pl.DataFrame', *, return_as_pandas: 'bool' = False) -> "Union[pl.DataFrame, 'pd.DataFrame']"

Compute expected points for provided plays.

Mirrors nflfastR's calculate_expected_points(). Drops and recomputes any existing ep / *_prob columns so the output is always fresh.

Parameters

Parameter	Type	Default	Description
pbp_data	DataFrame		Play-by-play DataFrame with nflverse columns. Required: season, posteam, home_team, roof, half_seconds_remaining, yardline_100, down, ydstogo, posteam_timeouts_remaining, defteam_timeouts_remaining.
return_as_pandas	bool	False	When True, return a pandas.DataFrame.

Returns

DataFrame with the original columns plus: td_prob, opp_td_prob, fg_prob, opp_fg_prob, safety_prob, opp_safety_prob, no_score_prob, and ep (expected points, clipped to [-10, 10]).

Example

from sportsdataverse.nfl import load_nfl_pbp
from sportsdataverse.nfl.ep_wp import calculate_expected_points

pbp = load_nfl_pbp([2023])
pbp_ep = calculate_expected_points(pbp)
print(pbp_ep.select("ep").head())

calculate_win_probability(pbp_data: 'pl.DataFrame', *, return_as_pandas: 'bool' = False) -> "Union[pl.DataFrame, 'pd.DataFrame']"

Compute win probability for provided plays.

Mirrors nflfastR's calculate_win_probability(). Uses the spread-adjusted model (wp_spread.ubj) when spread_line is non-null, and falls back to the naive model (wp_naive.ubj) for plays with a missing spread line. Drops and recomputes any existing wp / vegas_wp columns.

Parameters

Parameter	Type	Default	Description
pbp_data	DataFrame		Play-by-play DataFrame. Required: all EP columns plus score_differential, game_seconds_remaining, spread_line, receive_2h_ko.
return_as_pandas	bool	False	When True, return a pandas.DataFrame.

Returns

DataFrame with the original columns plus: wp (naive WP) and vegas_wp (spread-adjusted WP).

Example

from sportsdataverse.nfl import load_nfl_pbp
from sportsdataverse.nfl.ep_wp import calculate_win_probability

pbp = load_nfl_pbp([2023])
pbp_wp = calculate_win_probability(pbp)
print(pbp_wp.select("wp", "vegas_wp").head())

calculate_wpa(df: 'pl.DataFrame') -> 'pl.DataFrame'

Derive win probability added (WPA) from pre-scored WP point estimates.

This is the derivation half of NFLPlayProcess.__process_wpa lifted into a shared, model-free function so the same nflfastR-faithful WPA logic can be reused by the streaming enrich_nfl_pbp pipeline and by process_wpa itself. It performs **no** model inference — the caller must already have scored the per-play WP point estimates (wp_spread.ubj) for the start / touchback / end feature views and attached them as wp_before/wp_touchback/wp_after. This mirrors calculate_epa`, which likewise consumes pre-scored EP point estimates and leaves prediction to the orchestrator.

Derivation rules (mirror the original process_wpa`):

• Leading overlay (do not drop): on a kickoff (type.text in kickoff_vec) wp_before is replaced by wp_touchback — the win-probability scored from the touchback feature view — before any other column derives. This is the WP analogue of the EPA 0.92 scoring-attempt overlay and must fire first.
• def_wp_before = 1 - wp_before; home_wp_before / away_wp_before are the posteam->home perspective columns (the offense's wp_before flows to home when the start possession team is the home team, otherwise to the defense def_wp_before).
• wp_after is rewritten by the end-of-half / end-of-game / OT two-path: timeouts hold wp_before; a completed final play resolves to 1.0 / 0.0 by the winner; end-of-half and End Period / End of Half lead plays take lead_wp_before (or 1 - lead_wp_before on a possession change); a possession change otherwise flips the lead; everything else keeps the model wp_after.
• def_wp_after = 1 - wp_after; home_wp_after / away_wp_after use the end possession team for the perspective flip.
• wpa = wp_after - wp_before.

Every shift / forward reference is grouped .over("game_id") so a concatenated multi-game frame never leaks WP across game boundaries — the lead_wp_before / lead_wp_before2 shifts and the end-of-game game_play_number == max() lookup are all per-game. This differs from process_wpa` (which runs one game per instance and therefore needs no grouping).

Parameters

Parameter	Type	Default	Description
df	DataFrame		Play-by-play DataFrame that already carries the WP point estimates wp_before (start feature view), wp_touchback (touchback feature view) and wp_after (end feature view), plus the play-classification / perspective columns game_id, type.text, homeTeamId, start.pos_team.id, end.pos_team.id, start.pos_team_receives_2H_kickoff, change_of_pos_team, scoringPlay, kickoff_onside, end_of_half, status_type_completed, pos_score_diff_end, lead_play_type, lead_pos_team and game_play_number. See WPA_REQUIRED_COLUMNS. This function does **not** score WP itself — score it first via calculate_win_probability/ thewp_spread` feature pipeline.

Returns

The input frame with the WPA derivation applied: wp_before rewritten by the kickoff-touchback overlay; def_wp_before, home_wp_before, away_wp_before, lead_wp_before, lead_wp_before2, the rewritten wp_after, def_wp_after, home_wp_after, away_wp_after and wpa added; plus first-class lowercase aliases wp (= wp_before), def_wp (= def_wp_before), home_wp (= home_wp_before) and away_wp (= away_wp_before) for downstream contract parity (the per-play offense win probability is the pre-snap wp_before, matching nflfastR's wp semantics).

Example

# For most use cases, call the high-level entry point instead. ``enrich_nfl_pbp`` scores WP, derives WPA, and adds EP/EPA/CP/CPOE in one shot on any nflverse-shape frame

    from sportsdataverse.nfl import load_nfl_pbp
    from sportsdataverse.nfl.ep_wp import enrich_nfl_pbp

    pbp = load_nfl_pbp([2023])
    enriched = enrich_nfl_pbp(pbp)
    print(enriched.select("game_id", "wp", "def_wp", "home_wp", "away_wp", "wpa").head())

``calculate_wpa`` directly requires ESPN-internal columns
(``wp_before``, ``wp_touchback``, ``wp_after``, ``homeTeamId``,
``start.pos_team.id``, etc.) produced by ``NFLPlayProcess``.  It is
called internally by ``NFLPlayProcess.__process_wpa`` and by the
``enrich_nfl_pbp`` orchestrator — a naked
``calculate_wpa(load_nfl_pbp([2023]))`` will raise ``KeyError``
because those columns are absent from a nflverse frame.

calculate_xpass(pbp_data: 'pl.DataFrame', *, models_dir: 'Union[str, None]' = None, return_as_pandas: 'bool' = False) -> "Union[pl.DataFrame, 'pd.DataFrame']"

Compute expected dropback probability (xpass) and pass_oe.

Faithful polars port of nflfastR's add_xpass / prepare_xpass_data (helper_add_xpass.R). Scores a single binary:logistic XGBoost model (17 features, in XPASS_FEATURES order) over the rows that satisfy nflfastR's valid_play filter:

• season >= 2006 (before this the NFL did not mark scrambles), and
• play_type in {"no_play", "pass", "run"}, and
• none of posteam / down / defteam_timeouts_remaining / posteam_timeouts_remaining / yardline_100 / score_differential is null.

The era2..4 + outdoors / retractable / dome dummies and the home indicator are produced by make_cp_mutations(the same nflfastRmake_model_mutationslogic CP uses) rather than re-derived.wp/vegas_wpare the start-of-play win-probability columns and must already be present (run after the WP step / insideenrich_nfl_pbp`).

The booster ships with no embedded feature_names, so the DMatrix is built with XPASS_FEATURES as the column order — feeding the features in any other order silently yields wrong predictions.

Drops and recomputes any existing xpass / pass_oe columns.

Parameters

Parameter	Type	Default	Description
pbp_data	DataFrame		nflverse-format play-by-play DataFrame. Required: season, play_type, posteam, home_team, down, ydstogo, yardline_100, qtr, wp, vegas_wp, score_differential, half_seconds_remaining, posteam_timeouts_remaining, defteam_timeouts_remaining. Optional: roof (for the roof dummies), pass / rush (the 0/1 dropback / rush indicators used by pass_oe).
models_dir	Union[str, None]	None	Optional directory to load xpass_model.ubj from instead of downloading / caching it (offline or custom model).
return_as_pandas	bool	False	When True, return a pandas.DataFrame.

Returns

DataFrame with the original columns plus xpass (predicted pass probability, null outside the valid_play filter; float64) and pass_oe (100 * (pass - xpass), null when xpass is null and null when rush == 0 & pass == 0; float64).

Example

from sportsdataverse.nfl import load_nfl_pbp
from sportsdataverse.nfl.ep_wp import enrich_nfl_pbp, calculate_xpass

pbp = enrich_nfl_pbp(load_nfl_pbp([2023]))  # gives wp / vegas_wp
pbp_xp = calculate_xpass(pbp)
print(pbp_xp.select("xpass", "pass_oe").head())

# Pipeline next step

pbp_xp.filter(pl.col("play_type") == "pass").select("posteam", "xpass", "pass_oe").head()

calculate_xyac(pbp_data: 'pl.DataFrame', *, models_dir: 'Optional[Union[str, Path]]' = None, return_as_pandas: 'bool' = False) -> "Union[pl.DataFrame, 'pd.DataFrame']"

Compute expected yards after catch (xYAC) for intended pass plays.

Faithful polars port of nflfastR's add_xyac. Unlike a per-statistic regressor, xYAC is one multi:softprob model (num_class=76) that predicts a distribution over YAC buckets (yac = -5..70); the five output columns are derived from that distribution by re-scoring expected points on every outcome. ep is not required on the input — it is recomputed on the outcome rows via calculate_expected_points. The play's pre-snap ep (original_ep) is the EPA baseline; air_epa is also part of the baseline (xyac_epa = Σ((ep − original_ep)·prob) − air_epa). air_epa is optional: when present (the nflverse path) it is used verbatim so parity is byte-for-byte preserved; when absent (the Shield-native / ESPN path) it is computed from the already-scored yac == 0 (catch-spot) outcome — air_epa = ep(yac == 0) − original_ep — and, since it was genuinely missing, surfaced as an extra air_epa output column.

Inference filter (nflfastR valid_pass ∧ distance_to_goal != 0): complete_pass == 1 OR incomplete_pass == 1 OR interception == 1, air_yards in [-15, 70), non-null receiver_player_name and pass_location, and distance_to_goal != 0. Non-qualifying rows receive null in all five columns. Drops and recomputes any existing xYAC output columns.

The xYAC model (xyac_model.ubj, ~34 MB) is not bundled in the wheel: on first use it is downloaded from the nfl_model_artifacts GitHub release and cached under <cache_dir>/models/ (see sportsdataverse.nfl.get_config). Subsequent calls load it from the cache; clear_cache() deliberately preserves the models/ subdir so a data-cache clear does not force a re-download. Pass models_dir= to point at a local directory containing xyac_model.ubj (offline / custom model override). If the model is genuinely unavailable (no cache + no network) the underlying loader raises FileNotFoundError.

Parameters

Parameter	Type	Default	Description
pbp_data	DataFrame		nflverse-format play-by-play DataFrame. Required: air_yards, season, half_seconds_remaining, yardline_100, ydstogo, down, posteam, home_team, roof, ep, posteam_timeouts_remaining, defteam_timeouts_remaining, complete_pass, incomplete_pass, interception, pass_location, receiver_player_name. Optional: air_epa (used verbatim when present for byte-for-byte nflverse parity; computed from the yac == 0 outcome and added as an output column when absent), qb_hit.
models_dir	Optional[Union[str, Path]]	None	Optional directory to load xyac_model.ubj from instead of downloading/caching it (offline use or a custom-trained model). When None (default) the model is resolved bundled → cache → downloaded-from-release.
return_as_pandas	bool	False	When True, return a pandas.DataFrame.

Returns

DataFrame with the original columns plus the five nflfastR xYAC columns (Float64, null on non-qualifying rows): xyac_epa, xyac_mean_yardage, xyac_median_yardage, xyac_success, xyac_fd. When the input lacked air_epa and at least one qualifying pass was scored, a computed air_epa column (catch-spot air EPA) is also added.

Example

import polars as pl

from sportsdataverse.nfl import load_nfl_pbp
from sportsdataverse.nfl.ep_wp import calculate_xyac

pbp = load_nfl_pbp([2023])
pbp = calculate_xyac(pbp)
print(pbp.select("xyac_epa", "xyac_mean_yardage").head())

# Pipeline next step (one line)

pbp.filter(pl.col("xyac_epa").is_not_null()).select("xyac_epa", "xyac_fd").head()

clear_cache() -> 'None'

Clear both memory and filesystem caches.

Memory: empties the in-process dict. Filesystem: removes all entries under config.cache_dir. The directory itself is preserved so subsequent writes succeed without needing mkdir.

The models/ subdirectory is deliberately preserved — it holds download-on-demand model artifacts (e.g. the ~34 MB xyac_model.ubj) that are expensive to re-fetch. Clearing the data cache should not force a model re-download; delete <cache_dir>/models/ by hand to drop those.

Example

from sportsdataverse.nfl import clear_cache, load_nfl_pbp
clear_cache()
pbp = load_nfl_pbp(seasons=[2024])

# Pair with a cache-mode switch

from sportsdataverse.nfl import clear_cache, update_config
update_config(cache_mode="filesystem")
# ... lots of cached calls accumulate parquet files on disk ...
clear_cache()  # wipe disk + memory together

espn_nfl_teams(return_as_pandas=False, **kwargs) -> 'pl.DataFrame'

espn_nfl_teams - look up NFL teams

Parameters

Parameter	Type	Default	Description
return_as_pandas	bool	False	If True, returns a pandas dataframe. If False, returns a polars dataframe.

Returns

Polars dataframe containing teams for the requested league. This function caches by default, so if you want to refresh the data, use the command sportsdataverse.nfl.espn_nfl_teams.clear_cache().

col_name	type	description
team_abbreviation	character	Short team abbreviation (e.g. 'LAS').
team_alternate_color	character	Team alternate color (hex without leading '#').
team_color	character	Team primary color (hex without leading '#').
team_display_name	character	Full team display name.
team_id	character	Unique team identifier.
team_is_active	logical	TRUE if the team is currently active.
team_is_all_star	logical	TRUE if the row represents an All-Star team.
team_location	character	Team city or location string.
team_logos	integer	Team logo metadata.
team_name	character	Full team display name (e.g. 'Las Vegas Aces').
team_nickname	character	Team nickname.
team_short_display_name	character	Short team display name (e.g. 'Aces').
team_slug	character	URL-safe team identifier (e.g. 'lasvegas-aces' / 'aces').
team_uid	character	ESPN universal team identifier (UID format 's:40~l:...~t:...').

Example

from sportsdataverse.nfl import espn_nfl_teams
teams = espn_nfl_teams()
teams.shape

# Pandas round-trip

teams_pd = espn_nfl_teams(return_as_pandas=True)
teams_pd[["team_abbreviation", "team_display_name"]].head()

# Force a refresh after upstream ESPN updates

espn_nfl_teams.cache_clear()  # underlying lru_cache
teams = espn_nfl_teams()

get_2pt_wp(pbp_df: "Union[pl.DataFrame, 'pd.DataFrame']") -> 'pd.DataFrame'

Win probability of the PAT-vs-2pt choice after a touchdown (nfl4th get_2pt_wp).

For each row, scores the post-touchdown state under three scoring outcomes (0 / 1 / 2 added points) from the kicking-off team's ensuing-drive WP, and combines them with the 2-pt conversion probability (two_pt_model) and the PAT make probability (the FG model at yardline_100 = 15) into wp_td — the better of go-for-2 and kick-the-PAT.

Parameters

Parameter	Type	Default	Description
pbp_df	Union[DataFrame, 'DataFrame']		Play-by-play frame (polars or pandas) of post-touchdown states, already carrying the prepared state columns (see module docstring).

Returns

A pandas frame with go_index, yardline_100 (always 0) and wp_td. wp_td is NaN when the WP / 2-pt models are unavailable.

Example

from sportsdataverse.nfl.nfl_fourth_down import get_2pt_wp
out = get_2pt_wp(touchdown_states)
print(out[["go_index", "wp_td"]].head())

get_4th_down_probs(pbp_df: "Union[pl.DataFrame, 'pd.DataFrame']") -> 'pd.DataFrame'

Full 4th-down decision surface (nfl4th add_4th_probs) + recommendation.

Runs get_go_wp, get_fg_wp, get_punt_wp on the fourth-down rows and adds the combined option columns plus:

• go_boost -- nfl4th's headline number: 100 * (go_wp - max(fg_wp, punt_wp)) in percentage points (a NaN punt_wp is treated as 0).
• fourth_down_recommendation -- the max-WP choice among {go, punt, field_goal} (NaN options are excluded).
• go_wp_diff / punt_wp_diff / fg_wp_diff -- each option's WP minus the recommended option's WP (the recommended option's diff is 0, the others <= 0). NaN where the option WP is NaN.

Parameters

Parameter	Type	Default	Description
pbp_df	Union[DataFrame, 'DataFrame']		Play-by-play frame (polars or pandas) of fourth-down situations (the nflverse-shape output of load_nfl_pbp; see module docstring for required columns).

Returns

A pandas copy of pbp_df with the decision columns added. Empty input returns the input plus empty decision columns.

Example

from sportsdataverse.nfl import load_nfl_pbp
from sportsdataverse.nfl.nfl_fourth_down import get_4th_down_probs

pbp = load_nfl_pbp([2023])
fourth = pbp.filter((pl.col("down") == 4) & pl.col("yardline_100").is_not_null())
out = get_4th_down_probs(fourth)
print(out[["go_wp", "punt_wp", "fg_wp", "go_boost", "fourth_down_recommendation"]].head())

get_config() -> 'NflConfig'

Return the live NflConfig singleton.

The same object is returned on every call; mutate via update_config rather than reassigning fields directly so future hooks (e.g. logging on config change) have a single choke point.

Example

from sportsdataverse.nfl import get_config
cfg = get_config()
print(cfg.cache_mode, cfg.cache_duration, cfg.cache_dir)

# Pair with ``update_config`` to verify a change took effect

from sportsdataverse.nfl import update_config, get_config
update_config(cache_mode="off")
assert get_config().cache_mode == "off"

get_fg_wp(pbp_df: "Union[pl.DataFrame, 'pd.DataFrame']") -> 'pd.DataFrame'

Expected win probability of attempting a field goal (nfl4th get_fg_wp).

The make probability comes from the self-trained fg_model (a binary:logistic XGBoost re-train of the original mgcv GAM, features [yardline_100, fg_roof, fg_era]), shrunk by 0.9 for kicks at/beyond yardline_100 = 38 and zeroed at/beyond yardline_100 = 45 (>= ~63-yard kicks). The made-FG state (opponent receives a touchback kickoff at the 25, kicking team +3) and the missed-FG state (opponent takes over 8 yards back of the spot, capped at the 80) are each scored with win probability; fg_wp = make_prob * make_wp + (1 - make_prob) * miss_wp.

Parameters

Parameter	Type	Default	Description
pbp_df	Union[DataFrame, 'DataFrame']		Play-by-play frame (polars or pandas) of fourth-down situations.

Returns

A pandas copy of pbp_df plus fg_make_prob, make_fg_wp, miss_fg_wp and fg_wp (from the kicking team's perspective). All four are NaN when the FG model or WP model is unavailable.

Example

from sportsdataverse.nfl import load_nfl_pbp
from sportsdataverse.nfl.nfl_fourth_down import get_fg_wp

pbp = load_nfl_pbp([2023])
fourth = pbp.filter((pl.col("down") == 4) & pl.col("yardline_100").is_not_null())
out = get_fg_wp(fourth)
print(out[["fg_make_prob", "fg_wp"]].head())

get_go_wp(pbp_df: "Union[pl.DataFrame, 'pd.DataFrame']") -> 'pd.DataFrame'

Expected win probability of going for it on 4th down (nfl4th get_go_wp).

The fd_model 76-class yards-gained distribution is expanded per play; each outcome's hypothetical post-play game state (turnover-on-downs flip, +6 touchdown with the PAT/2-pt branch routed through get_2pt_wp, 6-second runoff, goal-to-go distance shrink) is scored with win probability and the end-of-game kneel-out clamps are applied; the option value is the prob-weighted WP.

Parameters

Parameter	Type	Default	Description
pbp_df	Union[DataFrame, 'DataFrame']		Play-by-play frame (polars or pandas) of fourth-down situations carrying the prepared state columns (see module docstring). The frame is prepared internally if it lacks the derived columns.

Returns

A pandas copy of pbp_df plus go_wp (prob-weighted WP of going for it), first_down_prob (P(conversion)), wp_succeed (mean WP over conversion outcomes) and wp_fail (mean WP over failure outcomes). All are NaN when the fourth-down / WP models are unavailable (FD_MODEL_AVAILABLE / WP_MODEL_AVAILABLE).

Example

from sportsdataverse.nfl import load_nfl_pbp
from sportsdataverse.nfl.nfl_fourth_down import get_go_wp

pbp = load_nfl_pbp([2023])
fourth = pbp.filter((pl.col("down") == 4) & pl.col("yardline_100").is_not_null())
out = get_go_wp(fourth)
print(out[["go_wp", "first_down_prob"]].head())

get_punt_wp(pbp_df: "Union[pl.DataFrame, 'pd.DataFrame']") -> 'pd.DataFrame'

Expected win probability of punting on 4th down (nfl4th get_punt_wp).

The punt landing distribution (punt_data: yardline_after / pct / muff per yardline_100) is joined per play; possession is flipped to the receiving team, with return-touchdown (yardline_after == 100) and muff (muff == 1) recoveries flipping the ball back to the punting team; each landing spot's ensuing-drive WP is scored and the option value is the prob-weighted WP from the punting team's perspective.

Parameters

Parameter	Type	Default	Description
pbp_df	Union[DataFrame, 'DataFrame']		Play-by-play frame (polars or pandas) of fourth-down situations.

Returns

A pandas copy of pbp_df plus punt_wp. punt_wp is NaN where the punt distribution has no support for the play's yardline_100 (inside the punting team's own 31, where the table is empty — matching the R reference's left-join NA behavior) or when the WP model is unavailable.

Example

from sportsdataverse.nfl import load_nfl_pbp
from sportsdataverse.nfl.nfl_fourth_down import get_punt_wp

pbp = load_nfl_pbp([2023])
fourth = pbp.filter((pl.col("down") == 4) & pl.col("yardline_100").is_not_null())
out = get_punt_wp(fourth)
print(out[["punt_wp"]].head())

nfl_clear_token_cache() -> 'None'

Drop the cached api.nfl.com token (forces a fresh mint on the next call).

nfl_game_details(game_id: 'Optional[str]' = None, headers: 'Optional[Dict[str, str]]' = None, raw: 'bool' = False) -> 'Dict'

Pull full api.nfl.com game details (drives + plays) by game id.

Hits /experience/v1/gamedetails/{game_id}; the payload is the shield data.viewer.gameDetail object (plays, drives, scoring summaries, line scores, possession, weather, attendance, ...).

Parameters

Parameter	Type	Default	Description
game_id	str	None	the uuid game id from nfl_game_schedule (e.g. '7d3e8f84-1312-11ef-afd1-646009f18b2e').
headers	Dict[str, str] | None	None	Pre-built header dict. Defaults to a fresh nfl_headers_gen call.
raw	bool	False	If True, return the full envelope ({"data": {...}}) untouched. If False (default), unwrap to the gameDetail object.

Returns

the gameDetail object (or the raw envelope when raw=True). Empty dict if the game has no detail payload.

Example

from sportsdataverse.nfl.nfl_games import nfl_game_details
detail = nfl_game_details(game_id="7d3e8f84-1312-11ef-afd1-646009f18b2e")
len(detail["plays"]), len(detail["drives"])

# Reuse headers across many calls (avoids re-minting tokens)

from sportsdataverse.nfl.nfl_games import nfl_game_details, nfl_headers_gen
hdrs = nfl_headers_gen()
detail = nfl_game_details(game_id="7d3e8f84-1312-11ef-afd1-646009f18b2e", headers=hdrs)

nfl_game_pbp(game_id: 'Optional[str]' = None, headers: 'Optional[Dict[str, str]]' = None, return_as_pandas: 'bool' = False)

Parsed api.nfl.com play-by-play -- one row per play (polars/pandas frame).

Tidy wrapper over nfl_game_details: flattens gameDetail.plays into a DataFrame (playId, quarter, down, yardsToGo, yardLine, playType, playDescription, possessionTeam_*, ...) and prepends the game context (game_id, home_team, visitor_team).

Parameters

Parameter	Type	Default	Description
game_id	str	None	uuid game id from nfl_week_games / nfl_game_schedule.
headers	Optional[Dict[str, str]]	None	reuse a nfl_headers_gen dict.
return_as_pandas	bool	False	return a pandas frame instead of polars.

Returns

A polars (or pandas) DataFrame, one row per play (empty frame if the game has no play-by-play yet).

col_name	type	description
game_id	character	Ten digit identifier for NFL game.
home_team	character	The home team. Note that this contains the designated home team for games which no team is playing at home such as Super Bowls or NFL International games.
visitor_team	character	Abbreviation or name of the visiting (away) team in this game.
clockTime	character	Game clock time at the start of the play, formatted as MM:SS within the quarter.
down	integer	The down for the given play.
driveNetYards	integer	Net yards gained on the current drive up to and including this play.
drivePlayCount	integer	Number of offensive plays run on the current drive up to and including this play.
driveSequenceNumber	integer	Sequential identifier for the drive within the game, starting at 1.
driveTimeOfPossession	character	Elapsed time of possession for the current drive, formatted as MM:SS.
endClockTime	character	Game clock time at the end of the play, formatted as MM:SS within the quarter.
endYardLine	character	Yard line where the ball was placed at the conclusion of the play.
firstDown	logical	Indicates whether the play resulted in a first down.
goalToGo	logical	Indicates whether the offensive team is in a goal-to-go situation at the start of the play.
isBigPlay	character	Classification label indicating whether this play is designated as a big play by the NFL.
latestPlay	character	Indicates whether this play is the most recent play recorded in the live feed.
nextPlayIsGoalToGo	logical	Indicates whether the next play will begin in a goal-to-go situation.
nextPlayType	character	Play type category anticipated or assigned to the next play in sequence.
orderSequence	integer	Sequential ordering value used to sort plays within the game in chronological order.
penaltyOnPlay	logical	Indicates whether a penalty was called on this play.
playClock	integer	Play clock value in seconds at the snap of the ball.
playReviewStatus	character	Status of any official review of the play (e.g., confirmed, overturned, stands).
playDeleted	logical	Indicates whether this play record has been marked as deleted or voided.
playDescription	character	Official text description of the play as provided by the NFL.
playDescriptionWithJerseyNumbers	character	Play description text augmented with player jersey numbers for participant identification.
playId	integer	Unique play event identifier (UUID).
playStats	integer	Internal NFL stat identifier linking this play to associated statistical records.
playType	character	Categorical classification of the play type (e.g., PASS, RUSH, PUNT, KICKOFF).
prePlayByPlay	character	Narrative text describing the game situation or setup immediately before this play.
quarter	integer	Game quarter in which the play occurred (1–4 for regulation, 5 for overtime).
scoringPlay	logical	Indicates whether this play resulted in points being scored.
scoringPlayType	character	Type of scoring event on this play (e.g., TD, FG, SAFETY, PAT).
scoringTeam	character	Abbreviation or identifier of the team that scored on this play, if applicable.
shortDescription	character	Short narrative summary of the play outcome (e.g., 'PENALTY', 'TOUCHDOWN').
specialTeamsPlay	logical	Indicates whether this play was a special teams play.
stPlayType	character	Special teams play sub-type classification (e.g., PUNT, KICKOFF, FG_ATTEMPT) when applicable.
timeOfDay	character	Wall-clock time of day when the play occurred, typically in local or Eastern time.
yardLine	character	Yard line on the field where the ball was placed at the start of the play.
yards	integer	The number of receiving yards
yardsToGo	integer	Number of yards needed to gain a first down at the start of the play.
possessionTeam_id	character	NFL.com Shield API identifier for the team with offensive possession on this play.
possessionTeam_abbreviation	character	Abbreviated team name for the team with offensive possession on this play.
possessionTeam_nickName	character	Nickname (mascot name) of the team with offensive possession on this play.
possessionTeam_franchise_primaryColor	character	Primary brand color for the possessing team's franchise, expressed as a hex color code.
possessionTeam_franchise_secondaryColor	character	Secondary brand color for the possessing team's franchise, expressed as a hex color code.
possessionTeam_franchise_tertiaryColor	character	Tertiary brand color for the possessing team's franchise, expressed as a hex color code.
possessionTeam_franchise_currentLogo	character	URL of the current primary logo for the possessing team's franchise.
possessionTeam_franchise_largeTypeColor	character	Large-type display color for the possessing team's franchise branding, as a hex code.
possessionTeam_franchise_decorativeElementsColor	character	Decorative elements color for the possessing team's franchise branding, as a hex code.
scoringTeam_id	character	NFL.com Shield API identifier for the team that scored on this play.
scoringTeam_abbreviation	character	Abbreviated team name for the team that scored on this play.
scoringTeam_nickName	character	Nickname (mascot name) of the team that scored on this play.
scoringTeam_franchise_primaryColor	character	Primary brand color for the scoring team's franchise, expressed as a hex color code.
scoringTeam_franchise_secondaryColor	character	Secondary brand color for the scoring team's franchise, expressed as a hex color code.
scoringTeam_franchise_tertiaryColor	character	Tertiary brand color for the scoring team's franchise, expressed as a hex color code.
scoringTeam_franchise_currentLogo	character	URL of the current primary logo for the scoring team's franchise.
scoringTeam_franchise_largeTypeColor	character	Large-type display color for the scoring team's franchise branding, as a hex code.
scoringTeam_franchise_decorativeElementsColor	character	Decorative elements color for the scoring team's franchise branding, as a hex code.

Example

from sportsdataverse.nfl import nfl_game_pbp
pbp = nfl_game_pbp(game_id="7d3e8f84-1312-11ef-afd1-646009f18b2e")
pbp.select(["quarter", "down", "yardsToGo", "playType", "playDescription"]).head()

nfl_game_schedule(season: 'int' = 2024, season_type: 'str' = 'REG', week: 'int' = 1, headers: 'Optional[Dict[str, str]]' = None, raw: 'bool' = False) -> 'Dict'

List api.nfl.com games for a season/week slice (/football/v2/games).

Parameters

Parameter	Type	Default	Description
season	int	2024	season year (e.g. 2024).
season_type	str	'REG'	season type. One of "PRE", "REG", "POST".
week	int	1	week number (1-18 regular season, 1-4 post-season).
headers	Dict[str, str] | None	None	Pre-built header dict (skip the auth roundtrip). Defaults to a fresh nfl_headers_gen call.
raw	bool	False	currently ignored; the function returns the parsed JSON payload.

Returns

payload with the games list under "games" plus "pagination". Each game carries id (the uuid game id used by nfl_game_details), homeTeam/awayTeam, date, status, externalIds (gsis etc.).

Example

from sportsdataverse.nfl.nfl_games import nfl_game_schedule
week_one = nfl_game_schedule(season=2024, season_type="REG", week=1)
first_id = week_one["games"][0]["id"]

nfl_headers_gen(token: 'Optional[str]' = None) -> 'Dict[str, str]'

Build the request-header dict expected by api.nfl.com.

Obtains a bearer token via nfl_token_gen (which caches + auto-renews, or honors NFL_ACCESS_TOKEN) unless token is supplied, and combines it with the browser-style headers the NFL.com web app sends. Token caching already avoids re-minting, so callers rarely need to thread token/headers by hand.

Parameters

Parameter	Type	Default	Description
token	Optional[str]	None	An existing access token to reuse; uses the cached/minted one when None.

Returns

Header dict ready to drop into requests.get.

Example

from sportsdataverse.nfl.nfl_games import nfl_headers_gen, nfl_game_schedule
hdrs = nfl_headers_gen()
week_one = nfl_game_schedule(season=2024, season_type="REG", week=1, headers=hdrs)
week_two = nfl_game_schedule(season=2024, season_type="REG", week=2, headers=hdrs)

nfl_ngs_gamecenter_overview(game_id, group: 'str' = 'passers', return_as_pandas: 'bool' = False)

NGS gamecenter overview for one game -- one row per player on a side.

Wraps /api/gamecenter/overview (keyed by NGS gameId). The payload splits each stat group into home and visitor entries; this function stacks both and tags every row with side ("home"/"visitor") plus the game's gameId. Note passers carries a single primary QB per side (two rows total) while rushers/receivers/passRushers are full lists.

Parameters

Parameter	Type	Default	Description
game_id			NGS gameId (e.g. "2024090500") from nfl_ngs_league_schedule.
group	str	'passers'	which player group -- one of "passers", "rushers", "receivers", "passRushers".
return_as_pandas	bool	False	return a pandas frame instead of polars.

Returns

A polars (or pandas) DataFrame, one row per player (both teams), with side and gameId columns prepended.

col_name	type	description
side	character	"home" or "visitor" -- which team's roster the row belongs to.
gameId	character	Unique identifier for the NFL game in the NGS/Shield data system.
esbId	character	Elias Sports Bureau identifier for the player, used to link to official NFL records and statistics.
teamId	character	Unique numeric identifier for the player's team in the NFL NGS/Shield data system.
teamAbbr	character	Two- or three-letter abbreviation identifying the player's team (e.g., 'KC', 'PHI').
shortName	character	Abbreviated display name of the player (e.g., 'P.Mahomes') used in compact UI contexts.
position	character	Primary position as reported by NFL.com
jerseyNumber	integer	Jersey number worn by the player during the game.
playerName	character	Full display name of the player as shown in the NFL Next Gen Stats gamecenter.
zones	integer	Serialized or structured representation of field zones targeted or covered by the player during the game.
passYards	integer	Total passing yards accumulated by the player (relevant for quarterbacks) in the NGS gamecenter overview.
touchdowns	integer	Total number of touchdowns scored or thrown by the player in the NGS gamecenter overview.
interceptions	integer	The number of interceptions thrown.
attempts	integer	The number of pass attempts as defined by the NFL.
completions	integer	The number of completed passes.
headshot	character	NFL headshot url for player

Example

from sportsdataverse.nfl import nfl_ngs_gamecenter_overview
ov = nfl_ngs_gamecenter_overview(game_id="2024090500", group="passers")
ov.select(["side", "playerName", "position"]).head()

nfl_ngs_leaders(category: 'str' = 'speed', season: 'int' = 2024, season_type: 'str' = 'REG', week: 'Optional[int]' = None, return_as_pandas: 'bool' = False)

NGS top-N "leaders" board for a single category (one row per leader play).

One parameterized wrapper over the single-list leader endpoints. Each record nests a leader (player/stat) block and a play (the play that produced the highlight) block, flattened to leader_* / play_* columns.

Categories (category= value -> endpoint):

• "speed" -> /leaders/speed/ballCarrier (fastest ball-carrier speeds)
• "distance_ballcarrier" -> /leaders/distance/ballCarrier
• "distance_tackle" -> /leaders/distance/tackle
• "time_sack" -> /leaders/time/sack
• "completion_season" / "completion_week" -> /leaders/expectation/completion/{season,week} (most-improbable completions)
• "ery_season" / "ery_week" -> /leaders/expectation/ery/{season,week} (expected rush yards over expectation)
• "yac_season" / "yac_week" -> /leaders/expectation/yac/{season,week} (yards after catch over expectation)

Parameters

Parameter	Type	Default	Description
category	str	'speed'	one of the keys above. Defaults to "speed".
season	int	2024	season year.
season_type	str	'REG'	"REG", "POST", or "PRE".
week	int | None	None	week filter -- required (and only used) by the *_week categories; ignored by season/non-expectation boards.
return_as_pandas	bool	False	return a pandas frame instead of polars.

Returns

A polars (or pandas) DataFrame, one row per leader entry.

col_name	type	description
leader_esbId	character	Elias Sports Bureau identifier for the statistical leader, used to link to official NFL records.
leader_firstName	character	First name of the statistical leader player.
leader_gsisId	character	NFL GSIS (Game Statistics and Information System) identifier for the statistical leader player.
leader_jerseyNumber	integer	Jersey number worn by the statistical leader player.
leader_lastName	character	Last name of the statistical leader player.
leader_playerName	character	Full display name of the statistical leader player as shown in NFL NGS.
leader_position	character	Specific position designation of the statistical leader (e.g., 'QB', 'WR', 'CB').
leader_positionGroup	character	Broader position group of the statistical leader (e.g., 'Offense', 'Defense', 'Special Teams').
leader_shortName	character	Abbreviated display name of the statistical leader (e.g., 'P.Mahomes') for compact UI usage.
leader_teamAbbr	character	Two- or three-letter abbreviation identifying the statistical leader's team.
leader_teamId	character	Unique numeric identifier for the statistical leader's team in the NFL NGS/Shield system.
leader_week	integer	NFL week number during which the statistical leader achieved the highlighted performance.
leader_yards	integer	Total yards associated with the statistical leader's highlighted play or performance metric.
leader_inPlayDist	double	Total in-play distance traveled (in yards) by the statistical leader during the highlighted play.
leader_maxSpeed	double	Maximum recorded speed (in miles per hour) reached by the statistical leader during the highlighted play.
leader_headshot	character	URL to the player's official headshot image as provided by the NFL NGS system.
play_gameId	integer	Unique identifier for the NFL game in which the highlighted play occurred.
play_playId	integer	Unique identifier for the specific play within the game in the NFL NGS/Shield system.
play_sequence	integer	Sequential order number of the play within the game or drive as recorded in the NGS system.
play_down	integer	Down number (1–4) on which the highlighted play occurred.
play_gameClock	character	Game clock time (MM:SS) at the start of the highlighted play.
play_gameKey	integer	Alternate numeric key for the NFL game in which the highlighted play occurred, used in official NFL systems.
play_health_playerTracking	character	Indicator of whether player-tracking data from the NGS health/tracking system is available for this play.
play_health_ballTracking	character	Indicator of whether ball-tracking data from the NGS health/tracking system is available for this play.
play_homeScore	integer	Home team's score at the time of the highlighted play.
play_isBigPlay	logical	Boolean flag indicating whether the highlighted play is classified as a 'big play' in the NGS system.
play_isEndQuarter	logical	Boolean flag indicating whether the highlighted play occurred at the end of a quarter.
play_isGoalToGo	logical	Boolean flag indicating whether the highlighted play occurred in a goal-to-go situation.
play_isPenalty	logical	Boolean flag indicating whether the highlighted play involved a penalty.
play_isSTPlay	logical	Boolean flag indicating whether the highlighted play was a special teams play.
play_isScoring	logical	Boolean flag indicating whether the highlighted play resulted in a score.
play_playDescription	character	Full text description of the highlighted play as recorded by official NFL scorers.
play_playState	character	State or status of the highlighted play (e.g., 'APPROVED', 'CHALLENGED') in the NGS system.
play_playStats	integer	Serialized statistical outcomes or stat codes associated with the highlighted play.
play_playType	character	Text label describing the type of the highlighted play (e.g., 'PASS', 'RUSH', 'KICK').
play_playTypeCode	integer	Numeric or short code identifying the play type of the highlighted play in the NGS system.
play_possessionTeamId	character	Unique identifier for the team that had possession of the ball during the highlighted play.
play_preSnapHomeScore	integer	Home team's score immediately before the snap on the highlighted play.
play_preSnapVisitorScore	integer	Visitor team's score immediately before the snap on the highlighted play.
play_quarter	integer	Quarter (1–4, or 5 for overtime) in which the highlighted play occurred.
play_timeOfDayUTC	character	Wall-clock timestamp in UTC representing the real-world time the highlighted play occurred.
play_visitorScore	integer	Visitor team's score at the time of the highlighted play.
play_yardline	character	Formatted yard line string (e.g., 'KC 35') indicating the field position of the highlighted play.
play_yardlineNumber	integer	Numeric yard line (1–50) indicating the field position of the highlighted play.
play_yardlineSide	character	Team abbreviation indicating which team's side of the field the highlighted play occurred on.
play_yardsToGo	integer	Number of yards needed for a first down at the start of the highlighted play.
play_absoluteYardlineNumber	integer	Absolute yard line number (1–100) on the field where the highlighted play occurred.
play_actualYardlineForFirstDown	double	Yard line the offense must reach to convert a first down on the highlighted play.
play_actualYardsToGo	double	Actual distance in yards the offense needed to gain a first down on the highlighted play.
play_endGameClock	character	Game clock time (MM:SS) at the end of the highlighted play.
play_isChangeOfPossession	logical	Boolean flag indicating whether the highlighted play resulted in a change of ball possession.
play_playDirection	character	Direction of the highlighted play on the field (e.g., left, middle, right).
play_startGameClock	character	Game clock time (MM:SS) at the start of the highlighted play.

Example

from sportsdataverse.nfl import nfl_ngs_leaders
fast = nfl_ngs_leaders(category="speed", season=2024, season_type="REG")
fast.select(["leader_playerName", "leader_maxSpeed", "play_playDescription"]).head()

nfl_ngs_league_schedule(season: 'int' = 2024, season_type: 'str' = 'REG', week: 'Optional[int]' = None, return_as_pandas: 'bool' = False)

NGS league schedule -- one row per game; source of NGS gameId values.

Wraps /api/league/schedule (which returns a top-level list of games). Each row carries gameId (the YYYYMMDDNN id used by the game-scoped functions here), gameKey, smartId (the api.nfl.com uuid), team abbreviations/ids/names, kickoff times, ngsGame (tracking-data flag) and season/seasonType/week.

Parameters

Parameter	Type	Default	Description
season	int	2024	season year.
season_type	str	'REG'	"REG", "POST", or "PRE".
week	int | None	None	optional single-week filter.
return_as_pandas	bool	False	return a pandas frame instead of polars.

Returns

A polars (or pandas) DataFrame, one row per scheduled game.

col_name	type	description
gameKey	integer	Legacy NFL game key used as an alternative identifier in the NGS scheduling system.
gameDate	character	Game date-time (ISO 8601, UTC).
gameId	integer	NFL Next Gen Stats integer identifier for the game.
gameTime	character	Scheduled kickoff time for the game in local or UTC format.
gameTimeEastern	character	Scheduled kickoff time for the game expressed in Eastern Time.
gameType	character	Game type identifier (3 for playoffs).
homeDisplayName	character	Full display name of the home team (e.g., 'Kansas City Chiefs').
homeNickname	character	Nickname (mascot name) of the home team (e.g., 'Chiefs').
homeTeamAbbr	character	Abbreviated team name for the home team at the schedule row level.
homeTeamId	character	NFL Next Gen Stats team identifier for the home team at the schedule row level.
isoTime	integer	Scheduled kickoff time expressed as a Unix timestamp (milliseconds since epoch).
networkChannel	character	Television network or channel broadcasting this game (e.g., 'NBC', 'ESPN').
ngsGame	logical	Indicates whether this game has full Next Gen Stats data collection enabled.
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
seasonType	character	Season segment in which the game occurs (e.g., 'REG' for regular season, 'POST' for playoffs).
smartId	character	NFL Next Gen Stats smart (UUID-style) identifier for this scheduled game.
visitorDisplayName	character	Full display name of the visiting (away) team (e.g., 'San Francisco 49ers').
visitorNickname	character	Nickname (mascot name) of the visiting team (e.g., '49ers').
visitorTeamAbbr	character	Abbreviated team name for the visiting team at the schedule row level.
visitorTeamId	character	NFL Next Gen Stats team identifier for the visiting team at the schedule row level.
week	integer	Season week.
weekNameAbbr	character	Abbreviated label for the week of the season (e.g., 'WK1', 'WC' for Wild Card).
liveDotsGame	logical	Indicates whether this game has live player-tracking dot data available via NGS.
validated	logical	Indicates whether the schedule entry has been validated and confirmed by the NFL.
releasedToClubs	logical	Indicates whether NGS data for this game has been released to team personnel.
homeTeam_teamId	character	NFL Next Gen Stats integer team identifier for the home team from the nested team object.
homeTeam_smartId	character	NFL Next Gen Stats smart (UUID-style) identifier for the home team.
homeTeam_logo	character	URL of the home team's logo from the nested team object.
homeTeam_abbr	character	Abbreviated team name for the home team from the nested team object.
homeTeam_cityState	character	City and state of the home team as provided in the nested team object.
homeTeam_fullName	character	Full official name of the home team from the nested team object.
homeTeam_nick	character	Nickname of the home team from the nested team object.
homeTeam_teamType	character	Classification of the home team type (e.g., 'NFL' for active franchises).
homeTeam_conferenceAbbr	character	Conference abbreviation for the home team from the nested team object (e.g., 'AFC').
homeTeam_divisionAbbr	character	Division abbreviation for the home team from the nested team object (e.g., 'AFC West').
site_smartId	character	NFL Next Gen Stats smart (UUID-style) identifier for the game venue.
site_siteId	integer	NFL Next Gen Stats integer identifier for the game venue.
site_siteFullName	character	Full official name of the game venue (e.g., 'Arrowhead Stadium').
site_siteCity	character	City where the game venue is located.
site_siteState	character	State (or country for international games) where the game venue is located.
site_postalCode	character	Postal (ZIP) code of the venue where the game is played.
site_roofType	character	Roof type of the game venue (e.g., 'OUTDOOR', 'DOME', 'RETRACTABLE').
visitorTeam_teamId	character	NFL Next Gen Stats integer team identifier for the visiting team from the nested team object.
visitorTeam_smartId	character	NFL Next Gen Stats smart (UUID-style) identifier for the visiting team.
visitorTeam_logo	character	URL of the visiting team's logo from the nested team object.
visitorTeam_abbr	character	Abbreviated team name for the visiting team from the nested team object.
visitorTeam_cityState	character	City and state of the visiting team as provided in the nested team object.
visitorTeam_fullName	character	Full official name of the visiting team from the nested team object.
visitorTeam_nick	character	Nickname of the visiting team from the nested team object.
visitorTeam_teamType	character	Classification of the visiting team type (e.g., 'NFL' for active franchises).
visitorTeam_conferenceAbbr	character	Conference abbreviation for the visiting team from the nested team object (e.g., 'NFC').
visitorTeam_divisionAbbr	character	Division abbreviation for the visiting team from the nested team object (e.g., 'NFC West').
score_time	character	Game clock time at the current state as reported by the live score sub-object.
score_phase	character	Current phase or status of the game as reported by the live score sub-object (e.g., 'FINAL', 'IN_PROGRESS').
score_visitorTeamScore_pointTotal	integer	Visitor team total points scored through the current game state, from the live score sub-object.
score_visitorTeamScore_pointQ1	integer	Visitor team points scored in the first quarter, from the live score sub-object.
score_visitorTeamScore_pointQ2	integer	Visitor team points scored in the second quarter, from the live score sub-object.
score_visitorTeamScore_pointQ3	integer	Visitor team points scored in the third quarter, from the live score sub-object.
score_visitorTeamScore_pointQ4	integer	Visitor team points scored in the fourth quarter, from the live score sub-object.
score_visitorTeamScore_pointOT	integer	Visitor team points scored in overtime, from the live score sub-object.
score_visitorTeamScore_timeoutsRemaining	integer	Number of timeouts remaining for the visitor team, from the live score sub-object.
score_homeTeamScore_pointTotal	integer	Home team total points scored through the current game state, from the live score sub-object.
score_homeTeamScore_pointQ1	integer	Home team points scored in the first quarter, from the live score sub-object.
score_homeTeamScore_pointQ2	integer	Home team points scored in the second quarter, from the live score sub-object.
score_homeTeamScore_pointQ3	integer	Home team points scored in the third quarter, from the live score sub-object.
score_homeTeamScore_pointQ4	integer	Home team points scored in the fourth quarter, from the live score sub-object.
score_homeTeamScore_pointOT	integer	Home team points scored in overtime, from the live score sub-object.
score_homeTeamScore_timeoutsRemaining	integer	Number of timeouts remaining for the home team, from the live score sub-object.

Example

from sportsdataverse.nfl import nfl_ngs_league_schedule
sched = nfl_ngs_league_schedule(season=2024, season_type="REG", week=1)
first_game_id = sched["gameId"][0]

nfl_ngs_league_schedule_current(return_as_pandas: 'bool' = False)

NGS schedule for the current week -- one row per game.

Wraps /api/league/schedule/current; the games are under the games key (alongside scalar season/seasonType/week describing the slice).

Parameters

Parameter	Type	Default	Description
return_as_pandas	bool	False	return a pandas frame instead of polars.

Returns

A polars (or pandas) DataFrame, one row per game in the current week.

col_name	type	description
gameKey	integer	Alternate numeric key for the NFL game used in official NFL NGS record-keeping.
gameDate	character	Game date-time (ISO 8601, UTC).
gameId	integer	Unique identifier for the NFL game in the NGS/Shield data system.
gameTime	character	Scheduled kickoff time for the game in local or ET representation.
gameTimeEastern	character	Scheduled kickoff time for the game in Eastern Time (ET), as published by the NFL.
gameType	character	Game type identifier (3 for playoffs).
homeDisplayName	character	Full display name of the home team (e.g., 'Kansas City Chiefs') for the scheduled game.
homeNickname	character	Nickname of the home team (e.g., 'Chiefs') for the scheduled game.
homeTeamAbbr	character	Two- or three-letter abbreviation identifying the home team at the top-level game record.
homeTeamId	character	Unique numeric identifier for the home team at the top-level game record.
isoTime	integer	ISO 8601 formatted datetime string representing the scheduled kickoff time of the game.
networkChannel	character	Broadcast network or channel airing the game (e.g., 'NBC', 'ESPN', 'FOX').
ngsGame	logical	Boolean or indicator flag marking whether this game entry is an official NGS-tracked game.
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
seasonType	character	Season type classification for the game (e.g., 'REG' for regular season, 'POST' for playoffs, 'PRE' for preseason).
smartId	character	Smart ID (NGS/Shield internal identifier) for the game record itself.
visitorDisplayName	character	Full display name of the visiting team (e.g., 'Philadelphia Eagles') for the scheduled game.
visitorNickname	character	Nickname of the visiting team (e.g., 'Eagles') for the scheduled game.
visitorTeamAbbr	character	Two- or three-letter abbreviation identifying the visiting team at the top-level game record.
visitorTeamId	character	Unique numeric identifier for the visiting team at the top-level game record.
week	integer	Season week.
weekNameAbbr	character	Abbreviated name or label for the NFL week (e.g., 'WK1', 'WLD' for Wild Card) in which the game is scheduled.
releasedToClubs	logical	Boolean flag indicating whether the schedule entry has been officially released to NFL clubs.
validated	logical	Boolean flag indicating whether the schedule entry has been validated by NFL operations.
homeTeam_teamId	character	Unique numeric identifier for the home team in the NFL NGS/Shield data system.
homeTeam_smartId	character	Smart ID (NGS/Shield internal identifier) for the home team.
homeTeam_logo	character	URL to the home team's official logo image as provided in the NGS schedule feed.
homeTeam_abbr	character	Two- or three-letter abbreviation for the home team (e.g., 'KC').
homeTeam_cityState	character	City and state string for the home team's primary market (e.g., 'Kansas City, MO').
homeTeam_fullName	character	Full official name of the home team (e.g., 'Kansas City Chiefs').
homeTeam_nick	character	Short nickname of the home team (e.g., 'Chiefs') as used in the NGS system.
homeTeam_teamType	character	Classification of the home team type (e.g., 'NFL' for a standard league franchise).
homeTeam_conferenceAbbr	character	Conference abbreviation for the home team (e.g., 'AFC' or 'NFC').
homeTeam_divisionAbbr	character	Division abbreviation for the home team (e.g., 'AFC West').
site_smartId	character	Smart ID (NGS/Shield internal identifier) for the game venue.
site_siteId	integer	Unique identifier for the venue or stadium in the NFL NGS/Shield data system.
site_siteFullName	character	Full official name of the stadium or venue where the game is scheduled (e.g., 'Arrowhead Stadium').
site_siteCity	character	City in which the game's venue (stadium) is located.
site_siteState	character	State (or country) in which the game's venue is located.
site_postalCode	character	Postal (ZIP) code of the stadium or venue where the game is scheduled to be played.
site_roofType	character	Roof type of the game venue (e.g., 'OPEN', 'DOME', 'RETRACTABLE') affecting playing conditions.
visitorTeam_teamId	character	Unique numeric identifier for the visiting team in the NFL NGS/Shield data system.
visitorTeam_smartId	character	Smart ID (NGS/Shield internal identifier) for the visiting team.
visitorTeam_logo	character	URL to the visiting team's official logo image as provided in the NGS schedule feed.
visitorTeam_abbr	character	Two- or three-letter abbreviation for the visiting team (e.g., 'PHI').
visitorTeam_cityState	character	City and state string for the visiting team's primary market (e.g., 'Philadelphia, PA').
visitorTeam_fullName	character	Full official name of the visiting team (e.g., 'Philadelphia Eagles').
visitorTeam_nick	character	Short nickname of the visiting team (e.g., 'Eagles') as used in the NGS system.
visitorTeam_teamType	character	Classification of the visiting team type (e.g., 'NFL' for a standard league franchise).
visitorTeam_conferenceAbbr	character	Conference abbreviation for the visiting team (e.g., 'AFC' or 'NFC').
visitorTeam_divisionAbbr	character	Division abbreviation for the visiting team (e.g., 'NFC East').
score_time	character	Game clock time or elapsed time associated with the current scoring state snapshot.
score_phase	character	Current phase or status of the game (e.g., 'FINAL', 'IN_PROGRESS', 'PREGAME').
score_visitorTeamScore_pointTotal	integer	Visitor team's total final score (sum of all quarters and overtime) for the game.
score_visitorTeamScore_pointQ1	integer	Visitor team's points scored in the first quarter of the game.
score_visitorTeamScore_pointQ2	integer	Visitor team's points scored in the second quarter of the game.
score_visitorTeamScore_pointQ3	integer	Visitor team's points scored in the third quarter of the game.
score_visitorTeamScore_pointQ4	integer	Visitor team's points scored in the fourth quarter of the game.
score_visitorTeamScore_pointOT	integer	Visitor team's points scored in overtime during the game, if applicable.
score_visitorTeamScore_timeoutsRemaining	integer	Number of timeouts remaining for the visitor team at the current game state.
score_homeTeamScore_pointTotal	integer	Home team's total final score (sum of all quarters and overtime) for the game.
score_homeTeamScore_pointQ1	integer	Home team's points scored in the first quarter of the game.
score_homeTeamScore_pointQ2	integer	Home team's points scored in the second quarter of the game.
score_homeTeamScore_pointQ3	integer	Home team's points scored in the third quarter of the game.
score_homeTeamScore_pointQ4	integer	Home team's points scored in the fourth quarter of the game.
score_homeTeamScore_pointOT	integer	Home team's points scored in overtime during the game, if applicable.
score_homeTeamScore_timeoutsRemaining	integer	Number of timeouts remaining for the home team at the current game state.

Example

from sportsdataverse.nfl import nfl_ngs_league_schedule_current
cur = nfl_ngs_league_schedule_current()
cur.select(["gameId", "homeTeamAbbr", "visitorTeamAbbr"]).head()

nfl_ngs_league_teams(return_as_pandas: 'bool' = False)

NGS team directory -- one row per team.

Wraps /api/league/teams (top-level list). Each row carries teamId, abbr, fullName, nick, conference/division, cityState, stadiumName, smartId, logo and site/ticket URLs.

Parameters

Parameter	Type	Default	Description
return_as_pandas	bool	False	return a pandas frame instead of polars.

Returns

A polars (or pandas) DataFrame, one row per team.

col_name	type	description
abbr	character	Official team abbreviation used by the NFL Next Gen Stats system (e.g., 'KC', 'NE').
cityState	character	City and state where the team is based (e.g., 'Kansas City, MO').
conferenceAbbr	character	Abbreviation of the conference the team belongs to (e.g., 'AFC', 'NFC').
fullName	character	Full name of the probable starting pitcher.
logo	character	Team or league logo URL.
nick	character	Team nickname or mascot name (e.g., 'Chiefs', 'Patriots').
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
smartId	character	NFL Next Gen Stats smart (UUID-style) identifier for the team.
stadiumName	character	Name of the team's home stadium (e.g., 'Arrowhead Stadium').
teamId	character	NFL Next Gen Stats integer identifier for the team.
teamSiteTicketUrl	character	URL for purchasing tickets via the team's official website.
teamSiteUrl	character	URL of the team's official website.
teamType	character	Classification of the team type within the NFL system (e.g., 'NFL' for active franchises).
ticketPhoneNumber	character	Phone number for ticket sales inquiries for this team.
yearFound	integer	Year the franchise was founded.
conference_id	character	Conference identifier.
conference_abbr	character	Conference abbreviation.
conference_fullName	character	Full name of the conference the team belongs to (e.g., 'American Football Conference').
division_id	character	Division MLBAM ID.
division_abbr	character	Division abbreviation.
division_fullName	character	Full name of the division the team belongs to (e.g., 'AFC West').
divisionAbbr	character	Abbreviation of the division the team belongs to (e.g., 'AFC West').

Example

from sportsdataverse.nfl import nfl_ngs_league_teams
teams = nfl_ngs_league_teams()
teams.select(["teamId", "abbr", "fullName", "conferenceAbbr"]).head()

nfl_ngs_microsite_chart(season: 'int' = 2024, season_type: 'str' = 'REG', week=None, chart_type=None, team_id=None, limit: 'int' = 100, offset: 'int' = 0, return_as_pandas: 'bool' = False)

NGS microsite chart catalogue -- one row per rendered player chart image.

Wraps /api/content/microsite/chart; records live under charts and each carries the chart imageName/type (qb-grid, pass, route, carry) plus the player and headline stats (passerRating, completions, etc.) and image-size URLs. Supports server-side paging.

Parameters

Parameter	Type	Default	Description
season	int	2024	season year.
season_type	str	'REG'	"REG", "POST", or "PRE".
week		None	optional week filter (the API accepts "all" by default).
chart_type		None	optional chart-type filter (e.g. "qb-grid", "pass").
team_id		None	optional team-id filter.
limit	int	100	page size (passed as limit).
offset	int	0	page offset.
return_as_pandas	bool	False	return a pandas frame instead of polars.

Returns

A polars (or pandas) DataFrame, one row per chart in the page.

col_name	type	description
imageName	character	Filename or label for the player image asset used on the NGS microsite chart.
esbId	character	Elias Sports Bureau identifier for the player featured on the NGS microsite chart.
firstName	character	Scorer first name (localized list).
gameId	integer	Unique identifier for the NFL game associated with the NGS microsite chart entry.
headshot	character	NFL headshot url for player
lastName	character	Scorer last name (localized list).
playerName	character	Full display name of the player featured on the NGS microsite chart.
position	character	Primary position as reported by NFL.com
receivingYards	integer	Total receiving yards for the player (receiver) featured on the NGS microsite chart.
receptions	integer	The number of pass receptions. Lateral receptions officially don't count as reception.
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
seasonType	character	Season type classification for the chart entry (e.g., 'REG' for regular season, 'POST' for playoffs).
teamId	character	Unique numeric identifier for the player's team in the NFL NGS/Shield data system.
timestamp	integer	Response timestamp (ISO 8601).
touchdowns	integer	Total number of touchdowns scored or thrown by the player featured on the NGS microsite chart.
type	character	Record type / category.
week	integer	Season week.
extraLargeImg	character	URL to the extra-large player image asset used on the NFL NGS microsite chart display.
playerNameSlug	character	URL-safe slug version of the player's name used in NGS microsite routing (e.g., 'patrick-mahomes').
smallImg	character	URL to the small player image asset used on the NFL NGS microsite chart display.
mediumImg	character	URL to the medium-sized player image asset used on the NFL NGS microsite chart display.
largeImg	character	URL to the large player image asset used on the NFL NGS microsite chart display.
carries	integer	The number of official rush attempts (incl. scrambles and kneel downs). Rushes after a lateral reception don't count as carry.
rushingYards	integer	Total rushing yards for the player featured on the NGS microsite chart.
attempts	integer	The number of pass attempts as defined by the NFL.
completionPercentage	double	Completion percentage for the quarterback featured on the NGS microsite chart.
completions	integer	The number of completed passes.
interceptions	integer	The number of interceptions thrown.
passerRating	double	NFL passer rating for the quarterback featured on the NGS microsite chart.
passingYards	integer	Total passing yards for the player (quarterback) featured on the NGS microsite chart.

Example

from sportsdataverse.nfl import nfl_ngs_microsite_chart
charts = nfl_ngs_microsite_chart(season=2024, season_type="REG", limit=25)
charts.select(["playerName", "type", "imageName"]).head()

nfl_ngs_microsite_chart_players(season: 'int' = 2024, season_type: 'str' = 'REG', return_as_pandas: 'bool' = False)

NGS microsite chart player index -- one row per player with a chart.

Wraps /api/content/microsite/chart/players; records live under players and carry esbId, firstName, lastName and playerName. Useful as the lookup list of who has charts available for a given season.

Parameters

Parameter	Type	Default	Description
season	int	2024	season year.
season_type	str	'REG'	"REG", "POST", or "PRE".
return_as_pandas	bool	False	return a pandas frame instead of polars.

Returns

A polars (or pandas) DataFrame, one row per player.

col_name	type	description
esbId	character	Elias Sports Bureau identifier for the player used in NFL Next Gen Stats microsite chart data.
firstName	character	Scorer first name (localized list).
lastName	character	Scorer last name (localized list).
playerName	character	Full display name of the player as shown in NFL Next Gen Stats microsite charts.

Example

from sportsdataverse.nfl import nfl_ngs_microsite_chart_players
who = nfl_ngs_microsite_chart_players(season=2024, season_type="REG")
who.select(["playerName", "esbId"]).head()

nfl_ngs_play_is_highlight(game_id, play_id, return_as_pandas: 'bool' = False)

Look up whether a single play is an NGS highlight -- one-row frame.

Wraps /api/plays/isHighlight (keyed by NGS gameId + playId). When the play is a highlight, the response's nested highlight block (the play metadata, the players involved, season/week/team) is flattened onto the row alongside the top-level gameId/playId/isHighlight flag. Pull a real (gameId, playId) pair from nfl_ngs_leaders -- each leader entry's play_gameId / play_playId is a known highlight.

Parameters

Parameter	Type	Default	Description
game_id			NGS gameId (e.g. "2024111800").
play_id			the play id within that game (e.g. 1214).
return_as_pandas	bool	False	return a pandas frame instead of polars.

Returns

A one-row polars (or pandas) DataFrame with gameId, playId, isHighlight and (when true) flattened highlight_* columns.

col_name	type	description
gameId	integer	NFL Shield API game identifier for the game this highlight record belongs to.
playId	integer	Unique play event identifier (UUID).
isHighlight	logical	Boolean flag indicating whether this play record has been designated as a highlight by the NFL Shield API.
highlight_gameId	integer	NFL Shield API game identifier associated with the specific highlight clip.
highlight_playId	integer	NFL Shield API play identifier for the specific play associated with the highlight clip.
highlight_play_playType	character	Text label for the type of play (e.g., 'PASS', 'RUSH', 'PUNT') as classified by the NFL Shield API.
highlight_play_gameId	integer	NFL Shield API game identifier embedded in the play detail record for the highlight.
highlight_play_gameKey	integer	NFL Shield internal game key integer used to identify the game in internal API routing.
highlight_play_yardlineSide	character	Team abbreviation indicating which team's side of the field the yardline is on.
highlight_play_absoluteYardlineNumber	integer	Absolute yard line number (1–100 from one end zone) of the line of scrimmage at the snap for the highlighted play.
highlight_play_yardlineNumber	integer	Numeric yard line (1–50 from nearest end zone) of the line of scrimmage at the snap.
highlight_play_timeOfDayUTC	character	Wall-clock timestamp in UTC at which the highlighted play occurred during the broadcast.
highlight_play_isPenalty	logical	Boolean flag indicating whether a penalty was assessed on the highlighted play.
highlight_play_homeScore	integer	Home team's score at the end of the highlighted play.
highlight_play_visitorScore	integer	Visiting team's score at the end of the highlighted play.
highlight_play_playStats	integer	Serialized or encoded statistical detail associated with the play outcomes (e.g., yards, player ids).
highlight_play_playId	integer	NFL Shield API unique integer identifier for the specific play within the game.
highlight_play_playDescription	character	Official NFL text description of the highlighted play (e.g., '(12:34) T.Brady pass short right to J.Edelman for 8 yards').
highlight_play_playTypeCode	integer	Integer code corresponding to the play type classification in the NFL Shield API.
highlight_play_quarter	integer	Quarter number (1–4, or 5 for overtime) during which the highlighted play occurred.
highlight_play_down	integer	Down number (1–4) at the time of the highlighted play.
highlight_play_yardsToGo	integer	Integer yards needed for a first down on the highlighted play.
highlight_play_actualYardsToGo	double	Exact decimal yards required for a first down on the highlighted play.
highlight_play_actualYardlineForFirstDown	double	Actual yard line marker needed for a first down on the highlighted play.
highlight_play_possessionTeamId	character	NFL Shield team identifier for the team that had possession during the highlighted play.
highlight_play_isGoalToGo	logical	Boolean flag indicating whether the highlighted play occurred in a goal-to-go situation.
highlight_play_health	character	Data quality or tracking health status string indicating the reliability of player/ball tracking data for this play.
highlight_play_endGameClock	character	Game clock time remaining at the end of the highlighted play in MM:SS format.
highlight_play_startGameClock	character	Game clock time remaining at the start of the highlighted play in MM:SS format.
highlight_play_playState	character	State or status of the play record (e.g., 'FINAL', 'LIVE') in the NFL Shield API at time of capture.
highlight_play_preSnapHomeScore	integer	Home team's score immediately before the snap of the highlighted play.
highlight_play_preSnapVisitorScore	integer	Visitor team's score immediately before the snap of the highlighted play.
highlight_play_sequence	integer	Sequential order integer of this play within the game, as used by the NFL Shield API.
highlight_play_gameClock	character	Game clock time remaining at the start of the highlighted play in MM:SS format.
highlight_play_yardline	character	Human-readable yardline string (e.g., 'KC 35') indicating where the play began.
highlight_play_isScoring	logical	Boolean flag indicating whether the highlighted play resulted in a score.
highlight_play_isEndQuarter	logical	Boolean flag indicating whether the highlighted play was the final play of a quarter.
highlight_play_isSTPlay	logical	Boolean flag indicating whether the highlighted play was a special teams play.
highlight_play_playDirection	character	Direction of the play (left, right, middle) as recorded in the NFL Shield tracking data.
highlight_play_isBigPlay	logical	Boolean flag indicating whether the play was classified as a 'big play' (e.g., long gain or turnover) by the NFL Shield API.
highlight_play_isChangeOfPossession	logical	Boolean flag indicating whether the highlighted play resulted in a change of possession.
highlight_players	integer	Serialized list or count of player tracking records associated with the highlight clip.
highlight_season	integer	NFL season year (e.g., 2024) in which the highlighted play occurred.
highlight_seasonType	character	Season phase of the highlighted play (e.g., 'REG' for regular season, 'POST' for playoffs).
highlight_teamAbbr	character	Team abbreviation of the team featured in or associated with the highlight clip.
highlight_teamId	character	NFL Shield team identifier for the team featured in the highlight clip.
highlight_week	integer	Week number within the season during which the highlighted play occurred.

Example

from sportsdataverse.nfl import nfl_ngs_leaders, nfl_ngs_play_is_highlight
lead = nfl_ngs_leaders(category="speed", season=2024, season_type="REG")
gid, pid = lead["play_gameId"][0], lead["play_playId"][0]
hl = nfl_ngs_play_is_highlight(game_id=gid, play_id=pid)
hl.select(["gameId", "playId", "isHighlight"]).head()

nfl_ngs_statboard(stat_type: 'str' = 'passing', season: 'int' = 2024, season_type: 'str' = 'REG', week: 'Optional[int]' = None, return_as_pandas: 'bool' = False)

NGS season/week statboard leaderboard for a stat family (one row per player).

Wraps /api/statboard/{passing,receiving,rushing}. Each record is a flat per-player stat line (e.g. for passing: completionPercentageAboveExpectation, avgTimeToThrow, aggressiveness, passerRating ...). The player's bio is nested under a player object and is flattened to player_* columns.

Parameters

Parameter	Type	Default	Description
stat_type	str	'passing'	one of "passing", "receiving", "rushing". (For the cross-stat highlight board use nfl_ngs_statboard_leaders.)
season	int	2024	season year, e.g. 2024.
season_type	str	'REG'	"REG", "POST", or "PRE".
week	int | None	None	single week to filter to; None returns the full-season board.
return_as_pandas	bool	False	return a pandas frame instead of polars.

Returns

A polars (or pandas) DataFrame, one row per qualifying player.

col_name	type	description
aggressiveness	double	Aggressiveness tracks the amount of passing attempts a quarterback makes that are into tight coverage, where there is a defender within 1 yard or less of the receiver at the time of completion or incompletion. AGG is shown as a % of attempts into tight windows over all passing attempts.
attempts	integer	The number of pass attempts as defined by the NFL.
avgAirDistance	double	Average air distance in yards on all pass attempts, measuring how far the ball travels through the air regardless of direction.
avgAirYardsDifferential	double	Average difference between intended air yards and completed air yards, measuring accuracy relative to target depth.
avgAirYardsToSticks	double	Average air yards relative to the first-down marker on pass attempts, where positive values indicate throws beyond the sticks.
avgCompletedAirYards	double	Average air yards on completed passes only, measuring depth of actual completions.
avgIntendedAirYards	double	Average depth of target on all pass attempts, regardless of completion.
avgTimeToThrow	double	Average time in seconds from snap to release for the passer, as tracked by NFL Next Gen Stats.
completionPercentage	double	Actual completion percentage for the passer during the period covered.
completionPercentageAboveExpectation	double	Completion percentage above the model-expected completion rate (CPAE), measuring accuracy relative to difficulty of throws attempted.
completions	integer	The number of completed passes.
expectedCompletionPercentage	double	Model-expected completion percentage based on factors such as target depth, separation, and defensive coverage as calculated by NFL Next Gen Stats.
gamesPlayed	integer	Number of games played by the passer during the period covered.
interceptions	integer	The number of interceptions thrown.
maxAirDistance	double	Maximum air distance in yards recorded on any single pass attempt during the period covered.
maxCompletedAirDistance	double	Maximum air distance in yards recorded on any single completed pass during the period covered.
passTouchdowns	integer	Total passing touchdowns thrown by the passer during the period covered.
passYards	integer	Total passing yards accumulated by the passer during the period covered.
passerRating	double	NFL passer rating (0–158.3 scale) for the passer during the period covered.
playerName	character	Display name of the passer as returned at the top-level statboard row.
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
seasonType	character	Season segment covered by this statboard record (e.g., 'REG' for regular season, 'POST' for playoffs).
position	character	Primary position as reported by NFL.com
teamId	character	NFL Next Gen Stats team identifier for the passer's team at the time of this record.
player_season	integer	NFL season year associated with this player's statboard record.
player_currentTeamId	character	NFL Next Gen Stats team identifier for the team the player is currently rostered on.
player_displayName	character	Full display name of the player as used in NFL Next Gen Stats records.
player_esbId	character	ESPN-Elias Sports Bureau (ESB) identifier for the player.
player_firstName	character	First name of the player.
player_footballName	character	Football name used by the player, which may differ from the legal first name.
player_gsisId	character	NFL GSIS (Game Statistics and Information System) identifier, the primary nflverse player key.
player_gsisItId	integer	NFL GSIS internal tracking integer identifier for the player.
player_headshot	character	URL to the player headshot image.
player_jerseyNumber	integer	Jersey number worn by the player.
player_lastName	character	Last name of the player.
player_position	character	Position of the player accordinng to NGS
player_positionGroup	character	Broad position group for the player (e.g., 'QB', 'WR') in the NGS player record.
player_shortName	character	Shortened display name for the player (e.g., 'P.Mahomes').
player_smartId	character	NFL Next Gen Stats smart (UUID-style) identifier for the player.
player_status	character	Current roster status of the player (e.g., 'ACT' for active, 'IR' for injured reserve).
player_uniformNumber	character	Uniform number worn by the player, stored as a string to preserve leading zeros if applicable.
player_ngsPosition	character	Player's position as classified by the NFL Next Gen Stats system.
player_ngsPositionGroup	character	Broader position group the player belongs to as classified by the NFL Next Gen Stats system.

Example

from sportsdataverse.nfl import nfl_ngs_statboard
qb = nfl_ngs_statboard(stat_type="passing", season=2024, season_type="REG")
qb.select(["playerName", "passerRating", "completionPercentageAboveExpectation"]).head()

nfl_ngs_statboard_leaders(season: 'int' = 2024, season_type: 'str' = 'REG', week: 'Optional[int]' = None, return_as_pandas: 'bool' = False)

NGS cross-stat "leaders" board, stacked long with a category column.

Wraps /api/statboard/leaders, which bundles several short top-N lists of mixed shape (fastestBallCarriers, fastestSacks, longestCompletions, highestSeparation, rushYardsOverExpected, completionPctAboveExpected, avgYACAboveExpected). Each list is normalized separately and concatenated diagonally (union of columns; missing cells become null), with a category column recording which board each row came from.

Parameters

Parameter	Type	Default	Description
season	int	2024	season year.
season_type	str	'REG'	"REG", "POST", or "PRE".
week	int | None	None	optional single-week filter.
return_as_pandas	bool	False	return a pandas frame instead of polars.

Returns

A polars (or pandas) DataFrame stacking every leader list, with a category column. Empty frame if no lists are present.

col_name	type	description
category	character	Broader category of player positions
leader_esbId	character	ESB (NFL's Enterprise Subscriber Base) identifier for the player featured in the leader record.
leader_firstName	character	First name of the player featured in the leader record.
leader_gsisId	character	GSIS (Game Statistics and Information System) identifier for the player featured in the leader record.
leader_jerseyNumber	integer	Jersey number of the player featured in the leader record.
leader_lastName	character	Last name of the player featured in the leader record.
leader_playerName	character	Full display name of the player featured in the leader record.
leader_position	character	Position designation of the player featured in the leader record (e.g., 'QB', 'WR').
leader_positionGroup	character	Broad position group of the player featured in the leader record (e.g., 'OFFENSE', 'DEFENSE').
leader_shortName	character	Abbreviated display name of the player featured in the leader record (e.g., 'P.Mahomes').
leader_teamAbbr	character	Team abbreviation of the player featured in the leader record (e.g., 'KC').
leader_teamId	character	NFL Shield team identifier for the team of the player featured in the leader record.
leader_week	integer	Week number associated with the weekly leader record.
leader_yards	integer	Yardage total associated with the featured leader statistic for this player.
leader_inPlayDist	double	Distance (in yards) the player traveled while the ball was in play on the featured leader statistic.
leader_maxSpeed	double	Maximum speed (in yards per second or mph) recorded by the player on the featured leader statistic play.
leader_headshot	character	URL to the headshot image of the player featured in the leader record.
play_gameId	integer	NFL Shield API game identifier for the game this play belongs to.
play_playId	integer	NFL Shield API unique integer identifier for this play within the game.
play_sequence	integer	Sequential order integer of this play within the game as used by the NFL Shield API.
play_down	integer	Down number (1–4) at the time of this play.
play_gameClock	character	Game clock time remaining at the start of this play in MM:SS format.
play_gameKey	integer	NFL Shield internal game key integer identifying the game in internal API routing.
play_health_playerTracking	character	Data quality status string for player-tracking data on this play (e.g., 'GOOD', 'PARTIAL').
play_health_ballTracking	character	Data quality status string for ball-tracking data on this play (e.g., 'GOOD', 'MISSING').
play_homeScore	integer	Home team's score at the end of this play.
play_isBigPlay	logical	Boolean flag indicating whether this play was classified as a 'big play' by the NFL Shield API.
play_isEndQuarter	logical	Boolean flag indicating whether this play was the final play of a quarter.
play_isGoalToGo	logical	Boolean flag indicating whether this play occurred in a goal-to-go situation.
play_isPenalty	logical	Boolean flag indicating whether a penalty was assessed on this play.
play_isSTPlay	logical	Boolean flag indicating whether this play was a special teams play.
play_isScoring	logical	Boolean flag indicating whether this play resulted in a score.
play_playDescription	character	Official NFL text description of the play (e.g., '(2:11) J.Allen scrambles right end for 9 yards').
play_playState	character	Status of the play record (e.g., 'FINAL', 'LIVE') in the NFL Shield API at time of capture.
play_playStats	integer	Serialized or encoded statistical detail for the play outcomes (e.g., yards, player IDs).
play_playType	character	Text label classifying the play type (e.g., 'PASS', 'RUSH', 'PUNT') from the NFL Shield API.
play_playTypeCode	integer	Integer code corresponding to the play type classification in the NFL Shield API.
play_possessionTeamId	character	NFL Shield team identifier for the team that had possession during this play.
play_preSnapHomeScore	integer	Home team's score immediately before the snap of this play.
play_preSnapVisitorScore	integer	Visiting team's score immediately before the snap of this play.
play_quarter	integer	Quarter number (1–4, or 5 for overtime) in which this play occurred.
play_timeOfDayUTC	character	Wall-clock timestamp in UTC at which this play occurred during the broadcast.
play_visitorScore	integer	Visiting team's score at the end of this play.
play_yardline	character	Human-readable yardline string (e.g., 'DAL 22') indicating where this play began.
play_yardlineNumber	integer	Numeric yard line (1–50 from nearest end zone) of the line of scrimmage at the snap.
play_yardlineSide	character	Team abbreviation indicating which team's side of the field the yardline is on.
play_yardsToGo	integer	Integer yards needed for a first down on this play.
play_absoluteYardlineNumber	integer	Absolute yard line number (1–100 from one end zone) of the line of scrimmage at the snap for this play.
play_actualYardlineForFirstDown	double	Actual yard line marker needed for a first down on this play.
play_actualYardsToGo	double	Exact decimal yards required for a first down on this play.
play_endGameClock	character	Game clock time remaining at the end of this play in MM:SS format.
play_isChangeOfPossession	logical	Boolean flag indicating whether this play resulted in a change of possession.
play_playDirection	character	Direction of the play (left, right, middle) as recorded in NFL Shield tracking data.
play_startGameClock	character	Game clock time remaining at the moment this play began, in MM:SS format.
leader_time	double	Elapsed time value associated with the leader record, such as time of possession or time-to-throw.
leader_seasonAvg	double	Season-to-date average of the featured leader statistic for this player.
leader_teamAvg	double	Team-level average of the featured statistic across all players on the team.
aggressiveness	double	Aggressiveness tracks the amount of passing attempts a quarterback makes that are into tight coverage, where there is a defender within 1 yard or less of the receiver at the time of completion or incompletion. AGG is shown as a % of attempts into tight windows over all passing attempts.
attempts	integer	The number of pass attempts as defined by the NFL.
avgAirDistance	double	Average air distance (in yards) the ball traveled on all pass attempts, from NFL Next Gen Stats.
avgAirYardsDifferential	double	Average difference between intended air yards and completed air yards per attempt, indicating whether the passer throws short or long relative to the targeted depth.
avgAirYardsToSticks	double	Average air yards relative to the first-down marker on pass attempts, from NFL Next Gen Stats (negative = short of sticks, positive = past sticks).
avgCompletedAirYards	double	Average air yards on completed passes only, from NFL Next Gen Stats.
avgIntendedAirYards	double	Average intended air yards per pass attempt (including incompletions), from NFL Next Gen Stats.
avgTimeToThrow	double	Average time in seconds from snap to throw for the passer, from NFL Next Gen Stats.
completionPercentage	double	Percentage of pass attempts completed by the passer, from NFL Next Gen Stats statboard leaders.
completionPercentageAboveExpectation	double	Passer's actual completion percentage minus their expected completion percentage based on target depth, coverage, and game context (CPOE), from NFL Next Gen Stats.
completions	integer	The number of completed passes.
expectedCompletionPercentage	double	Model-predicted completion percentage for the passer based on depth of target, receiver separation, and coverage, from NFL Next Gen Stats.
gamesPlayed	integer	Number of games played by the player in the given season or time period.
interceptions	integer	The number of interceptions thrown.
maxAirDistance	double	Maximum air distance (in yards) recorded on a single pass attempt by the passer, from NFL Next Gen Stats.
maxCompletedAirDistance	double	Maximum air distance on a completed pass for the passer, from NFL Next Gen Stats.
passTouchdowns	integer	Total passing touchdowns recorded by the player in the given period.
passYards	integer	Total passing yards recorded by the player in the given period.
passerRating	double	NFL passer rating (0–158.3 scale) for the quarterback in the given period.
playerName	character	Full display name of the player associated with this NGS statboard leader record.
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
seasonType	character	Phase of the season for this record (e.g., 'REG' for regular season, 'POST' for playoffs).
position	character	Primary position as reported by NFL.com
teamId	character	NFL Shield team identifier for the team associated with this leader statistic.
player_season	integer	NFL season year in which this player record is valid.
player_currentTeamId	character	NFL Shield team identifier for the player's current team.
player_displayName	character	Player's full display name as used by NFL Next Gen Stats (e.g., 'Patrick Mahomes').
player_esbId	character	ESB (Enterprise Subscriber Base) identifier assigned to this player by the NFL.
player_firstName	character	Player's first name as recorded in the NFL Next Gen Stats player roster.
player_footballName	character	Player's preferred football name (may differ from legal first name) used on the field and in broadcasts.
player_gsisId	character	GSIS (Game Statistics and Information System) identifier, the primary NFL player identifier used across nflverse datasets.
player_gsisItId	integer	GSIS integrated tracking identifier, an alternate integer-form player ID used in the NFL Shield tracking system.
player_jerseyNumber	integer	Player's jersey number as worn on the field.
player_lastName	character	Player's last name as recorded in the NFL Next Gen Stats player roster.
player_position	character	Position of the player accordinng to NGS
player_positionGroup	character	Official NFL position group for the player (e.g., 'OFFENSE', 'DEFENSE', 'SPECIAL_TEAMS').
player_shortName	character	Abbreviated player name used in display contexts (e.g., 'P.Mahomes').
player_status	character	Player's current roster status (e.g., 'ACT' for active, 'IR' for injured reserve).
player_uniformNumber	character	Player's uniform number as a string, matching what appears on the jersey.
player_headshot	character	URL to the player headshot image.
player_smartId	character	NFL Smart ID — a system-agnostic unique identifier for the player used across NFL Shield systems.
player_ngsPosition	character	Player's position as classified by NFL Next Gen Stats (may differ from official NFL position; e.g., NGS uses 'ILB' vs 'LB').
player_ngsPositionGroup	character	Broad position group assigned by NFL Next Gen Stats (e.g., 'QB', 'WR', 'DB', 'DL').
avgCushion	double	Average distance (in yards) between the receiver and the nearest defender at the snap, from NFL Next Gen Stats receiver tracking.
avgExpectedYAC	double	Average yards after catch expected based on game situation and receiver location, from NFL Next Gen Stats models.
avgSeparation	double	Average separation (in yards) between the receiver and the nearest defender at the moment of catch or incompletion, from NFL Next Gen Stats.
avgYAC	double	Average yards after catch per reception, from NFL Next Gen Stats receiver tracking.
avgYACAboveExpectation	double	Average yards after catch above statistical expectation (YAC - expected YAC), from NFL Next Gen Stats models.
catchPercentage	double	Percentage of targets caught by the receiver, from NFL Next Gen Stats statboard leaders.
percentShareOfIntendedAirYards	double	Receiver's share (as a percentage) of the team's total intended air yards, from NFL Next Gen Stats.
recTouchdowns	integer	Total receiving touchdowns recorded by the player in the given period.
receptions	integer	The number of pass receptions. Lateral receptions officially don't count as reception.
targets	integer	The number of pass plays where the player was the targeted receiver.
yards	integer	The number of receiving yards
avgTimeToLos	double	Average time (in seconds) for the ball carrier to reach the line of scrimmage on rush attempts, from NFL Next Gen Stats.
expectedRushYards	double	Model-predicted rushing yards based on field position, personnel, and pre-snap alignment, from NFL Next Gen Stats.
rushAttempts	integer	Total rushing attempts by the player in the given period.
rushPctOverExpected	double	Percentage of rush attempts on which the player gained more yards than the model-predicted expectation, from NFL Next Gen Stats.
rushTouchdowns	integer	Total rushing touchdowns recorded by the player in the given period.
rushYards	integer	Total rushing yards recorded by the player in the given period.
rushYardsOverExpected	double	Total rushing yards gained above statistical expectation (RYOE) in the given period, from NFL Next Gen Stats.
rushYardsOverExpectedPerAtt	double	Average rushing yards over expected per attempt (RYOE/attempt), from NFL Next Gen Stats.

Example

from sportsdataverse.nfl import nfl_ngs_statboard_leaders
bd = nfl_ngs_statboard_leaders(season=2024, season_type="REG")
bd["category"].unique().to_list()

nfl_players_crosswalk(*, return_as_pandas: 'bool' = False) -> "Union[pl.DataFrame, 'pd.DataFrame']"

Pure-consumer ID crosswalk sliced from load_nfl_players.

Reads nflverse's published players master and projects it down to just the cross-system identifier columns it carries (gsis_id, esb_id, espn_id, pfr_id, pff_id, otc_id, nfl_id, smart_id — whichever the parquet exposes) plus full_name and position, deduped on gsis_id. It is a convenience for joining nflverse identity IDs onto PBP / rosters / stats frames without carrying the full ~40-column master.

Parameters

Parameter	Type	Default	Description
return_as_pandas	bool	False	If True, return a pandas.DataFrame; otherwise a polars.DataFrame (default).

Returns

A one-row-per-gsis_id DataFrame of cross-system IDs + full_name / position. A failed / empty players load yields a zero-row frame carrying the same column set (never a raise).

Example

from sportsdataverse.nfl import nfl_players_crosswalk
xwalk = nfl_players_crosswalk()
print(xwalk.columns)

# Join nflverse IDs onto a PBP frame (one line)

pbp.join(nfl_players_crosswalk(), left_on="passer_player_id", right_on="gsis_id", how="left")

nfl_token_gen(client_key: 'Optional[str]' = None, client_secret: 'Optional[str]' = None, force_refresh: 'bool' = False) -> 'str'

Return a valid api.nfl.com bearer token, minting + caching as needed.

The token is cached in-process and reused until ~2 min before its own JWT exp, then transparently re-minted -- so callers never have to think about expiry or refresh. The first call (or any call after expiry / force_refresh) mints a fresh token via the anonymous device-token grant at /identity/v3/token.

Resolution order (all overrides optional):

1. NFL_ACCESS_TOKEN env var -- returned verbatim, skipping minting and caching (you supply + manage the token). Ignored if explicit credentials are passed.
2. Credentials: explicit client_key/client_secret args -> NFL_CLIENT_KEY/NFL_CLIENT_SECRET env vars -> the bundled public WEB_DESKTOP web-app pair.

Parameters

Parameter	Type	Default	Description
client_key	Optional[str]	None	Override the client key (else env var, else the web default).
client_secret	Optional[str]	None	Override the client secret (else env var, else the default).
force_refresh	bool	False	Mint a new token even if a cached one is still valid.

Returns

The bearer accessToken string.

Example

from sportsdataverse.nfl.nfl_games import nfl_token_gen
token = nfl_token_gen()                # mints + caches
assert nfl_token_gen() == token        # served from cache
assert isinstance(token, str) and token.startswith("ey")

nfl_week_games(season: 'int' = 2024, season_type: 'str' = 'REG', week: 'int' = 1, headers: 'Optional[Dict[str, str]]' = None, return_as_pandas: 'bool' = False)

Parsed api.nfl.com week schedule -- one row per game (polars/pandas frame).

Tidy wrapper over nfl_game_schedule: flattens the games list into a DataFrame with id (uuid game id), season/seasonType/week, date, status_*, and homeTeam_* / awayTeam_* columns.

Parameters

Parameter	Type	Default	Description
season	int	2024	season year. season_type (str): "PRE"/"REG"/"POST".
season_type	str	'REG'
week	int	1	week number. headers: reuse a nfl_headers_gen dict.
headers	Optional[Dict[str, str]]	None
return_as_pandas	bool	False	return a pandas frame instead of polars.

Returns

A polars (or pandas) DataFrame, one row per game.

col_name	type	description
id	character	ID of the player in the 'name' column.
category	character	Broader category of player positions
date	character	Date in YYYY-MM-DD format.
time	character	Time at start of play provided in string format as minutes:seconds remaining in the quarter.
gameType	character	Game type identifier (3 for playoffs).
international	logical	Boolean flag indicating whether this game is designated as an international (outside the United States) game.
neutralSite	logical	Whether the game is at a neutral site.
season	integer	4 digit number indicating to which season(s) the specified timeframe belongs to.
seasonType	character	Phase of the season in which this game takes place (e.g., 'REG', 'POST', 'PRE').
status	character	Status label.
week	integer	Season week.
weekType	character	Classification of the week type within the season (e.g., 'REG', 'WC', 'DIV', 'CONF', 'SB').
externalIds	character	Serialized list of external system identifiers (e.g., partner IDs, league IDs) mapped to this game.
ticketUrl	character	URL to the official ticket purchasing page for this game.
ticketVendors	character	Serialized list of authorized ticket vendors or marketplaces for this game.
extensions	character	Serialized JSON or string of additional extension metadata attached to the game record by the NFL Shield API.
version	integer	API response version integer returned by the NFL Shield API for this game record.
homeTeam_id	character	NFL Shield team identifier for the home team.
homeTeam_currentLogo	character	URL for the home team's current primary logo image as served by the NFL Shield API.
homeTeam_fullName	character	Full name of the home team (e.g., 'Dallas Cowboys').
awayTeam_id	character	NFL Shield team identifier for the away team.
awayTeam_currentLogo	character	URL for the away team's current primary logo image as served by the NFL Shield API.
awayTeam_fullName	character	Full name of the away team (e.g., 'Kansas City Chiefs').
broadcastInfo_homeNetworkChannels	character	Serialized list of TV/streaming channels designated as the home team's local broadcast.
broadcastInfo_awayNetworkChannels	character	Serialized list of TV/streaming channels designated as the away team's local broadcast.
broadcastInfo_internationalWatchOptions	character	Serialized list of international streaming or broadcast options for viewers outside the United States.
broadcastInfo_streamingNetworks	character	Serialized list of streaming platforms (e.g., Peacock, Amazon Prime Video) carrying the game.
broadcastInfo_territory	character	Geographic territory or market designation for which this broadcast record applies.
broadcastInfo_audioNetworks	character	Serialized list of radio networks carrying the game's audio broadcast.
venue_id	character	Unique venue identifier.
venue_name	character	Venue name.
venue_city	character	Venue city.
venue_country	character	Country name or ISO code indicating where the game's venue is located.

Example

from sportsdataverse.nfl import nfl_week_games
sched = nfl_week_games(season=2024, season_type="REG", week=1)
sched.select(["id", "homeTeam_fullName", "awayTeam_fullName"]).head()

reset_config() -> 'NflConfig'

Reset the active config to its env-var-derived defaults.

Convenience for tests / interactive sessions that want to undo a chain of update_config() calls without restarting the interpreter.

Example

from sportsdataverse.nfl import update_config, reset_config
update_config(cache_mode="off", timeout=5)
# ... do work ...
reset_config()  # back to env-derived defaults

scoreboard_event_parsing(event)

Normalize one ESPN scoreboard event into a flatter shape.

Splits the competitors list into home / away siblings, hoists notes / broadcast metadata onto the competition root, and drops the fields the schedule helper does not need (odds, leaders, geoBroadcasts, etc.). Used internally by espn_nfl_schedule.

Parameters

Parameter	Type	Default	Description
event	Dict		A single events[i] dict from the ESPN scoreboard endpoint.

Returns

The mutated event dict with normalized home / away / broadcast keys.

Example

from sportsdataverse.dl_utils import download
from sportsdataverse.nfl.nfl_schedule import scoreboard_event_parsing
url = "http://site.api.espn.com/apis/site/v2/sports/football/nfl/scoreboard"
payload = download(url=url).json()
for ev in payload.get("events", []):
    scoreboard_event_parsing(ev)
    ev["competitions"][0]["home"]["abbreviation"]

scrape_ngs_season(stat_type: 'str', season: 'int', *, include_season_totals: 'bool' = True, return_as_pandas: 'bool' = False)

Scrape a full season of NGS statboard data, shaped like the nflverse parquet.

Port of nflverse ngs-data's R save_ngs_type: loop the regular-season weeks (1..max_reg where max_reg = 18 for season >= 2021 else 17) plus the playoff weeks (max_reg+1 .. max_reg+5, fetched with season_type="POST"), stack them diagonally, and -- when include_season_totals -- prepend the season-aggregate rows (NGS week=0, a REG call with no week param) tagged week=0. Duplicate rows (NGS returned dupes for some 2022 weeks) are de-duplicated on (season, week, player_gsis_id).

Output columns match the published nflverse NGS parquet read by sportsdataverse.nfl.load_nfl_nextgen_stats (snake_case, team_abbr resolved). It will not be byte-identical -- nflverse post-processes (column pruning / ordering) -- but the core metric columns and the player/team/week keys align.

NGS statboard rows are player-week aggregates, NOT per-play rows.

Parameters

Parameter	Type	Default	Description
stat_type	str		one of "passing", "rushing", "receiving".
season	int		season year (NGS coverage starts in 2016).
include_season_totals	bool	True	also fetch the season-aggregate (week=0) rows. Defaults to True (matches ngs-data, whose week loop starts at 0).
return_as_pandas	bool	False	return a pandas frame instead of polars.

Returns

A polars (or pandas) DataFrame stacking every week (and, by default, the season totals) for the requested stat_type and season. An EMPTY frame carrying the documented key schema if nothing is returned.

col_name	type	description
aggressiveness	double	Percentage of pass attempts thrown into tight windows (defender within one yard of the receiver at completion or incompletion), from NFL Next Gen Stats. Passing only.
attempts	integer	Pass attempts (including incompletions) recorded for the passer during the slice. Passing only.
avg_air_distance	double	Average air distance in yards on all pass attempts, measuring how far the ball travels through the air regardless of direction. Passing only.
avg_air_yards_differential	double	Average difference between intended air yards and completed air yards, measuring accuracy relative to target depth. Passing only.
avg_air_yards_to_sticks	double	Average air yards relative to the first-down marker on pass attempts (positive = beyond the sticks). Passing only.
avg_completed_air_yards	double	Average air yards on completed passes only, measuring depth of actual completions. Passing only.
avg_intended_air_yards	double	Average depth of target on all pass attempts, regardless of completion. Passing/receiving.
avg_time_to_throw	double	Average time in seconds from snap to release for the passer, as tracked by NFL Next Gen Stats. Passing only.
completion_percentage	double	Actual completion percentage for the passer during the slice. Passing only.
completion_percentage_above_expectation	double	Completion percentage above the model-expected completion rate (CPOE/CPAE). Passing only.
completions	integer	Completed passes for the passer during the slice. Passing only.
expected_completion_percentage	double	Model-expected completion percentage based on target depth, separation, and coverage, from NFL Next Gen Stats. Passing only.
games_played	integer	Number of games played by the player during the slice covered by the row.
interceptions	integer	Interceptions thrown by the passer during the slice. Passing only.
max_air_distance	double	Maximum air distance in yards recorded on any single pass attempt during the slice. Passing only.
max_completed_air_distance	double	Maximum air distance in yards recorded on any single completed pass during the slice. Passing only.
pass_touchdowns	integer	Passing touchdowns thrown by the passer during the slice. Passing only.
pass_yards	integer	Total passing yards accumulated by the passer during the slice. Passing only.
passer_rating	double	NFL passer rating (0–158.3 scale) for the passer during the slice. Passing only.
player_name	character	Display name of the player as returned at the top-level statboard row.
season	integer	NFL season year for the row.
season_type	character	Season segment for the row ('REG' for regular-season weeks and the week-0 aggregate, 'POST' for playoff weeks).
week	integer	NFL Next Gen Stats week tag; 0 is the season aggregate, 1..max_reg are regular-season weeks, and higher values are continuous playoff weeks.
position	character	Player position as returned at the top-level statboard row (e.g., 'QB', 'WR', 'RB').
team_id	character	NFL Next Gen Stats team identifier for the player's team on this row; the key joined to resolve team_abbr.
player_season	integer	NFL season year recorded in the nested player record.
player_season_type	character	Season segment recorded in the nested player record.
player_week	integer	Week recorded in the nested player record (the loop week tag overrides this on the top-level week column).
player_jersey_number	integer	Jersey number worn by the player.
player_last_name	character	Last name of the player.
player_football_name	character	Football name used by the player, which may differ from the legal first name.
player_first_name	character	First name of the player.
player_position	character	Player position from the nested player record.
player_gsis_it_id	integer	NFL GSIS internal tracking integer identifier for the player.
player_gsis_id	character	NFL GSIS (Game Statistics and Information System) identifier, the primary nflverse player key.
player_esb_id	character	Elias Sports Bureau (ESB) identifier for the player.
player_display_name	character	Full display name of the player as used in NFL Next Gen Stats records.
player_short_name	character	Shortened display name for the player (e.g., 'P.Mahomes').
player_uniform_number	character	Uniform number worn by the player, stored as a string to preserve leading zeros if applicable.
player_status	character	Current roster status of the player (e.g., 'ACT' for active, 'IR' for injured reserve).
player_current_team_id	character	NFL Next Gen Stats team identifier for the team the player is currently rostered on.
player_smart_id	character	NFL Next Gen Stats smart (UUID-style) identifier for the player.
player_headshot	character	URL template for the player's headshot image.
player_position_group	character	Broad position group for the player (e.g., 'QB', 'WR') in the NGS player record.
player_ngs_position	character	Player's position as classified by the NFL Next Gen Stats system.
player_ngs_position_group	character	Broader position group the player belongs to as classified by the NFL Next Gen Stats system.
team_abbr	character	Team abbreviation resolved from team_id via the NGS team directory (relocated franchise abbreviations dropped to keep the mapping one-to-one).

Example

from sportsdataverse.nfl import scrape_ngs_season
pas = scrape_ngs_season("passing", 2023)
pas.select(["season", "week", "player_display_name", "team_abbr"]).head()

# Regular-season weeks only (skip the week-0 totals)

wk = scrape_ngs_season("receiving", 2023, include_season_totals=False)

# Column-compatible with the published parquet

from sportsdataverse.nfl import load_nfl_nextgen_stats
published = load_nfl_nextgen_stats(seasons=[2023], stat_type="passing")
shared = set(pas.columns) & set(published.columns)

scrape_ngs_week(stat_type: 'str', season: 'int', week: 'int', season_type: 'str' = 'REG', *, return_as_pandas: 'bool' = False)

Scrape one (season, week) NGS statboard slice, shaped like the nflverse parquet.

Port of nflverse ngs-data's R load_week_ngs: fetch a single statboard slice via nfl_ngs_statboard, resolve team_abbr from the team directory, snake-case every column, and tag the row with the loop week. week=0 is the season-aggregate row (a season_type="REG" call with no week query param); weeks 1..max_reg are regular-season, and the playoff weeks (max_reg+1 upward) are fetched with season_type="POST".

NGS statboard rows are player-week aggregates (avg_intended_air_yards, completion_percentage_above_expectation, avg_time_to_throw, ...), NOT per-play rows -- this is a season-stats source, not a per-play air-yards / completion-probability source.

Parameters

Parameter	Type	Default	Description
stat_type	str		one of "passing", "rushing", "receiving".
season	int		season year (NGS coverage starts in 2016).
week	int		NGS week. 0 -> season aggregate; 1..max_reg -> REG; higher -> POST. The supplied value is what tags the returned rows.
season_type	str	'REG'	"REG" or "POST"; the caller (or scrape_ngs_season) selects this per week. Defaults to "REG".
return_as_pandas	bool	False	return a pandas frame instead of polars.

Returns

A polars (or pandas) DataFrame of player-week NGS rows with snake-cased columns + a resolved team_abbr. An EMPTY frame carrying the documented key schema (not an exception) when the API yields no stats.

col_name	type	description
aggressiveness	double	Percentage of pass attempts thrown into tight windows (defender within one yard of the receiver at completion or incompletion), from NFL Next Gen Stats. Passing only.
attempts	integer	Pass attempts (including incompletions) recorded for the passer during the week. Passing only.
avg_air_distance	double	Average air distance in yards on all pass attempts, measuring how far the ball travels through the air regardless of direction. Passing only.
avg_air_yards_differential	double	Average difference between intended air yards and completed air yards, measuring accuracy relative to target depth. Passing only.
avg_air_yards_to_sticks	double	Average air yards relative to the first-down marker on pass attempts (positive = beyond the sticks). Passing only.
avg_completed_air_yards	double	Average air yards on completed passes only, measuring depth of actual completions. Passing only.
avg_intended_air_yards	double	Average depth of target on all pass attempts, regardless of completion. Passing/receiving.
avg_time_to_throw	double	Average time in seconds from snap to release for the passer, as tracked by NFL Next Gen Stats. Passing only.
completion_percentage	double	Actual completion percentage for the passer during the week. Passing only.
completion_percentage_above_expectation	double	Completion percentage above the model-expected completion rate (CPOE/CPAE). Passing only.
completions	integer	Completed passes for the passer during the week. Passing only.
expected_completion_percentage	double	Model-expected completion percentage based on target depth, separation, and coverage, from NFL Next Gen Stats. Passing only.
games_played	integer	Number of games played by the player during the slice covered by the row.
interceptions	integer	Interceptions thrown by the passer during the week. Passing only.
max_air_distance	double	Maximum air distance in yards recorded on any single pass attempt during the week. Passing only.
max_completed_air_distance	double	Maximum air distance in yards recorded on any single completed pass during the week. Passing only.
pass_touchdowns	integer	Passing touchdowns thrown by the passer during the week. Passing only.
pass_yards	integer	Total passing yards accumulated by the passer during the week. Passing only.
passer_rating	double	NFL passer rating (0–158.3 scale) for the passer during the week. Passing only.
player_name	character	Display name of the player as returned at the top-level statboard row.
season	integer	NFL season year for the row.
season_type	character	Season segment for the row ('REG' or 'POST') as supplied by the caller.
week	integer	NFL Next Gen Stats week tag supplied by the caller; 0 is the season aggregate, 1..max_reg are regular-season weeks, and higher values are continuous playoff weeks.
position	character	Player position as returned at the top-level statboard row (e.g., 'QB', 'WR', 'RB').
team_id	character	NFL Next Gen Stats team identifier for the player's team on this row; the key joined to resolve team_abbr.
player_season	integer	NFL season year recorded in the nested player record.
player_season_type	character	Season segment recorded in the nested player record.
player_week	integer	Week recorded in the nested player record (the loop week tag overrides this on the top-level week column).
player_jersey_number	integer	Jersey number worn by the player.
player_last_name	character	Last name of the player.
player_football_name	character	Football name used by the player, which may differ from the legal first name.
player_first_name	character	First name of the player.
player_position	character	Player position from the nested player record.
player_gsis_it_id	integer	NFL GSIS internal tracking integer identifier for the player.
player_gsis_id	character	NFL GSIS (Game Statistics and Information System) identifier, the primary nflverse player key.
player_esb_id	character	Elias Sports Bureau (ESB) identifier for the player.
player_display_name	character	Full display name of the player as used in NFL Next Gen Stats records.
player_short_name	character	Shortened display name for the player (e.g., 'P.Mahomes').
player_uniform_number	character	Uniform number worn by the player, stored as a string to preserve leading zeros if applicable.
player_status	character	Current roster status of the player (e.g., 'ACT' for active, 'IR' for injured reserve).
player_current_team_id	character	NFL Next Gen Stats team identifier for the team the player is currently rostered on.
player_smart_id	character	NFL Next Gen Stats smart (UUID-style) identifier for the player.
player_headshot	character	URL template for the player's headshot image.
player_position_group	character	Broad position group for the player (e.g., 'QB', 'WR') in the NGS player record.
player_ngs_position	character	Player's position as classified by the NFL Next Gen Stats system.
player_ngs_position_group	character	Broader position group the player belongs to as classified by the NFL Next Gen Stats system.
team_abbr	character	Team abbreviation resolved from team_id via the NGS team directory (relocated franchise abbreviations dropped to keep the mapping one-to-one).

Example

from sportsdataverse.nfl import scrape_ngs_week
wk1 = scrape_ngs_week("passing", 2023, week=1)
wk1.select(["season", "week", "player_display_name", "team_abbr"]).head()

# Season-aggregate row (week 0)

tot = scrape_ngs_week("rushing", 2023, week=0)

update_config(**kwargs: 'object') -> 'NflConfig'

Update the active config in place.

Returns

The (mutated) global config object, for chaining or inspection.

Example

from sportsdataverse.nfl import update_config
update_config(cache_mode="filesystem", cache_duration=3600)

# Disable caching for development

update_config(cache_mode="off")

# Point cache at a custom directory

update_config(cache_dir="~/sdv-cache")