Skip to main content
Breaking in 4.0.0
  • Integer id columns are decimal strings — Every id column of integers (id, *_id, game_pk, playerId, …) comes back as exact decimal strings on every surface — parsers, producers, loaders, HockeyTech analytics. Compare and join ids as strings; Number(row.game_id) for a safe number back. (changelog)
  • Typed wrapper params and loader rows; strict mode — TypeScript only. A param the endpoint does not have, a missing required path param or a non-boolean bool param is a type error; parser rows are Record<string, unknown>; loader rows are their generated row types. (changelog)

sdv.nba — dataset loaders

45 loaders reading the published sportsdataverse-data releases (parquet) — the JS mirror of sportsdataverse-py's load_* functions. Each is a camelCase export plus its snake_case alias and resolves to an array of plain row objects (or { [column]: values[] } with format: "columns").

  • Size: row objects cost ~60-100 bytes per cell on the heap, so before decoding each season a loader adds its rows × columns (from the parquet footer) to a running total and checks it against maxCells — by default heap limit / 100 for rows (≈45M cells on Node's default 4 GB heap) and heap limit / 30 for format: "columns" — and throws a catchable SdvError instead of running out of memory. Play-by-play is the usual case: pass columns, use format: "columns", or raise the heap (node --max-old-space-size=8192).
  • Seasons: seasons takes one season or a list. A season with no published asset (HTTP 404) is skipped with a warning (loaders marked so in their seasons row — sdv-py's hand-written ones — throw NoDataError instead); any other failure raises AssetFetchError (a failed download is never an empty season); a season below the loader's floor raises SeasonNotFoundError before anything is fetched. Multi-season results union the columns, null-filling gaps, and cast a column whose type changed between seasons to the common type (an integer id that became a string → strings, "123" not "123.0"), as sdv-py's diagonal_relaxed concat does.
  • Integers: an id column (id, *_id, *_ids, game_pk, athlete_id_1, id_play, playerId, homeTeamId, start.team.id, …) of integers comes back as exact decimal strings in every row and every season, whatever width the release stores it with (INT32, INT64, or a DOUBLE holding integers: "401628579101849903", "39", never "39.0"), so ids join across seasons and across releases. Code-like id columns (type_id, status_id, …) are strings too. An id column that is not exact integers (a fraction, a DOUBLE past 2^53) is left as read with one warning. Any other INT64 column comes back as number when every value is a safe integer, otherwise as BigInt with one warning (code SDV_INT64) per column per process.
  • Runtime: Node only. Downloads go through the releases transport family (see Transport, auth & errors); each asset is downloaded whole, then decoded.
LoaderReleaseSeasons
loadNbaPbpespn_nba_pbp2002+
loadNbaPlayerBoxscoreespn_nba_player_boxscores1994+
loadNbaScheduleespn_nba_schedules1994+
loadNbaTeamBoxscoreespn_nba_team_boxscores1994+
loadNbaGameRostersespn_nba_game_rosters2002+
loadNbaOfficialsespn_nba_officials2002+
loadNbaShotsespn_nba_shots2002+
loadNbaStandingsespn_nba_standings2002+
loadNbaPlayerSeasonStatsespn_nba_player_season_stats1994+
loadNbaTeamSeasonStatsespn_nba_team_season_stats2002+
loadNbaDraftespn_nba_draft2003+
loadNbaRostersespn_nba_rosters2025+
loadNbaStatsSchedulesnba_stats_schedules1996+
loadNbaStatsCoachesnba_stats_coaches1996+
loadNbaStatsDraftnba_stats_draft1996+
loadNbaStatsGameRostersnba_stats_game_rosters1996+
loadNbaStatsLineupsnba_stats_lineups2007+
loadNbaStatsLineupsV3 (deprecated)nba_stats_game_lineups1996+
loadNbaStatsOfficialsnba_stats_officials1996+
loadNbaStatsPbpnba_stats_pbp1996+
loadNbaStatsPossessionsnba_stats_possessions1996+
loadNbaStatsGameLineupsnba_stats_game_lineups1996+
loadNbaStatsGameMatchupsnba_stats_game_matchups2017+
loadNbaStatsInjuryReportnba_stats_injury_report2018+
loadNbaStatsPbpV3 (deprecated)nba_stats_pbp1996+
loadNbaStatsPlayerBoxscoresnba_stats_player_boxscores1996+
loadNbaStatsPlayerGameLogsnba_stats_player_game_logs1996+
loadNbaStatsPlayerSeasonStatsnba_stats_player_season_stats1996+
loadNbaStatsPossessionsV3 (deprecated)nba_stats_possessions1996+
loadNbaStatsRostersnba_stats_rosters1996+
loadNbaStatsShotsnba_stats_shots1996+
loadNbaStatsStandingsnba_stats_standings1996+
loadNbaStatsTeamBoxscoresnba_stats_team_boxscores1996+
loadNbaStatsTeamSeasonStatsnba_stats_team_season_stats1996+
loadNbaPlayerCrosswalknba_crosswalk2002+
loadNbaScheduleCrosswalknba_crosswalk2002+
loadNbaTeamCrosswalknba_crosswalk2002+
loadNbaPlayerCoreespn_nba_player_core1994+
loadNbaPlayerImpactnba_player_impact1996+
loadNbaGroupsnba_groupsone asset
loadNbaGroupSeasonsnba_groupsone asset
loadNbaGroupAliasesnba_groupsone asset
loadNbaTeamGroupSeasonsnba_groups1971+
loadNbaTeamAliasesnba_groupsone asset
loadNbaPlayerAliasesnba_groupsone asset

loadNbaPbp​

Release: espn_nba_pbp · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/espn_nba_pbp/play_by_play_{season}.parquet

optiontyperequireddescription
seasonsnumber | number[]yesseason(s) to load (>= 2002)
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaPbp({ seasons: 2024, columns: ['game_id', 'sequence_number', 'type_text', 'text', 'score_value'] });
// snake_case alias (py/R parity): sdv.nba.load_nba_pbp(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaPbpRow (exported from the package root).

col_nametypedescription
game_play_numbernumberSequential play number within the game.
idstringId.
sequence_numbernumberSequence number representing a shot-possession (V3 PBP).
type_idstringType identifier (numeric).
type_textstringDisplay text for the type field.
textstringText description of the play / record.
away_scorenumberAway team's score after the play. Rebuilt from the scoring plays in feed order when that reaches the official final (score_source 'rebuilt'), otherwise ESPN's per-play value (score_source 'espn'). Seasons published before the repair carry ESPN's per-play value.
home_scorenumberHome team's score after the play. Rebuilt from the scoring plays in feed order when that reaches the official final (score_source 'rebuilt'), otherwise ESPN's per-play value (score_source 'espn'). Seasons published before the repair carry ESPN's per-play value.
period_numbernumberNumeric period (1-4 for quarters; 5+ for OT).
period_display_valuestringPeriod display label (e.g. '1st Quarter', 'OT').
clock_display_valuestringGame clock display string (e.g. '8:32').
scoring_playbooleanTRUE if the play resulted in points scored.
score_valuenumberPoint value of the play (2 / 3 / 1).
team_idstringUnique team identifier.
athlete_id_1stringPrimary athlete identifier (e.g. shooter).
athlete_id_2stringSecondary athlete identifier (e.g. assister / fouler).
athlete_id_3stringAthlete id 3.
wallclockstringWallclock.
shooting_playbooleanTRUE if the play was a shooting attempt.
coordinate_x_rawnumberX coordinate as returned by the API before any adjustment.
coordinate_y_rawnumberY coordinate as returned by the API before any adjustment.
points_attemptednumberPoint value attempted on the shot (2 or 3; 1 for a free throw).
short_descriptionstringShort text description of the play from ESPN.
game_idstringUnique game identifier.
seasonnumberSeason year.
season_typenumberSeason type (1=pre-season, 2=regular season, 3=postseason, 4=off-season for ESPN; or string label for WNBA Stats).
home_team_idstringUnique identifier for the home team.
home_team_namestringHome team name.
home_team_mascotstringHome team mascot.
home_team_abbrevstringHome team three-letter abbreviation.
home_team_name_altstringAlternate home team name.
away_team_idstringUnique identifier for the away team.
away_team_namestringAway team name.
away_team_mascotstringAway team mascot.
away_team_abbrevstringAway team three-letter abbreviation.
away_team_name_altstringAlternate away team name.
game_spreadnumberGame spread (signed; positive = home favored).
home_favoritebooleanTRUE if the home team is the betting favorite.
game_spread_availablebooleanTRUE if a point spread was available.
home_team_spreadnumberHome team's point spread.
qtrnumberQuarter (1-4) or OT period (5+).
timestringTime / clock value.
clock_minutesnumberClock minutes split out for convenience.
clock_secondsnumberClock seconds split out for convenience.
home_timeout_calledbooleanWhether the play is a timeout called by the home team.
away_timeout_calledbooleanWhether the play is a timeout called by the away team.
halfnumberHalf of the game (1 or 2).
game_halfnumberHalf of the game (1 or 2).
lag_qtrnumberQuarter lag (the previous-play's quarter).
lead_qtrnumberQuarter lead (the next-play's quarter).
lag_halfnumberA lag column on the half
lead_halfnumberA lead column on the half
start_quarter_seconds_remainingnumberSeconds remaining in the period at the start of the play.
start_half_seconds_remainingnumberSeconds remaining in the half at the start of the play.
start_game_seconds_remainingnumberSeconds remaining in the game at the start of the play.
end_quarter_seconds_remainingnumberSeconds remaining in the period at the end of the play.
end_half_seconds_remainingnumberSeconds remaining in the half at the end of the play.
end_game_seconds_remainingnumberSeconds remaining in the game at the end of the play.
periodnumberPeriod of the game (1-4 quarters; 5+ for OT).
coordinate_xnumberX coordinate on the court (half-court layout).
coordinate_ynumberY coordinate on the court (half-court layout).
game_dateDateGame date (YYYY-MM-DD).
game_date_timeDateGame start date/time (ISO 8601).
athlete_name_1stringName of the primary athlete involved in the play (pairs with athlete_id_1).
athlete_name_2stringName of the secondary athlete involved in the play (e.g. the assister or fouled player).
athlete_name_3stringName of the tertiary athlete involved in the play.
type_abbreviationstringType abbreviation.
score_sourcestringWhere this game's home_score/away_score come from: 'rebuilt' (summed from the scoring plays and equal to the official final on the last play) or 'espn' (ESPN's per-play values, kept when the rebuild misses the final or the game has no final). Absent from seasons published before the repair. Select it by name: it is the play-by-play builder's last column, but a release can carry columns after it.

loadNbaPlayerBoxscore​

Release: espn_nba_player_boxscores · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/espn_nba_player_boxscores/player_box_{season}.parquet

optiontyperequireddescription
seasonsnumber | number[]yesseason(s) to load (>= 1994)
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaPlayerBoxscore({ seasons: 2024 });
// snake_case alias (py/R parity): sdv.nba.load_nba_player_boxscore(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaPlayerBoxscoreRow (exported from the package root).

col_nametypedescription
game_idstringUnique game identifier.
seasonnumberSeason year.
season_typenumberSeason type (1=pre-season, 2=regular season, 3=postseason, 4=off-season for ESPN; or string label for WNBA Stats).
game_dateDateGame date (YYYY-MM-DD).
game_date_timeDateGame start date/time (ISO 8601).
athlete_idstringUnique athlete identifier (ESPN).
athlete_display_namestringAthlete display name (full).
team_idstringUnique team identifier.
team_namestringFull team display name (e.g. 'Las Vegas Aces').
team_locationstringTeam city or location string.
team_short_display_namestringShort team display name (e.g. 'Aces').
minutesnumberMinutes played, formatted MM:SS (V3 PT-duration parsed) or decimal minutes (V2).
field_goals_madenumberField goals made (2-pt + 3-pt).
field_goals_attemptednumberField goal attempts (2-pt + 3-pt).
three_point_field_goals_madenumberThree-point field goals made.
three_point_field_goals_attemptednumberThree-point field goal attempts.
free_throws_madenumberFree throws made.
free_throws_attemptednumberFree throw attempts.
offensive_reboundsnumberOffensive rebounds.
defensive_reboundsnumberDefensive rebounds.
reboundsnumberTotal rebounds.
assistsnumberTotal assists.
stealsnumberTotal steals.
blocksnumberTotal blocks.
turnoversnumberTotal turnovers.
foulsnumberPersonal fouls.
plus_minusstringPlus/minus point differential while on court.
pointsnumberPoints scored.
starterbooleanTRUE if the player was in the starting lineup; FALSE otherwise.
ejectedbooleanTRUE if the player was ejected from the game.
did_not_playbooleanTRUE if the player did not appear in the game.
reasonstringReason.
activebooleanTRUE if the row represents an active record (player / team / season).
athlete_jerseystringAthlete jersey number.
athlete_short_namestringAthlete short display name.
athlete_headshot_hrefstringAthlete headshot image URL.
athlete_position_namestringAthlete position ('Guard', 'Forward', 'Center').
athlete_position_abbreviationstringAthlete position abbreviation (G / F / C).
team_display_namestringFull team display name.
team_uidstringESPN universal team identifier (UID format 's:40~l:...~t:...').
team_slugstringURL-safe team identifier (e.g. 'lasvegas-aces' / 'aces').
team_logostringTeam logo image URL.
team_abbreviationstringShort team abbreviation (e.g. 'LAS').
team_colorstringTeam primary color (hex without leading '#').
team_alternate_colorstringTeam alternate color (hex without leading '#').
home_awaystringGame venue label ('home' or 'away').
team_winnerbooleanTRUE if the team won this game.
team_scorenumberTeam's score / final score.
opponent_team_idstringUnique identifier for the opponent team.
opponent_team_namestringOpponent team display name.
opponent_team_locationstringOpponent team city / location.
opponent_team_display_namestringOpponent team full display name.
opponent_team_abbreviationstringOpponent team abbreviation.
opponent_team_logostringOpponent team logo URL.
opponent_team_colorstringOpponent team primary color (hex).
opponent_team_alternate_colorstringOpponent team alternate color (hex).
opponent_team_scorenumberOpponent team's score.

loadNbaSchedule​

Release: espn_nba_schedules · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/espn_nba_schedules/nba_schedule_{season}.parquet

optiontyperequireddescription
seasonsnumber | number[]yesseason(s) to load (>= 1994)
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaSchedule({ seasons: 2024 });
// snake_case alias (py/R parity): sdv.nba.load_nba_schedule(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaScheduleRow (exported from the package root).

col_nametypedescription
idstringId.
uidstringESPN UID string.
datestringDate in YYYY-MM-DD format.
attendancenumberReported attendance.
time_validbooleanTime valid.
neutral_sitebooleanNeutral site.
conference_competitionbooleanConference competition.
play_by_play_availableboolean
recentbooleanRecent.
start_datestringStart date (YYYY-MM-DD).
broadcaststringBroadcast information string.
highlightsstring
notes_typestringNotes type.
notes_headlinestringNotes headline.
broadcast_marketstringBroadcast market label (e.g. 'national', 'home').
broadcast_namestringBroadcast name.
type_idstringType identifier (numeric).
type_abbreviationstringType abbreviation.
venue_idstringUnique venue identifier.
venue_full_namestringVenue full name.
venue_address_citystringVenue address city.
venue_indoorbooleanTRUE if the venue is indoors.
status_clocknumberStatus clock.
status_display_clockstringStatus display clock.
status_periodnumberStatus period.
status_type_idstringESPN's numeric game-status code, passed through unchanged. ESPN publishes no reference for these codes, so read status_type_name rather than hard-coding numbers.
status_type_namestringESPN's game-status enum (STATUS_FINAL, STATUS_SCHEDULED, STATUS_POSTPONED, STATUS_CANCELED, ...), passed through unchanged. ESPN does not document the full set, and no value certifies an ordinary final as distinct from a forfeit or an awarded result.
status_type_statestringESPN's coarse game state, passed through unchanged: pre, in or post. post covers every game that is over, whatever the reason.
status_type_completedbooleanESPN's completed flag, passed through unchanged: true once ESPN considers the game over. It does not separate an ordinary final from a forfeited, protested, replayed or awarded result; ESPN documents no such distinction.
status_type_descriptionstringESPN's display text for the game status (for example "Final" or "Postponed"), passed through unchanged.
status_type_detailstringESPN's longer status text, passed through unchanged; an overtime final reads like "Final/OT".
status_type_short_detailstringESPN's short status text for scoreboards, passed through unchanged.
format_regulation_periodsnumberFormat regulation periods.
home_idstringUnique identifier for home.
home_uidstringHome team's uid.
home_locationstringHome team's location.
home_namestringHome name.
home_abbreviationstringHome team's abbreviation.
home_display_namestringHome display name.
home_short_display_namestringHome short display name.
home_colorstringColor code (hex) for home.
home_alternate_colorstringColor code (hex) for home alternate.
home_is_activebooleanHome team's is active.
home_venue_idstringUnique identifier for home venue.
home_logostringHome team logo URL.
home_scorenumberHome team's points as ESPN reports them: once the game is final, the full-game total including every overtime period. ESPN corrects a score in place, so this is the value ESPN showed when the release was last built. Not a final score until status_type_completed is true.
home_winnerbooleanHome team's winner.
home_linescoresstringPeriod-by-period points for the home team, stringified from ESPN's linescores array.
home_recordsstringHome team's records at game time (overall, home, away), stringified from ESPN.
away_idstringUnique identifier for away.
away_uidstringAway team's uid.
away_locationstringAway team's location.
away_namestringAway name.
away_abbreviationstringAway team's abbreviation.
away_display_namestringAway display name.
away_short_display_namestringAway short display name.
away_colorstringColor code (hex) for away.
away_alternate_colorstringColor code (hex) for away alternate.
away_is_activebooleanAway team's is active.
away_venue_idstringUnique identifier for away venue.
away_logostringAway team logo URL.
away_scorenumberAway team's points as ESPN reports them: once the game is final, the full-game total including every overtime period. ESPN corrects a score in place, so this is the value ESPN showed when the release was last built. Not a final score until status_type_completed is true.
away_winnerbooleanAway team's winner.
away_linescoresstringPeriod-by-period points for the away team, stringified from ESPN's linescores array.
away_recordsstringAway team's records at game time (overall, home, away), stringified from ESPN.
game_idstringUnique game identifier.
seasonnumberSeason year.
season_typenumberSeason type (1=pre-season, 2=regular season, 3=postseason, 4=off-season for ESPN; or string label for WNBA Stats).
venue_address_statestringVenue address state / region.
status_type_alt_detailstringStatus type alt detail.
game_jsonboolean
game_json_urlstring
game_date_timeDateGame start date/time (ISO 8601).
game_dateDateGame date (YYYY-MM-DD).
PBPboolean
team_boxbooleanTeam box.
player_boxbooleanPlayer box.

loadNbaTeamBoxscore​

Release: espn_nba_team_boxscores · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/espn_nba_team_boxscores/team_box_{season}.parquet

optiontyperequireddescription
seasonsnumber | number[]yesseason(s) to load (>= 1994)
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaTeamBoxscore({ seasons: 2024 });
// snake_case alias (py/R parity): sdv.nba.load_nba_team_boxscore(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaTeamBoxscoreRow (exported from the package root).

col_nametypedescription
game_idstringUnique game identifier.
seasonnumberSeason year.
season_typenumberSeason type (1=pre-season, 2=regular season, 3=postseason, 4=off-season for ESPN; or string label for WNBA Stats).
game_dateDateGame date (YYYY-MM-DD).
game_date_timeDateGame start date/time (ISO 8601).
team_idstringUnique team identifier.
team_uidstringESPN universal team identifier (UID format 's:40~l:...~t:...').
team_slugstringURL-safe team identifier (e.g. 'lasvegas-aces' / 'aces').
team_locationstringTeam city or location string.
team_namestringFull team display name (e.g. 'Las Vegas Aces').
team_abbreviationstringShort team abbreviation (e.g. 'LAS').
team_display_namestringFull team display name.
team_short_display_namestringShort team display name (e.g. 'Aces').
team_colorstringTeam primary color (hex without leading '#').
team_alternate_colorstringTeam alternate color (hex without leading '#').
team_logostringTeam logo image URL.
team_home_awaystringTeam home away.
team_scorenumberTeam's score / final score.
team_winnerbooleanTRUE if the team won this game.
assistsnumberTotal assists.
blocksnumberTotal blocks.
defensive_reboundsnumberDefensive rebounds.
fast_break_pointsstringFast-break points scored.
field_goal_pctnumberField goal percentage (0-1).
field_goals_madenumberField goals made (2-pt + 3-pt).
field_goals_attemptednumberField goal attempts (2-pt + 3-pt).
flagrant_foulsnumberTotal flagrant fouls.
foulsnumberPersonal fouls.
free_throw_pctnumberFree throw percentage (0-1).
free_throws_madenumberFree throws made.
free_throws_attemptednumberFree throw attempts.
largest_leadstringLargest lead during the game.
offensive_reboundsnumberOffensive rebounds.
points_in_paintstringPoints scored in the paint.
stealsnumberTotal steals.
team_turnoversnumberTeam turnovers (turnovers credited to the team rather than a player).
technical_foulsnumberTotal technical fouls.
three_point_field_goal_pctnumberThree-point field goal percentage (0-1).
three_point_field_goals_madenumberThree-point field goals made.
three_point_field_goals_attemptednumberThree-point field goal attempts.
total_reboundsnumberTotal rebounds.
total_technical_foulsnumberTotal technical fouls (player + team).
total_turnoversnumberTotal turnovers (player + team).
turnover_pointsstringTurnover points.
turnoversnumberTotal turnovers.
opponent_team_idstringUnique identifier for the opponent team.
opponent_team_uidstringOpponent team uid.
opponent_team_slugstringOpponent team slug.
opponent_team_locationstringOpponent team city / location.
opponent_team_namestringOpponent team display name.
opponent_team_abbreviationstringOpponent team abbreviation.
opponent_team_display_namestringOpponent team full display name.
opponent_team_short_display_namestringOpponent team short display name.
opponent_team_colorstringOpponent team primary color (hex).
opponent_team_alternate_colorstringOpponent team alternate color (hex).
opponent_team_logostringOpponent team logo URL.
opponent_team_scorenumberOpponent team's score.
lead_changesstringLead changes.
lead_percentagestringPercentage of the game the team held the lead, from ESPN's team box score stats.

loadNbaGameRosters​

Release: espn_nba_game_rosters · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/espn_nba_game_rosters/game_rosters_{season}.parquet

optiontyperequireddescription
seasonsnumber | number[]yesseason(s) to load (>= 2002)
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaGameRosters({ seasons: 2002 });
// snake_case alias (py/R parity): sdv.nba.load_nba_game_rosters(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaGameRostersRow (exported from the package root).

col_nametypedescription
seasonnumberSeason year.
game_idstringUnique game identifier.
team_idstringUnique team identifier.
team_slugstringURL-safe team identifier (e.g. 'lasvegas-aces' / 'aces').
team_abbreviationstringShort team abbreviation (e.g. 'LAS').
team_display_namestringFull team display name.
home_awaystringGame venue label ('home' or 'away').
athlete_idstringUnique athlete identifier (ESPN).
athlete_uidstringESPN athlete UID (universal identifier).
athlete_guidstringESPN athlete GUID.
athlete_display_namestringAthlete display name (full).
athlete_short_namestringAthlete short display name.
athlete_first_namestring
athlete_last_namestring
athlete_jerseystringAthlete jersey number.
athlete_positionstringAthlete position.
athlete_headshotstringURL of the player's headshot image.
starterbooleanTRUE if the player was in the starting lineup; FALSE otherwise.
did_not_playbooleanTRUE if the player did not appear in the game.
activebooleanTRUE if the row represents an active record (player / team / season).
ejectedbooleanTRUE if the player was ejected from the game.
reasonstringReason.

loadNbaOfficials​

Release: espn_nba_officials · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/espn_nba_officials/officials_{season}.parquet

optiontyperequireddescription
seasonsnumber | number[]yesseason(s) to load (>= 2002)
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaOfficials({ seasons: 2002 });
// snake_case alias (py/R parity): sdv.nba.load_nba_officials(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaOfficialsRow (exported from the package root).

col_nametypedescription
seasonnumberSeason year.
game_idstringUnique game identifier.
official_full_namestringFull name of an on-court game official as ESPN publishes it in the summary gameInfo.officials array; it is identical to official_display_name in every released NBA row.
official_display_namestringESPN's display-form name for the official, falling back to the full name when ESPN omits it; it never diverges from official_full_name in the released NBA data.
official_positionstringESPN's label for the official's assignment slot; the NBA summary feed only ever ships Referee, so this reads the same on every released row.
official_position_idstringESPN's numeric identifier for the official's assignment slot, constant at 40 (Referee) across every released NBA season.
official_ordernumberThe official's listing index within the game's crew as ESPN orders them, normally 1 through 3 for a three-person crew; a fourth entry appears in a small share of recent-season games and the sequence is not guaranteed to be gap-free.

loadNbaShots​

Release: espn_nba_shots · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/espn_nba_shots/shots_{season}.parquet

optiontyperequireddescription
seasonsnumber | number[]yesseason(s) to load (>= 2002)
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaShots({ seasons: 2002 });
// snake_case alias (py/R parity): sdv.nba.load_nba_shots(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaShotsRow (exported from the package root).

col_nametypedescription
game_idstringUnique game identifier.
seasonnumberSeason year.
period_numbernumberNumeric period (1-4 for quarters; 5+ for OT).
clock_display_valuestringGame clock display string (e.g. '8:32').
team_idstringUnique team identifier.
athlete_id_1stringPrimary athlete identifier (e.g. shooter).
athlete_id_2stringSecondary athlete identifier (e.g. assister / fouler).
type_idstringType identifier (numeric).
type_textstringDisplay text for the type field.
scoring_playbooleanTRUE if the play resulted in points scored.
score_valuenumberPoint value of the play (2 / 3 / 1).
coordinate_xnumberX coordinate on the court (half-court layout).
coordinate_ynumberY coordinate on the court (half-court layout).
coordinate_x_rawnumberX coordinate as returned by the API before any adjustment.
coordinate_y_rawnumberY coordinate as returned by the API before any adjustment.
athlete_name_1stringName of the shooter on the attempt.
athlete_name_2stringName of the secondary athlete on the attempt (e.g. the assister).
team_namestringFull team display name (e.g. 'Las Vegas Aces').
team_mascotstringMascot (nickname) portion of the shooting team's name.
team_abbrevstringAbbreviation for team.

loadNbaStandings​

Release: espn_nba_standings · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/espn_nba_standings/standings_{season}.parquet

optiontyperequireddescription
seasonsnumber | number[]yesseason(s) to load (>= 2002)
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaStandings({ seasons: 2002 });
// snake_case alias (py/R parity): sdv.nba.load_nba_standings(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaStandingsRow (exported from the package root).

col_nametypedescription
seasonnumberSeason year.
group_idstringESPN group id.
group_namestring
group_abbreviationstring
group_short_namestringESPN's short name for the standings grouping the team sits in, read from the group node's shortName; the NBA standings payload supplies only the group name and abbreviation, so this is null throughout.
team_idstringUnique team identifier.
team_uidstringESPN universal team identifier (UID format 's:40~l:...~t:...').
team_slugstringURL-safe team identifier (e.g. 'lasvegas-aces' / 'aces').
team_locationstringTeam city or location string.
team_namestringFull team display name (e.g. 'Las Vegas Aces').
team_abbreviationstringShort team abbreviation (e.g. 'LAS').
team_display_namestringFull team display name.
team_short_display_namestringShort team display name (e.g. 'Aces').
team_colorstringTeam primary color (hex without leading '#').
team_alternate_colorstringTeam alternate color (hex without leading '#').
team_logostringTeam logo image URL.
stat_namestringStat key.
stat_display_namestringStat display name.
stat_short_display_namestring
stat_descriptionstringESPN's prose gloss for the standings statistic on this row; for the clincher stat it is not a fixed label but the team's actual status text, such as Clinched Playoff Berth or Eliminated From Playoff.
stat_abbreviationstringESPN's short code for the standings statistic, such as PCT, GB or OPP PPG; it is null for the four record-style splits (Home, Road, vs. Conf., vs. Div.), which ship no abbreviation.
stat_typestringStat type code (e.g. "win", "loss").
display_valuestringDisplay-formatted value.
valuenumberNumeric or string value field.

loadNbaPlayerSeasonStats​

Release: espn_nba_player_season_stats · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/espn_nba_player_season_stats/player_season_stats_{season}.parquet

optiontyperequireddescription
seasonsnumber | number[]yesseason(s) to load (>= 1994)
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaPlayerSeasonStats({ seasons: 2025 });
// snake_case alias (py/R parity): sdv.nba.load_nba_player_season_stats(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaPlayerSeasonStatsRow (exported from the package root).

col_nametypedescription
seasonnumberSeason year.
athlete_idstringUnique athlete identifier (ESPN).
athlete_display_namestringAthlete display name (full).
athlete_position_abbreviationstringAthlete position abbreviation (G / F / C).
athlete_jerseystringAthlete jersey number.
team_idstringUnique team identifier.
team_slugstringURL-safe team identifier (e.g. 'lasvegas-aces' / 'aces').
team_display_namestringFull team display name.
categorystringCategory label.
stat_labelstring
stat_namestringStat key.
stat_display_namestringStat display name.
stat_descriptionstringESPN's prose definition of the statistic on this row, for example the ratio of field goals made to field goals attempted; for the paired Made-Attempted stats it is the two definitions joined with a hyphen.
display_valuestringDisplay-formatted value.
valuenumberNumeric or string value field.

loadNbaTeamSeasonStats​

Release: espn_nba_team_season_stats · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/espn_nba_team_season_stats/team_season_stats_{season}.parquet

optiontyperequireddescription
seasonsnumber | number[]yesseason(s) to load (>= 2002)
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaTeamSeasonStats({ seasons: 2025 });
// snake_case alias (py/R parity): sdv.nba.load_nba_team_season_stats(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaTeamSeasonStatsRow (exported from the package root).

col_nametypedescription
seasonnumberSeason year.
team_idstringUnique team identifier.
team_slugstringURL-safe team identifier (e.g. 'lasvegas-aces' / 'aces').
team_abbreviationstringShort team abbreviation (e.g. 'LAS').
team_display_namestringFull team display name.
team_short_display_namestringShort team display name (e.g. 'Aces').
team_colorstringTeam primary color (hex without leading '#').
team_alternate_colorstringTeam alternate color (hex without leading '#').
team_logostringTeam logo image URL.
categorystringCategory label.
stat_labelstring
stat_namestringStat key.
stat_display_namestringStat display name.
stat_descriptionstringESPN's prose definition of the team statistic on this row, for example the average number of assists a team records per turnover.
display_valuestringDisplay-formatted value.
valuenumberNumeric or string value field.

loadNbaDraft​

Release: espn_nba_draft · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/espn_nba_draft/draft_{season}.parquet

optiontyperequireddescription
seasonsnumber | number[]yesseason(s) to load (>= 2003)
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaDraft({ seasons: 2025 });
// snake_case alias (py/R parity): sdv.nba.load_nba_draft(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaDraftRow (exported from the package root).

col_nametypedescription
seasonnumberSeason year.
roundnumberTournament / playoff round.
round_display_namestringESPN's display label for the draft round a pick belongs to, read from the round object's displayName; the NBA payload supplies no round metadata, so this is null for every released season 2003 through 2025.
picknumberPick number within the round.
overall_picknumberOverall pick.
pick_tradedstringESPN's traded flag for the pick, carried through as a string rather than a boolean; every released NBA row reads FALSE, so it does not currently identify traded picks.
pick_notesstringFree-text note ESPN can attach to a pick, read from the pick's notes or note field; the NBA draft payload never populates it, so it is null across all released seasons.
athlete_idstringUnique athlete identifier (ESPN).
athlete_uidstringESPN athlete UID (universal identifier).
athlete_guidstringESPN athlete GUID.
athlete_first_namestring
athlete_last_namestring
athlete_full_namestring
athlete_display_namestringAthlete display name (full).
athlete_short_namestringAthlete short display name.
athlete_heightstring
athlete_weightstring
athlete_position_abbreviationstringAthlete position abbreviation (G / F / C).
athlete_position_namestringAthlete position ('Guard', 'Forward', 'Center').
athlete_headshot_hrefstringAthlete headshot image URL.
college_idstringUnique identifier for college.
college_namestringCollege / pre-draft team.
college_short_namestring
college_abbreviationstringAbbreviation of the drafted player's college taken from the pick's nested college block; the ESPN NBA draft feed omits that block entirely, so this and the other college columns are null throughout.
team_idstringUnique team identifier.
team_uidstringESPN universal team identifier (UID format 's:40~l:...~t:...').
team_slugstringURL-safe team identifier (e.g. 'lasvegas-aces' / 'aces').
team_locationstringTeam city or location string.
team_namestringFull team display name (e.g. 'Las Vegas Aces').
team_abbreviationstringShort team abbreviation (e.g. 'LAS').
team_display_namestringFull team display name.
team_short_display_namestringShort team display name (e.g. 'Aces').
team_colorstringTeam primary color (hex without leading '#').
team_alternate_colorstringTeam alternate color (hex without leading '#').
team_logostringTeam logo image URL.

loadNbaRosters​

Release: espn_nba_rosters · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/espn_nba_rosters/rosters_{season}.parquet

optiontyperequireddescription
seasonsnumber | number[]yesseason(s) to load (>= 2025)
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaRosters({ seasons: 2025 });
// snake_case alias (py/R parity): sdv.nba.load_nba_rosters(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaRostersRow (exported from the package root).

col_nametypedescription
seasonnumberSeason year.
team_idstringUnique team identifier.
team_slugstringURL-safe team identifier (e.g. 'lasvegas-aces' / 'aces').
team_abbreviationstringShort team abbreviation (e.g. 'LAS').
team_display_namestringFull team display name.
team_short_display_namestringShort team display name (e.g. 'Aces').
team_colorstringTeam primary color (hex without leading '#').
team_alternate_colorstringTeam alternate color (hex without leading '#').
team_logostringTeam logo image URL.
athlete_idstringUnique athlete identifier (ESPN).
uidstringESPN UID string.
guidstringStable cross-league team GUID.
full_namestringPlayer's full name.
display_namestringDisplay name.
short_namestringShort display name.
first_namestringPlayer's first name.
last_namestringPlayer's last name.
jerseystringJersey number worn by the player.
position_abbreviationstringPosition abbreviation ('G' / 'F' / 'C').
position_namestringListed roster position ('Guard', 'Forward', 'Center').
position_idstringUnique position identifier.
heightstringPlayer height (string e.g. '6-2' or inches).
weightstringPlayer weight in pounds.
agestringPlayer age (in years).
date_of_birthstringDate of birth (YYYY-MM-DD).
birth_place_citystringBirth place city.
birth_place_statestringBirth place state.
birth_place_countrystringBirth place country.
experience_yearsstringExperience years.
experience_display_valuestringExperience display value.
headshot_hrefstringHeadshot image URL.
headshot_altstringAlternative-text label for the headshot.
link_webstringWeb link / URL.
status_idstringStatus identifier.
status_namestringStatus label.
status_typestringStatus type.

loadNbaStatsSchedules​

Release: nba_stats_schedules · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/nba_stats_schedules/nba_schedule_{season + 1}.parquet

Pass the season's START year (e.g. 1996 for the 1996-97 season); the asset is keyed by the END year and the loader translates internally, so a season column in the rows carries that END year.

optiontyperequireddescription
seasonsnumber | number[]yesseason(s) to load (>= 1996)
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaStatsSchedules({ seasons: 2025 });
// snake_case alias (py/R parity): sdv.nba.load_nba_stats_schedules(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaStatsSchedulesRow (exported from the package root).

col_nametypedescription
game_idstringUnique game identifier.
seasonnumber | bigintSeason year.
season_typestringSeason type (1=pre-season, 2=regular season, 3=postseason, 4=off-season for ESPN; or string label for WNBA Stats).
game_datestringGame date (YYYY-MM-DD).
matchupstringMatchup.
home_team_idstringUnique identifier for the home team.
home_team_abbreviationstring
home_team_namestringHome team name.
home_ptsnumber | bigintFinal points scored by the home team.
home_wlstringHome team's result for the game (W or L).
away_team_idstringUnique identifier for the away team.
away_team_abbreviationstring
away_team_namestringAway team name.
away_ptsnumber | bigintFinal points scored by the away team.
away_wlstringAway team's result for the game (W or L).

loadNbaStatsCoaches​

Release: nba_stats_coaches · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/nba_stats_coaches/coaches_{season + 1}.parquet

Pass the season's START year (e.g. 1996 for the 1996-97 season); the asset is keyed by the END year and the loader translates internally, so a season column in the rows carries that END year.

optiontyperequireddescription
seasonsnumber | number[]yesseason(s) to load (>= 1996)
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaStatsCoaches({ seasons: 2025 });
// snake_case alias (py/R parity): sdv.nba.load_nba_stats_coaches(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaStatsCoachesRow (exported from the package root).

col_nametypedescription
team_idstringUnique team identifier.
seasonnumberSeason year.
coach_idstringESPN coach id.
first_namestringPlayer's first name.
last_namestringPlayer's last name.
coach_namestringCoach's full name.
is_assistantnumber | bigintNumeric flag from the NBA Stats API distinguishing assistants from head coaches.
coach_typestringCoach role description (e.g. "Head Coach", "Assistant Coach").
sort_sequencenumber | bigintSort order of the coach within the team's staff listing.
sub_sort_sequencenumber | bigintSecondary sort order within the coach type.
season_typestringSeason type (1=pre-season, 2=regular season, 3=postseason, 4=off-season for ESPN; or string label for WNBA Stats).

loadNbaStatsDraft​

Release: nba_stats_draft · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/nba_stats_draft/draft_{season + 1}.parquet

Coverage

seasons is the draft year: load_nba_stats_draft(2025) returns the June 2025 draft, read from draft_2026.parquet, whose rows carry season 2026 (the END year of the 2025-26 season the picks enter). season is Int32 in this asset, not the Int64 of the other nba_stats loaders.

Pass the season's START year (e.g. 1996 for the 1996-97 season); the asset is keyed by the END year and the loader translates internally, so a season column in the rows carries that END year.

optiontyperequireddescription
seasonsnumber | number[]yesseason(s) to load (>= 1996)
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaStatsDraft({ seasons: 2025 });
// snake_case alias (py/R parity): sdv.nba.load_nba_stats_draft(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaStatsDraftRow (exported from the package root).

col_nametypedescription
person_idstringstats.nba.com player identifier (PERSON_ID) of the pick; Int64, joins to player_id elsewhere.
player_namestringDrafted player's name.
seasonnumberEND year of the season the draft class enters (2026 = the June 2025 draft) -- one more than the seasons argument, which is the draft year. Int32 in this asset.
round_numbernumber | bigintDraft round (1 or 2).
round_picknumber | bigintPick number within the round.
overall_picknumber | bigintOverall pick number.
draft_typestringSelection type as stats.nba.com reports it (e.g. 'Draft').
team_idstringstats.nba.com identifier of the team that made the pick; Int64.
team_citystringCity of the team that made the pick.
team_namestringNickname of the team that made the pick.
team_abbreviationstringTricode of the team that made the pick.
organizationstringSchool, club or team the player was drafted from.
organization_typestringKind of organization drafted from (e.g. 'College/University', 'Other Team/Club').
player_profile_flagnumber | bigint1 when stats.nba.com has a player profile page for the pick, else 0.

loadNbaStatsGameRosters​

Release: nba_stats_game_rosters · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/nba_stats_game_rosters/game_rosters_{season + 1}.parquet

Pass the season's START year (e.g. 1996 for the 1996-97 season); the asset is keyed by the END year and the loader translates internally, so a season column in the rows carries that END year.

optiontyperequireddescription
seasonsnumber | number[]yesseason(s) to load (>= 1996)
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaStatsGameRosters({ seasons: 2025 });
// snake_case alias (py/R parity): sdv.nba.load_nba_stats_game_rosters(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaStatsGameRostersRow (exported from the package root).

col_nametypedescription
player_idstringUnique player identifier.
first_namestringPlayer's first name.
last_namestringPlayer's last name.
jersey_numstringJersey number worn by the player.
team_idstringUnique team identifier.
team_citystringTeam city or region (e.g. 'Las Vegas').
team_namestringFull team display name (e.g. 'Las Vegas Aces').
team_abbreviationstringShort team abbreviation (e.g. 'LAS').
seasonnumberSeason year.
game_idstringUnique game identifier.
season_type_idstringSeason-type digit: the 3rd character of game_id (and the leading digit of season_id). 1 = preseason, 2 = regular season, 3 = All-Star, 4 = playoffs, 5 = play-in, 6 = NBA Cup final, 9 = international.

loadNbaStatsLineups​

Release: nba_stats_lineups · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/nba_stats_lineups/lineups_{season + 1}.parquet

Pass the season's START year (e.g. 2007 for the 2007-08 season); the asset is keyed by the END year and the loader translates internally, so a season column in the rows carries that END year.

optiontyperequireddescription
seasonsnumber | number[]yesseason(s) to load (>= 2007)
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaStatsLineups({ seasons: 2025 });
// snake_case alias (py/R parity): sdv.nba.load_nba_stats_lineups(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaStatsLineupsRow (exported from the package root).

col_nametypedescription
group_setstringLineup grouping label from the NBA Stats API (e.g. "Lineups").
group_idstringESPN group id.
group_namestring
team_idstringUnique team identifier.
team_abbreviationstringShort team abbreviation (e.g. 'LAS').
gpnumber | bigintGames played.
wnumber | bigintWins.
lnumber | bigintLosses.
w_pctnumberWins percentage (0-1 decimal).
minnumberMinutes played.
e_off_ratingnumberEstimated offensive rating (NBA Stats estimated-metrics family) over the split.
off_ratingnumberOffensive rating (points scored per 100 possessions) over the split.
e_def_ratingnumberEstimated defensive rating (NBA Stats estimated-metrics family) over the split.
def_ratingnumberDefensive rating (points allowed per 100 possessions) over the split.
e_net_ratingnumberEstimated net rating (NBA Stats estimated-metrics family) over the split.
net_ratingnumberNet rating (off rating - def rating).
ast_pctnumberAssist percentage.
ast_tonumberAssist-to-turnover ratio over the split.
ast_rationumberAssist ratio (assists per 100 possessions used) over the split.
oreb_pctnumberOffensive rebound percentage over the split, as a decimal.
dreb_pctnumberDefensive rebound percentage over the split, as a decimal.
reb_pctnumberTotal rebound percentage over the split, as a decimal.
tm_tov_pctnumberTeam turnover percentage (turnovers per 100 possessions) over the split, as a decimal.
efg_pctnumberEffective field goal percentage over the split, as a decimal.
ts_pctnumberTrue shooting percentage (0-1).
e_pacenumberEstimated pace (NBA Stats estimated-metrics family) over the split.
pacenumberPossessions per 48 minutes.
pace_per40numberPace per40.
possnumber | bigintPoss.
pienumberPlayer Impact Estimate (0-1).
gp_ranknumber | bigintLeague rank of the row's games played for the season and split.
w_ranknumber | bigintLeague rank of the row's wins for the season and split.
l_ranknumber | bigintLeague rank of the row's losses for the season and split.
w_pct_ranknumber | bigintLeague rank of the row's win percentage for the season and split.
min_ranknumber | bigintLeague rank of the row's minutes played for the season and split.
off_rating_ranknumber | bigintLeague rank of the row's offensive rating (points scored per 100 possessions) for the season and split.
def_rating_ranknumber | bigintLeague rank of the row's defensive rating (points allowed per 100 possessions) for the season and split.
net_rating_ranknumber | bigintLeague rank of the row's net rating (offensive minus defensive rating) for the season and split.
ast_pct_ranknumber | bigintLeague rank of the row's assist percentage (share of teammate field goals assisted while on the floor) for the season and split.
ast_to_ranknumber | bigintLeague rank of the row's assist-to-turnover ratio for the season and split.
ast_ratio_ranknumber | bigintLeague rank of the row's assist ratio (assists per 100 possessions used) for the season and split.
oreb_pct_ranknumber | bigintLeague rank of the row's offensive rebound percentage for the season and split.
dreb_pct_ranknumber | bigintLeague rank of the row's defensive rebound percentage for the season and split.
reb_pct_ranknumber | bigintLeague rank of the row's total rebound percentage for the season and split.
tm_tov_pct_ranknumber | bigintLeague rank of the row's team turnover percentage (turnovers per 100 possessions) for the season and split.
efg_pct_ranknumber | bigintLeague rank of the row's effective field goal percentage for the season and split.
ts_pct_ranknumber | bigintLeague rank of the row's true shooting percentage for the season and split.
pace_ranknumber | bigintLeague rank of the row's pace (possessions per 48 minutes) for the season and split.
pie_ranknumber | bigintLeague rank of the row's Player Impact Estimate (PIE, the NBA Stats catch-all impact metric) for the season and split.
sum_time_playednumber | bigintTotal time the five-man lineup was on the floor across the split.
seasonnumberSeason year.
season_typestringSeason type (1=pre-season, 2=regular season, 3=postseason, 4=off-season for ESPN; or string label for WNBA Stats).
measure_typestringNBA Stats measure type the row was pulled from (e.g. Base, Advanced, Misc, Scoring, Opponent, Usage, Defense).
per_modestringNBA Stats per-mode of the row values (e.g. Totals, PerGame, Per100Possessions).
fgmnumberField goals made.
fganumberField goal attempts.
fg_pctnumberField goal percentage (0-1).
fg3mnumberThree-point field goals made.
fg3anumberThree-point field goal attempts.
fg3_pctnumberThree-point field goal percentage (0-1).
ftmnumberFree throws made.
ftanumberFree throw attempts.
ft_pctnumberFree throw percentage (0-1).
orebnumberOffensive rebounds.
drebnumberDefensive rebounds.
rebnumberRebounds per game.
astnumberAssists.
tovnumberTurnovers.
stlnumberSteals.
blknumberBlocks.
blkanumberShot attempts blocked by opponents (blocks against).
pfnumberPersonal fouls.
pfdnumberPersonal fouls drawn.
ptsnumberPoints scored.
plus_minusnumberPlus/minus point differential while on court.
fgm_ranknumber | bigintLeague rank of the row's field goals made for the season and split.
fga_ranknumber | bigintLeague rank of the row's field goals attempted for the season and split.
fg_pct_ranknumber | bigintLeague rank of the row's field goal percentage for the season and split.
fg3m_ranknumber | bigintLeague rank of the row's three-point field goals made for the season and split.
fg3a_ranknumber | bigintLeague rank of the row's three-point field goals attempted for the season and split.
fg3_pct_ranknumber | bigintLeague rank of the row's three-point field goal percentage for the season and split.
ftm_ranknumber | bigintLeague rank of the row's free throws made for the season and split.
fta_ranknumber | bigintLeague rank of the row's free throws attempted for the season and split.
ft_pct_ranknumber | bigintLeague rank of the row's free throw percentage for the season and split.
oreb_ranknumber | bigintLeague rank of the row's offensive rebounds for the season and split.
dreb_ranknumber | bigintLeague rank of the row's defensive rebounds for the season and split.
reb_ranknumber | bigintLeague rank of the row's total rebounds for the season and split.
ast_ranknumber | bigintLeague rank of the row's assists for the season and split.
tov_ranknumber | bigintLeague rank of the row's turnovers for the season and split.
stl_ranknumber | bigintLeague rank of the row's steals for the season and split.
blk_ranknumber | bigintLeague rank of the row's blocked shots for the season and split.
blka_ranknumber | bigintLeague rank of the row's shot attempts blocked by opponents (blocks against) for the season and split.
pf_ranknumber | bigintLeague rank of the row's personal fouls committed for the season and split.
pfd_ranknumber | bigintLeague rank of the row's personal fouls drawn for the season and split.
pts_ranknumber | bigintLeague rank of the row's points scored for the season and split.
plus_minus_ranknumber | bigintLeague rank of the row's plus-minus point differential while on the floor for the season and split.
pts_off_tovnumberPoints scored off opponent turnovers over the split.
pts_2nd_chancenumberSecond-chance points over the split.
pts_fbnumberFast-break points over the split.
pts_paintnumberPoints in the paint over the split.
opp_pts_off_tovnumberOpponent points scored off opponent turnovers allowed over the split.
opp_pts_2nd_chancenumberOpponent second-chance points allowed over the split.
opp_pts_fbnumberOpponent fast-break points allowed over the split.
opp_pts_paintnumberOpponent points in the paint allowed over the split.
pts_off_tov_ranknumber | bigintLeague rank of the row's points scored off opponent turnovers for the season and split.
pts_2nd_chance_ranknumber | bigintLeague rank of the row's second-chance points for the season and split.
pts_fb_ranknumber | bigintLeague rank of the row's fast-break points for the season and split.
pts_paint_ranknumber | bigintLeague rank of the row's points in the paint for the season and split.
opp_pts_off_tov_ranknumber | bigintLeague rank of the row's opponent points scored off opponent turnovers for the season and split.
opp_pts_2nd_chance_ranknumber | bigintLeague rank of the row's opponent second-chance points for the season and split.
opp_pts_fb_ranknumber | bigintLeague rank of the row's opponent fast-break points for the season and split.
opp_pts_paint_ranknumber | bigintLeague rank of the row's opponent points in the paint for the season and split.
opp_fgmnumberOpponent field goals made allowed over the split.
opp_fganumberOpponent field goals attempted allowed over the split.
opp_fg_pctnumberOpponent field goal percentage allowed over the split.
opp_fg3mnumberOpponent three-point field goals made allowed over the split.
opp_fg3anumberOpponent three-point field goals attempted allowed over the split.
opp_fg3_pctnumberOpponent three-point field goal percentage allowed over the split.
opp_ftmnumberOpponent free throws made allowed over the split.
opp_ftanumberOpponent free throws attempted allowed over the split.
opp_ft_pctnumberOpponent free throw percentage allowed over the split.
opp_orebnumberOpponent offensive rebounds allowed over the split.
opp_drebnumberOpponent defensive rebounds allowed over the split.
opp_rebnumberOpponent total rebounds allowed over the split.
opp_astnumberOpponent assists allowed over the split.
opp_tovnumberOpponent turnovers allowed over the split.
opp_stlnumberOpponent steals allowed over the split.
opp_blknumberOpponent blocked shots allowed over the split.
opp_blkanumberOpponent shot attempts blocked by opponents (blocks against) allowed over the split.
opp_pfnumberOpponent personal fouls committed allowed over the split.
opp_pfdnumberOpponent personal fouls drawn allowed over the split.
opp_ptsnumberOpponent points.
opp_fgm_ranknumber | bigintLeague rank of the row's opponent field goals made for the season and split.
opp_fga_ranknumber | bigintLeague rank of the row's opponent field goals attempted for the season and split.
opp_fg_pct_ranknumber | bigintLeague rank of the row's opponent field goal percentage for the season and split.
opp_fg3m_ranknumber | bigintLeague rank of the row's opponent three-point field goals made for the season and split.
opp_fg3a_ranknumber | bigintLeague rank of the row's opponent three-point field goals attempted for the season and split.
opp_fg3_pct_ranknumber | bigintLeague rank of the row's opponent three-point field goal percentage for the season and split.
opp_ftm_ranknumber | bigintLeague rank of the row's opponent free throws made for the season and split.
opp_fta_ranknumber | bigintLeague rank of the row's opponent free throws attempted for the season and split.
opp_ft_pct_ranknumber | bigintLeague rank of the row's opponent free throw percentage for the season and split.
opp_oreb_ranknumber | bigintLeague rank of the row's opponent offensive rebounds for the season and split.
opp_dreb_ranknumber | bigintLeague rank of the row's opponent defensive rebounds for the season and split.
opp_reb_ranknumber | bigintLeague rank of the row's opponent total rebounds for the season and split.
opp_ast_ranknumber | bigintLeague rank of the row's opponent assists for the season and split.
opp_tov_ranknumber | bigintLeague rank of the row's opponent turnovers for the season and split.
opp_stl_ranknumber | bigintLeague rank of the row's opponent steals for the season and split.
opp_blk_ranknumber | bigintLeague rank of the row's opponent blocked shots for the season and split.
opp_blka_ranknumber | bigintLeague rank of the row's opponent shot attempts blocked by opponents (blocks against) for the season and split.
opp_pf_ranknumber | bigintLeague rank of the row's opponent personal fouls committed for the season and split.
opp_pfd_ranknumber | bigintLeague rank of the row's opponent personal fouls drawn for the season and split.
opp_pts_ranknumber | bigintLeague rank of the row's opponent points scored for the season and split.
pct_fga_2ptnumberShare of field goal attempts taken as two-pointers, as a decimal.
pct_fga_3ptnumberShare of field goal attempts taken as three-pointers, as a decimal.
pct_pts_2ptnumberShare of points scored on two-point field goals, as a decimal.
pct_pts_2pt_mrnumberShare of points scored on mid-range two-pointers, as a decimal.
pct_pts_3ptnumberShare of points scored on three-pointers, as a decimal.
pct_pts_fbnumberShare of points scored on fast breaks, as a decimal.
pct_pts_ftnumberShare of points scored at the free throw line, as a decimal.
pct_pts_off_tovnumberShare of points scored off opponent turnovers, as a decimal.
pct_pts_paintnumberShare of points scored in the paint, as a decimal.
pct_ast_2pmnumberPercentage of made two-pointers that were assisted, as a decimal.
pct_uast_2pmnumberPercentage of made two-pointers that were unassisted, as a decimal.
pct_ast_3pmnumberPercentage of made three-pointers that were assisted, as a decimal.
pct_uast_3pmnumberPercentage of made three-pointers that were unassisted, as a decimal.
pct_ast_fgmnumberPercentage of made field goals that were assisted, as a decimal.
pct_uast_fgmnumberPercentage of made field goals that were unassisted, as a decimal.
pct_fga_2pt_ranknumber | bigintLeague rank of the row's share of field goal attempts taken as two-pointers for the season and split.
pct_fga_3pt_ranknumber | bigintLeague rank of the row's share of field goal attempts taken as three-pointers for the season and split.
pct_pts_2pt_ranknumber | bigintLeague rank of the row's share of points scored on two-point field goals for the season and split.
pct_pts_2pt_mr_ranknumber | bigintLeague rank of the row's share of points scored on mid-range two-pointers for the season and split.
pct_pts_3pt_ranknumber | bigintLeague rank of the row's share of points scored on three-pointers for the season and split.
pct_pts_fb_ranknumber | bigintLeague rank of the row's share of points scored on fast breaks for the season and split.
pct_pts_ft_ranknumber | bigintLeague rank of the row's share of points scored at the free throw line for the season and split.
pct_pts_off_tov_ranknumber | bigintLeague rank of the row's share of points scored off opponent turnovers for the season and split.
pct_pts_paint_ranknumber | bigintLeague rank of the row's share of points scored in the paint for the season and split.
pct_ast_2pm_ranknumber | bigintLeague rank of the row's percentage of made two-pointers that were assisted for the season and split.
pct_uast_2pm_ranknumber | bigintLeague rank of the row's percentage of made two-pointers that were unassisted for the season and split.
pct_ast_3pm_ranknumber | bigintLeague rank of the row's percentage of made three-pointers that were assisted for the season and split.
pct_uast_3pm_ranknumber | bigintLeague rank of the row's percentage of made three-pointers that were unassisted for the season and split.
pct_ast_fgm_ranknumber | bigintLeague rank of the row's percentage of made field goals that were assisted for the season and split.
pct_uast_fgm_ranknumber | bigintLeague rank of the row's percentage of made field goals that were unassisted for the season and split.

loadNbaStatsLineupsV3​

Deprecated

Use loadNbaStatsGameLineups — same seasons convention. Emits a one-time DeprecationWarning.

Release: nba_stats_game_lineups · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/nba_stats_game_lineups/nba_lineups_{season + 1}.parquet

Pass the season's START year (e.g. 1996 for the 1996-97 season); the asset is keyed by the END year and the loader translates internally, so a season column in the rows carries that END year.

optiontyperequireddescription
seasonsnumber | number[]yesseason(s) to load (>= 1996)
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaStatsLineupsV3({ seasons: 2025 });
// snake_case alias (py/R parity): sdv.nba.load_nba_stats_lineups_v3(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaStatsGameLineupsRow (exported from the package root).

col_nametypedescription
game_idstringUnique game identifier.
action_numbernumber | bigintSequential action number within a game (V3 PBP).
periodnumber | bigintPeriod of the game (1-4 quarters; 5+ for OT).
home_player_1number | bigintNBA Stats player id of the home on-court player 1 of 5 for the row.
home_player_2number | bigintNBA Stats player id of the home on-court player 2 of 5 for the row.
home_player_3number | bigintNBA Stats player id of the home on-court player 3 of 5 for the row.
home_player_4number | bigintNBA Stats player id of the home on-court player 4 of 5 for the row.
home_player_5number | bigintNBA Stats player id of the home on-court player 5 of 5 for the row.
away_player_1number | bigintNBA Stats player id of the away on-court player 1 of 5 for the row.
away_player_2number | bigintNBA Stats player id of the away on-court player 2 of 5 for the row.
away_player_3number | bigintNBA Stats player id of the away on-court player 3 of 5 for the row.
away_player_4number | bigintNBA Stats player id of the away on-court player 4 of 5 for the row.
away_player_5number | bigintNBA Stats player id of the away on-court player 5 of 5 for the row.
seasonnumber | bigintSeason year.

loadNbaStatsOfficials​

Release: nba_stats_officials · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/nba_stats_officials/officials_{season + 1}.parquet

Pass the season's START year (e.g. 1996 for the 1996-97 season); the asset is keyed by the END year and the loader translates internally, so a season column in the rows carries that END year.

optiontyperequireddescription
seasonsnumber | number[]yesseason(s) to load (>= 1996)
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaStatsOfficials({ seasons: 2025 });
// snake_case alias (py/R parity): sdv.nba.load_nba_stats_officials(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaStatsOfficialsRow (exported from the package root).

col_nametypedescription
official_idstringUnique official / referee identifier.
first_namestringPlayer's first name.
last_namestringPlayer's last name.
jersey_numstringJersey number worn by the player.
seasonnumberSeason year.
game_idstringUnique game identifier.
season_type_idstringSeason-type digit: the 3rd character of game_id (and the leading digit of season_id). 1 = preseason, 2 = regular season, 3 = All-Star, 4 = playoffs, 5 = play-in, 6 = NBA Cup final, 9 = international.

loadNbaStatsPbp​

Release: nba_stats_pbp · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/nba_stats_pbp/nba_play_by_play_{season + 1}.parquet

Pass the season's START year (e.g. 1996 for the 1996-97 season); the asset is keyed by the END year and the loader translates internally, so a season column in the rows carries that END year.

optiontyperequireddescription
seasonsnumber | number[]yesseason(s) to load (>= 1996)
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaStatsPbp({ seasons: 2025, columns: ['game_id', 'period', 'clock', 'event_type', 'description'] });
// snake_case alias (py/R parity): sdv.nba.load_nba_stats_pbp(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaStatsPbpRow (exported from the package root).

col_nametypedescription
order_indexnumber | bigintStable within-game ordering index for events after pbpstats-style reordering of the raw feed.
action_numbernumber | bigintSequential action number within a game (V3 PBP).
clockstringGame clock value.
periodnumber | bigintPeriod of the game (1-4 quarters; 5+ for OT).
team_idstringUnique team identifier.
team_tricodestringThree-letter team code (e.g. 'LAS' / 'NYL').
person_idstringUnique player identifier (V3 endpoints).
player_namestringPlayer name.
player_name_istringPlayer name i.
x_legacynumber | bigintV2-format X coordinate (preserved for V3-to-V2 compatibility).
y_legacynumber | bigintV2-format Y coordinate (preserved for V3-to-V2 compatibility).
shot_distancenumber | bigintShot distance from the basket, in feet.
shot_resultstringShot result ('Made' / 'Missed').
is_field_goalnumber | bigint1 if the action was a field goal; 0 otherwise.
score_homestringScore home.
score_awaystringScore away.
points_totalnumber | bigintRunning total of points scored.
locationstringLocation.
descriptionstringLong-form description text.
action_typestringAction type label (e.g. 'Made Shot', 'Substitution').
sub_typestringAction sub-type label.
video_availablenumber | bigintVideo available.
shot_valuenumber | bigintPoint value of the shot (2 or 3).
action_idstringUnique action identifier within a game (V3 PBP).
game_idstringUnique game identifier.
seconds_remainingnumberSeconds remaining in the period.
event_typestringEvent / play type code (V2 PBP).
is_made_shotbooleanWhether the event is a made field goal.
is_missed_shotbooleanWhether the event is a missed field goal.
is_free_throwbooleanWhether the event is a free throw attempt.
is_reboundbooleanWhether the event is a rebound.
is_turnoverboolean
is_foulbooleanWhether the event is a foul.
is_substitutionbooleanWhether the event is a substitution.
is_jump_ballbooleanWhether the event is a jump ball.
is_timeoutbooleanWhether the event is a timeout.
is_periodbooleanWhether the event is a period start or end marker.
possession_numbernumber | bigintPossession number.
off_player_1number | bigintNBA Stats player id of offensive on-court player 1 of 5 during the event.
off_player_2number | bigintNBA Stats player id of offensive on-court player 2 of 5 during the event.
off_player_3number | bigintNBA Stats player id of offensive on-court player 3 of 5 during the event.
off_player_4number | bigintNBA Stats player id of offensive on-court player 4 of 5 during the event.
off_player_5number | bigintNBA Stats player id of offensive on-court player 5 of 5 during the event.
def_player_1number | bigintNBA Stats player id of defensive on-court player 1 of 5 during the event.
def_player_2number | bigintNBA Stats player id of defensive on-court player 2 of 5 during the event.
def_player_3number | bigintNBA Stats player id of defensive on-court player 3 of 5 during the event.
def_player_4number | bigintNBA Stats player id of defensive on-court player 4 of 5 during the event.
def_player_5number | bigintNBA Stats player id of defensive on-court player 5 of 5 during the event.
seasonnumber | bigintSeason year.

loadNbaStatsPossessions​

Release: nba_stats_possessions · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/nba_stats_possessions/nba_possessions_{season + 1}.parquet

Pass the season's START year (e.g. 1996 for the 1996-97 season); the asset is keyed by the END year and the loader translates internally, so a season column in the rows carries that END year.

optiontyperequireddescription
seasonsnumber | number[]yesseason(s) to load (>= 1996)
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaStatsPossessions({ seasons: 2025 });
// snake_case alias (py/R parity): sdv.nba.load_nba_stats_possessions(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaStatsPossessionsRow (exported from the package root).

col_nametypedescription
game_idstringUnique game identifier.
periodnumber | bigintPeriod of the game (1-4 quarters; 5+ for OT).
possession_numbernumber | bigintPossession number.
offense_team_idstringUnique identifier for offense team.
defense_team_idstringNBA Stats team id of the defending team for the possession.
start_order_indexnumber | bigintorder_index of the play-by-play event that starts the possession.
end_order_indexnumber | bigintorder_index of the play-by-play event that ends the possession.
start_seconds_remainingnumberSeconds remaining in the period when the possession started.
end_seconds_remainingnumberSeconds remaining in the period when the possession ended.
pointsnumber | bigintPoints the offense scored in the possession (its score step), plus any score it made while on defense since its previous possession (a technical free throw, say), credited here with its shots.
is_second_chancebooleanWhether the row is a second-chance continuation following an offensive rebound.
number_in_periodnumber | bigintSequential possession number for the offense within the period.
possession_start_typestringHow the possession began (e.g. off a made shot, defensive rebound, turnover, or period start).
count_as_possessionbooleanWhether the row counts as a true possession for per-possession rate stats.
fg2anumber | bigintTwo-point field goal attempts during the possession.
fg2mnumber | bigintTwo-point field goals made during the possession.
fg3anumber | bigintThree-point field goal attempts.
fg3mnumber | bigintThree-point field goals made.
ftanumber | bigintFree throw attempts.
ftmnumber | bigintFree throws made.
orebnumber | bigintOffensive rebounds.
drebnumber | bigintDefensive rebounds.
tovnumber | bigintTurnovers.
off_player_1number | bigintNBA Stats player id of offensive on-court player 1 of 5 for the possession.
off_player_2number | bigintNBA Stats player id of offensive on-court player 2 of 5 for the possession.
off_player_3number | bigintNBA Stats player id of offensive on-court player 3 of 5 for the possession.
off_player_4number | bigintNBA Stats player id of offensive on-court player 4 of 5 for the possession.
off_player_5number | bigintNBA Stats player id of offensive on-court player 5 of 5 for the possession.
def_player_1number | bigintNBA Stats player id of defensive on-court player 1 of 5 for the possession.
def_player_2number | bigintNBA Stats player id of defensive on-court player 2 of 5 for the possession.
def_player_3number | bigintNBA Stats player id of defensive on-court player 3 of 5 for the possession.
def_player_4number | bigintNBA Stats player id of defensive on-court player 4 of 5 for the possession.
def_player_5number | bigintNBA Stats player id of defensive on-court player 5 of 5 for the possession.
lineup_sourcestringProvenance of the on-court lineup identification for the row (how the five-man units were resolved).
seasonnumber | bigintSeason year.

loadNbaStatsGameLineups​

Release: nba_stats_game_lineups · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/nba_stats_game_lineups/nba_lineups_{season + 1}.parquet

Pass the season's START year (e.g. 1996 for the 1996-97 season); the asset is keyed by the END year and the loader translates internally, so a season column in the rows carries that END year.

optiontyperequireddescription
seasonsnumber | number[]yesseason(s) to load (>= 1996)
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaStatsGameLineups({ seasons: 2025 });
// snake_case alias (py/R parity): sdv.nba.load_nba_stats_game_lineups(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaStatsGameLineupsRow (exported from the package root).

col_nametypedescription
game_idstringUnique game identifier.
action_numbernumber | bigintSequential action number within a game (V3 PBP).
periodnumber | bigintPeriod of the game (1-4 quarters; 5+ for OT).
home_player_1number | bigintNBA Stats player id of the home on-court player 1 of 5 for the row.
home_player_2number | bigintNBA Stats player id of the home on-court player 2 of 5 for the row.
home_player_3number | bigintNBA Stats player id of the home on-court player 3 of 5 for the row.
home_player_4number | bigintNBA Stats player id of the home on-court player 4 of 5 for the row.
home_player_5number | bigintNBA Stats player id of the home on-court player 5 of 5 for the row.
away_player_1number | bigintNBA Stats player id of the away on-court player 1 of 5 for the row.
away_player_2number | bigintNBA Stats player id of the away on-court player 2 of 5 for the row.
away_player_3number | bigintNBA Stats player id of the away on-court player 3 of 5 for the row.
away_player_4number | bigintNBA Stats player id of the away on-court player 4 of 5 for the row.
away_player_5number | bigintNBA Stats player id of the away on-court player 5 of 5 for the row.
seasonnumber | bigintSeason year.

loadNbaStatsGameMatchups​

Release: nba_stats_game_matchups · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/nba_stats_game_matchups/game_matchups_{season + 1}.parquet

Pass the season's START year (e.g. 2017 for the 2017-18 season); the asset is keyed by the END year and the loader translates internally, so a season column in the rows carries that END year.

optiontyperequireddescription
seasonsnumber | number[]yesseason(s) to load (>= 2017)
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaStatsGameMatchups({ seasons: 2025 });
// snake_case alias (py/R parity): sdv.nba.load_nba_stats_game_matchups(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaStatsGameMatchupsRow (exported from the package root).

col_nametypedescription
off_team_idstringTeam id of the offensive player. Taken from the payload's team block.
off_team_citystringCity/market of the offensive player's team ("Indiana").
off_team_namestringNickname of the offensive player's team ("Pacers") -- pair with off_team_city for the full club name.
off_team_tricodestringThree-letter abbreviation of the offensive player's team ("IND").
off_team_slugstringURL slug of the offensive player's team ("pacers").
def_team_idstringTeam id of the defender, read from the game envelope's homeTeamId/awayTeamId rather than the nested team object, which is 0 on uncovered captures.
sidestringWhich side of the game the row's team was on: "home" or "away". In game_matchups, where a row carries two teams, it is the OFFENSIVE player's team (off_team_id) -- the defender is always the other side.
off_person_idstringstats.nba.com person id of the offensive player -- the one being guarded.
off_first_namestringFirst name of the offensive player.
off_family_namestringFamily name of the offensive player.
off_name_istringOffensive player's abbreviated display name ("B. Mathurin").
off_player_slugstringURL slug of the offensive player ("bennedict-mathurin").
off_positionstringStarting position of the offensive player as the payload reports it; empty for players who did not start.
off_commentstringAvailability note on the offensive player (DNP reason); empty when they played.
off_jersey_numstringJersey number of the offensive player, as a string (it can carry a leading zero, e.g. "00").
def_person_idstringstats.nba.com person id of the defender guarding the offensive player.
def_first_namestringFirst name of the defender.
def_family_namestringFamily name of the defender.
def_name_istringDefender's abbreviated display name ("J. Allen").
def_player_slugstringURL slug of the defender ("jarrett-allen").
def_jersey_numstringJersey number of the defender, as a string (it can carry a leading zero).
matchup_minutesstringTime the pair were matched up, as the payload's MM:SS string; use matchup_minutes_sort for arithmetic.
matchup_minutes_sortnumberThe same matchup time in seconds, as a float -- the sortable/summable form.
partial_possessionsnumberPossessions credited to the matchup. Fractional because a possession is split across every defender who guarded the ball-handler during it, which is why matchup counting stats do not sum exactly to a player's game totals.
percentage_defender_total_timenumberShare of the defender's floor time spent guarding this offensive player.
percentage_offensive_total_timenumberShare of the offensive player's floor time spent guarded by this defender.
percentage_total_time_both_onnumberShare of the time both players were on the floor together that they were matched up.
switches_onnumber | bigintTimes the defense switched this defender onto the offensive player.
player_pointsnumber | bigintPoints the offensive player scored while guarded by this defender.
team_pointsnumber | bigintPoints the offensive player's team scored while this matchup was on.
matchup_assistsnumber | bigintAssists by the offensive player while guarded by this defender.
matchup_potential_assistsnumber | bigintPasses by the offensive player that would have been assists had the shot fallen, while guarded by this defender.
matchup_turnoversnumber | bigintTurnovers by the offensive player while guarded by this defender.
matchup_blocksnumber | bigintShots by the offensive player blocked by this defender.
matchup_field_goals_madenumber | bigintField goals made by the offensive player against this defender.
matchup_field_goals_attemptednumber | bigintField goals attempted by the offensive player against this defender.
matchup_field_goals_percentagenumberField-goal percentage of the offensive player against this defender.
matchup_three_pointers_madenumber | bigintThree-pointers made by the offensive player against this defender.
matchup_three_pointers_attemptednumber | bigintThree-pointers attempted by the offensive player against this defender.
matchup_three_pointers_percentagenumberThree-point percentage of the offensive player against this defender.
help_blocksnumber | bigintBlocks by this defender on the offensive player when helping off another assignment rather than as the primary defender.
help_field_goals_madenumber | bigintField goals the offensive player made against this defender in help defense.
help_field_goals_attemptednumber | bigintField goals the offensive player attempted against this defender in help defense.
help_field_goals_percentagenumberField-goal percentage allowed by this defender in help defense.
matchup_free_throws_madenumber | bigintFree throws made by the offensive player on trips drawn against this defender.
matchup_free_throws_attemptednumber | bigintFree throws attempted by the offensive player on trips drawn against this defender.
shooting_foulsnumber | bigintShooting fouls committed by this defender on the offensive player.
game_idstringUnique game identifier.
seasonnumberSeason year.
season_type_idstringSeason-type digit: the 3rd character of game_id (and the leading digit of season_id). 1 = preseason, 2 = regular season, 3 = All-Star, 4 = playoffs, 5 = play-in, 6 = NBA Cup final, 9 = international.

loadNbaStatsInjuryReport​

Release: nba_stats_injury_report · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/nba_stats_injury_report/injury_report_{season + 1}.parquet

Pass the season's START year (e.g. 2018 for the 2018-19 season); the asset is keyed by the END year and the loader translates internally, so a season column in the rows carries that END year.

optiontyperequireddescription
seasonsnumber | number[]yesseason(s) to load (>= 2018)
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaStatsInjuryReport({ seasons: 2025 });
// snake_case alias (py/R parity): sdv.nba.load_nba_stats_injury_report(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaStatsInjuryReportRow (exported from the package root).

col_nametypedescription
seasonnumber | bigintSeason as its ENDING year (2026 = the 2025-26 season), the published asset's stamp.
report_datetimeDatePublish time of the report version the row comes from (US Eastern, naive). The league re-publishes the report through the day, so a listing repeats once per version.
game_dateDateGame date as the report prints it (US Eastern).
game_timestringTip time as the report prints it ("07:30 (ET)", no AM/PM).
report_matchupstringGame as the report prints it, away@home tricodes ("CLE@ATL").
away_team_abbrstringAway team tricode as the report prints it.
home_team_abbrstringHome team tricode as the report prints it.
teamstringTeam the listing belongs to, full name as printed ("Cleveland Cavaliers").
report_player_namestringPlayer as the report prints it, "Last, First" ("James, LeBron"); null on a team's NOT YET SUBMITTED row.
current_statusstringListed status in that version: Available, Probable, Questionable, Doubtful or Out; null on a NOT YET SUBMITTED row.
reasonstringReason as printed ("Injury/Illness - Left Ankle; Sprain"); "NOT YET SUBMITTED" for a team that had not filed by that version.
categorystringCategory column of the early report format; last printed 2019-11-14, null after.
previous_statusstringPrevious Status column of the early report format; last printed 2019-12-17, null after.
game_idstringstats.nba.com game id, joined on the game date and tricodes; null for a game postponed off its printed date, a playoff game that was not needed, or a Summer League game.
team_idstringstats.nba.com team id of the listing's team, from its printed name.
person_idstringstats.nba.com person id matched from the printed name (see match_method); null when unmatched.
match_methodstringHow person_id was matched: exact (full name on the team's boxscores that season), family (family name and first initial, unique on the team), league (name unique league-wide that season), roster (name on the team's roster, which lists players who have not played), or null.

loadNbaStatsPbpV3​

Deprecated

Use loadNbaStatsPbp — same seasons convention. Emits a one-time DeprecationWarning.

Release: nba_stats_pbp · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/nba_stats_pbp/nba_play_by_play_{season + 1}.parquet

Pass the season's START year (e.g. 1996 for the 1996-97 season); the asset is keyed by the END year and the loader translates internally, so a season column in the rows carries that END year.

optiontyperequireddescription
seasonsnumber | number[]yesseason(s) to load (>= 1996)
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaStatsPbpV3({ seasons: 2025, columns: ['game_id', 'period', 'clock', 'event_type', 'description'] });
// snake_case alias (py/R parity): sdv.nba.load_nba_stats_pbp_v3(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaStatsPbpRow (exported from the package root).

col_nametypedescription
order_indexnumber | bigintStable within-game ordering index for events after pbpstats-style reordering of the raw feed.
action_numbernumber | bigintSequential action number within a game (V3 PBP).
clockstringGame clock value.
periodnumber | bigintPeriod of the game (1-4 quarters; 5+ for OT).
team_idstringUnique team identifier.
team_tricodestringThree-letter team code (e.g. 'LAS' / 'NYL').
person_idstringUnique player identifier (V3 endpoints).
player_namestringPlayer name.
player_name_istringPlayer name i.
x_legacynumber | bigintV2-format X coordinate (preserved for V3-to-V2 compatibility).
y_legacynumber | bigintV2-format Y coordinate (preserved for V3-to-V2 compatibility).
shot_distancenumber | bigintShot distance from the basket, in feet.
shot_resultstringShot result ('Made' / 'Missed').
is_field_goalnumber | bigint1 if the action was a field goal; 0 otherwise.
score_homestringScore home.
score_awaystringScore away.
points_totalnumber | bigintRunning total of points scored.
locationstringLocation.
descriptionstringLong-form description text.
action_typestringAction type label (e.g. 'Made Shot', 'Substitution').
sub_typestringAction sub-type label.
video_availablenumber | bigintVideo available.
shot_valuenumber | bigintPoint value of the shot (2 or 3).
action_idstringUnique action identifier within a game (V3 PBP).
game_idstringUnique game identifier.
seconds_remainingnumberSeconds remaining in the period.
event_typestringEvent / play type code (V2 PBP).
is_made_shotbooleanWhether the event is a made field goal.
is_missed_shotbooleanWhether the event is a missed field goal.
is_free_throwbooleanWhether the event is a free throw attempt.
is_reboundbooleanWhether the event is a rebound.
is_turnoverboolean
is_foulbooleanWhether the event is a foul.
is_substitutionbooleanWhether the event is a substitution.
is_jump_ballbooleanWhether the event is a jump ball.
is_timeoutbooleanWhether the event is a timeout.
is_periodbooleanWhether the event is a period start or end marker.
possession_numbernumber | bigintPossession number.
off_player_1number | bigintNBA Stats player id of offensive on-court player 1 of 5 during the event.
off_player_2number | bigintNBA Stats player id of offensive on-court player 2 of 5 during the event.
off_player_3number | bigintNBA Stats player id of offensive on-court player 3 of 5 during the event.
off_player_4number | bigintNBA Stats player id of offensive on-court player 4 of 5 during the event.
off_player_5number | bigintNBA Stats player id of offensive on-court player 5 of 5 during the event.
def_player_1number | bigintNBA Stats player id of defensive on-court player 1 of 5 during the event.
def_player_2number | bigintNBA Stats player id of defensive on-court player 2 of 5 during the event.
def_player_3number | bigintNBA Stats player id of defensive on-court player 3 of 5 during the event.
def_player_4number | bigintNBA Stats player id of defensive on-court player 4 of 5 during the event.
def_player_5number | bigintNBA Stats player id of defensive on-court player 5 of 5 during the event.
seasonnumber | bigintSeason year.

loadNbaStatsPlayerBoxscores​

Release: nba_stats_player_boxscores · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/nba_stats_player_boxscores/player_boxscores_{season + 1}.parquet

Pass the season's START year (e.g. 1996 for the 1996-97 season); the asset is keyed by the END year and the loader translates internally, so a season column in the rows carries that END year.

optiontyperequireddescription
seasonsnumber | number[]yesseason(s) to load (>= 1996)
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaStatsPlayerBoxscores({ seasons: 2025 });
// snake_case alias (py/R parity): sdv.nba.load_nba_stats_player_boxscores(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaStatsPlayerBoxscoresRow (exported from the package root).

col_nametypedescription
team_idstringUnique team identifier.
team_citystringCity/market of the team ("Indiana"); pair with team_name for the full club name.
team_namestringFull team display name (e.g. 'Las Vegas Aces').
team_tricodestringThree-letter team code (e.g. 'LAS' / 'NYL').
team_slugstringURL slug of the team ("pacers").
sidestringSide label (e.g. 'home', 'away', or 'overUnder').
person_idstringUnique player identifier (V3 endpoints).
first_namestringPlayer's first name.
family_namestringPlayer's family / last name.
name_istringInitialed name (e.g. 'A. Wilson').
player_slugstringURL-safe player identifier.
positionstringListed roster position (G, F, C, etc.).
commentstringPlayer status / inactive reason (e.g. 'DNP - Coach's Decision', 'Inactive').
jersey_numstringJersey number worn by the player.
minutesstringMinutes played, formatted MM:SS (V3 PT-duration parsed) or decimal minutes (V2).
field_goals_madenumber | bigintField goals made (2-pt + 3-pt).
field_goals_attemptednumber | bigintField goal attempts (2-pt + 3-pt).
field_goals_percentagenumberField goal percentage (0-1 decimal).
three_pointers_madenumber | bigintThree-point field goals made.
three_pointers_attemptednumber | bigintThree-point field goal attempts.
three_pointers_percentagenumberThree-point field goal percentage (0-1 decimal).
free_throws_madenumber | bigintFree throws made.
free_throws_attemptednumber | bigintFree throw attempts.
free_throws_percentagenumberFree throw percentage (0-1 decimal).
rebounds_offensivenumber | bigintOffensive rebounds.
rebounds_defensivenumber | bigintDefensive rebounds.
rebounds_totalnumber | bigintTotal rebounds.
assistsnumber | bigintTotal assists.
stealsnumber | bigintTotal steals.
blocksnumber | bigintTotal blocks.
turnoversnumber | bigintTotal turnovers.
fouls_personalnumber | bigintPersonal fouls.
pointsnumber | bigintPoints scored.
plus_minus_pointsnumberPlus/minus point differential while on court.
game_idstringUnique game identifier.
seasonnumberSeason year.
season_type_idstringSeason-type digit: the 3rd character of game_id (and the leading digit of season_id). 1 = preseason, 2 = regular season, 3 = All-Star, 4 = playoffs, 5 = play-in, 6 = NBA Cup final, 9 = international.

loadNbaStatsPlayerGameLogs​

Release: nba_stats_player_game_logs · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/nba_stats_player_game_logs/player_game_logs_{season + 1}.parquet

Pass the season's START year (e.g. 1996 for the 1996-97 season); the asset is keyed by the END year and the loader translates internally, so a season column in the rows carries that END year.

optiontyperequireddescription
seasonsnumber | number[]yesseason(s) to load (>= 1996)
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaStatsPlayerGameLogs({ seasons: 2025 });
// snake_case alias (py/R parity): sdv.nba.load_nba_stats_player_game_logs(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaStatsPlayerGameLogsRow (exported from the package root).

col_nametypedescription
season_idstringUnique season identifier.
team_idstringUnique team identifier.
team_abbreviationstringShort team abbreviation (e.g. 'LAS').
team_namestringFull team display name (e.g. 'Las Vegas Aces').
game_idstringUnique game identifier.
game_datestringGame date (YYYY-MM-DD).
matchupstringMatchup.
wlstringWl.
minnumber | bigintMinutes played.
fgmnumber | bigintField goals made.
fganumber | bigintField goal attempts.
fg_pctnumberField goal percentage (0-1).
fg3mnumber | bigintThree-point field goals made.
fg3anumber | bigintThree-point field goal attempts.
fg3_pctnumberThree-point field goal percentage (0-1).
ftmnumber | bigintFree throws made.
ftanumber | bigintFree throw attempts.
ft_pctnumberFree throw percentage (0-1).
orebnumber | bigintOffensive rebounds.
drebnumber | bigintDefensive rebounds.
rebnumber | bigintRebounds per game.
astnumber | bigintAssists.
stlnumber | bigintSteals.
blknumber | bigintBlocks.
tovnumber | bigintTurnovers.
pfnumber | bigintPersonal fouls.
ptsnumber | bigintPoints scored.
plus_minusnumber | bigintPlus/minus point differential while on court.
video_availablenumber | bigintVideo available.
seasonnumberSeason year.
season_typestringSeason type (1=pre-season, 2=regular season, 3=postseason, 4=off-season for ESPN; or string label for WNBA Stats).

loadNbaStatsPlayerSeasonStats​

Release: nba_stats_player_season_stats · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/nba_stats_player_season_stats/player_season_stats_{season + 1}.parquet

Pass the season's START year (e.g. 1996 for the 1996-97 season); the asset is keyed by the END year and the loader translates internally, so a season column in the rows carries that END year.

optiontyperequireddescription
seasonsnumber | number[]yesseason(s) to load (>= 1996)
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaStatsPlayerSeasonStats({ seasons: 2025 });
// snake_case alias (py/R parity): sdv.nba.load_nba_stats_player_season_stats(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaStatsPlayerSeasonStatsRow (exported from the package root).

col_nametypedescription
player_idstringUnique player identifier.
player_namestringPlayer name.
nicknamestringTeam or athlete nickname.
team_idstringUnique team identifier.
team_abbreviationstringShort team abbreviation (e.g. 'LAS').
agenumberPlayer age (in years).
gpnumber | bigintGames played.
wnumber | bigintWins.
lnumber | bigintLosses.
w_pctnumberWins percentage (0-1 decimal).
minnumberMinutes played.
e_off_ratingnumberEstimated offensive rating (NBA Stats estimated-metrics family) over the split.
off_ratingnumberOffensive rating (points scored per 100 possessions) over the split.
sp_work_off_ratingnumberOffensive rating carried in the stats API's SP_WORK column set (mirrors off_rating) over the split.
e_def_ratingnumberEstimated defensive rating (NBA Stats estimated-metrics family) over the split.
def_ratingnumberDefensive rating (points allowed per 100 possessions) over the split.
sp_work_def_ratingnumberDefensive rating carried in the stats API's SP_WORK column set (mirrors def_rating) over the split.
e_net_ratingnumberEstimated net rating (NBA Stats estimated-metrics family) over the split.
net_ratingnumberNet rating (off rating - def rating).
sp_work_net_ratingnumberNet rating carried in the stats API's SP_WORK column set (mirrors net_rating) over the split.
ast_pctnumberAssist percentage.
ast_tonumberAssist-to-turnover ratio over the split.
ast_rationumberAssist ratio (assists per 100 possessions used) over the split.
oreb_pctnumberOffensive rebound percentage over the split, as a decimal.
dreb_pctnumberDefensive rebound percentage over the split, as a decimal.
reb_pctnumberTotal rebound percentage over the split, as a decimal.
tm_tov_pctnumberTeam turnover percentage (turnovers per 100 possessions) over the split, as a decimal.
e_tov_pctnumberEstimated turnover percentage (NBA Stats estimated-metrics family) over the split, as a decimal.
efg_pctnumberEffective field goal percentage over the split, as a decimal.
ts_pctnumberTrue shooting percentage (0-1).
usg_pctnumberUsage percentage (share of team plays used while on the floor) over the split, as a decimal.
e_usg_pctnumberEstimated usage percentage (NBA Stats estimated-metrics family) over the split, as a decimal.
e_pacenumberEstimated pace (NBA Stats estimated-metrics family) over the split.
pacenumberPossessions per 48 minutes.
pace_per40numberPace per40.
sp_work_pacenumberPace carried in the stats API's SP_WORK column set (mirrors pace) over the split.
pienumberPlayer Impact Estimate (0-1).
possnumber | bigintPoss.
fgmnumberField goals made.
fganumberField goal attempts.
fgm_pgnumberField goals made per game over the split.
fga_pgnumberField goals attempted per game over the split.
fg_pctnumberField goal percentage (0-1).
gp_ranknumber | bigintLeague rank of the row's games played for the season and split.
w_ranknumber | bigintLeague rank of the row's wins for the season and split.
l_ranknumber | bigintLeague rank of the row's losses for the season and split.
w_pct_ranknumber | bigintLeague rank of the row's win percentage for the season and split.
min_ranknumber | bigintLeague rank of the row's minutes played for the season and split.
e_off_rating_ranknumber | bigintLeague rank of the row's estimated offensive rating (NBA Stats estimated-metrics family) for the season and split.
off_rating_ranknumber | bigintLeague rank of the row's offensive rating (points scored per 100 possessions) for the season and split.
sp_work_off_rating_ranknumber | bigintLeague rank of the row's offensive rating carried in the stats API's SP_WORK column set (mirrors off_rating) for the season and split.
e_def_rating_ranknumber | bigintLeague rank of the row's estimated defensive rating (NBA Stats estimated-metrics family) for the season and split.
def_rating_ranknumber | bigintLeague rank of the row's defensive rating (points allowed per 100 possessions) for the season and split.
sp_work_def_rating_ranknumber | bigintLeague rank of the row's defensive rating carried in the stats API's SP_WORK column set (mirrors def_rating) for the season and split.
e_net_rating_ranknumber | bigintLeague rank of the row's estimated net rating (NBA Stats estimated-metrics family) for the season and split.
net_rating_ranknumber | bigintLeague rank of the row's net rating (offensive minus defensive rating) for the season and split.
sp_work_net_rating_ranknumber | bigintLeague rank of the row's net rating carried in the stats API's SP_WORK column set (mirrors net_rating) for the season and split.
ast_pct_ranknumber | bigintLeague rank of the row's assist percentage (share of teammate field goals assisted while on the floor) for the season and split.
ast_to_ranknumber | bigintLeague rank of the row's assist-to-turnover ratio for the season and split.
ast_ratio_ranknumber | bigintLeague rank of the row's assist ratio (assists per 100 possessions used) for the season and split.
oreb_pct_ranknumber | bigintLeague rank of the row's offensive rebound percentage for the season and split.
dreb_pct_ranknumber | bigintLeague rank of the row's defensive rebound percentage for the season and split.
reb_pct_ranknumber | bigintLeague rank of the row's total rebound percentage for the season and split.
tm_tov_pct_ranknumber | bigintLeague rank of the row's team turnover percentage (turnovers per 100 possessions) for the season and split.
e_tov_pct_ranknumber | bigintLeague rank of the row's estimated turnover percentage (NBA Stats estimated-metrics family) for the season and split.
efg_pct_ranknumber | bigintLeague rank of the row's effective field goal percentage for the season and split.
ts_pct_ranknumber | bigintLeague rank of the row's true shooting percentage for the season and split.
usg_pct_ranknumber | bigintLeague rank of the row's usage percentage (share of team plays used while on the floor) for the season and split.
e_usg_pct_ranknumber | bigintLeague rank of the row's estimated usage percentage (NBA Stats estimated-metrics family) for the season and split.
e_pace_ranknumber | bigintLeague rank of the row's estimated pace (NBA Stats estimated-metrics family) for the season and split.
pace_ranknumber | bigintLeague rank of the row's pace (possessions per 48 minutes) for the season and split.
sp_work_pace_ranknumber | bigintLeague rank of the row's pace carried in the stats API's SP_WORK column set (mirrors pace) for the season and split.
pie_ranknumber | bigintLeague rank of the row's Player Impact Estimate (PIE, the NBA Stats catch-all impact metric) for the season and split.
fgm_ranknumber | bigintLeague rank of the row's field goals made for the season and split.
fga_ranknumber | bigintLeague rank of the row's field goals attempted for the season and split.
fgm_pg_ranknumber | bigintLeague rank of the row's field goals made per game for the season and split.
fga_pg_ranknumber | bigintLeague rank of the row's field goals attempted per game for the season and split.
fg_pct_ranknumber | bigintLeague rank of the row's field goal percentage for the season and split.
team_countnumber | bigintNumber of distinct teams aggregated into the split row.
seasonnumberSeason year.
season_typestringSeason type (1=pre-season, 2=regular season, 3=postseason, 4=off-season for ESPN; or string label for WNBA Stats).
measure_typestringNBA Stats measure type the row was pulled from (e.g. Base, Advanced, Misc, Scoring, Opponent, Usage, Defense).
per_modestringNBA Stats per-mode of the row values (e.g. Totals, PerGame, Per100Possessions).
fg3mnumberThree-point field goals made.
fg3anumberThree-point field goal attempts.
fg3_pctnumberThree-point field goal percentage (0-1).
ftmnumberFree throws made.
ftanumberFree throw attempts.
ft_pctnumberFree throw percentage (0-1).
orebnumberOffensive rebounds.
drebnumberDefensive rebounds.
rebnumberRebounds per game.
astnumberAssists.
tovnumberTurnovers.
stlnumberSteals.
blknumberBlocks.
blkanumberShot attempts blocked by opponents (blocks against).
pfnumberPersonal fouls.
pfdnumberPersonal fouls drawn.
ptsnumberPoints scored.
plus_minusnumberPlus/minus point differential while on court.
nba_fantasy_ptsnumberFantasy points under the NBA's fantasy scoring formula.
dd2number | bigintDouble-doubles recorded over the split.
td3number | bigintTriple-doubles recorded over the split.
wnba_fantasy_ptsnumberFantasy points under the WNBA's fantasy scoring formula.
fg3m_ranknumber | bigintLeague rank of the row's three-point field goals made for the season and split.
fg3a_ranknumber | bigintLeague rank of the row's three-point field goals attempted for the season and split.
fg3_pct_ranknumber | bigintLeague rank of the row's three-point field goal percentage for the season and split.
ftm_ranknumber | bigintLeague rank of the row's free throws made for the season and split.
fta_ranknumber | bigintLeague rank of the row's free throws attempted for the season and split.
ft_pct_ranknumber | bigintLeague rank of the row's free throw percentage for the season and split.
oreb_ranknumber | bigintLeague rank of the row's offensive rebounds for the season and split.
dreb_ranknumber | bigintLeague rank of the row's defensive rebounds for the season and split.
reb_ranknumber | bigintLeague rank of the row's total rebounds for the season and split.
ast_ranknumber | bigintLeague rank of the row's assists for the season and split.
tov_ranknumber | bigintLeague rank of the row's turnovers for the season and split.
stl_ranknumber | bigintLeague rank of the row's steals for the season and split.
blk_ranknumber | bigintLeague rank of the row's blocked shots for the season and split.
blka_ranknumber | bigintLeague rank of the row's shot attempts blocked by opponents (blocks against) for the season and split.
pf_ranknumber | bigintLeague rank of the row's personal fouls committed for the season and split.
pfd_ranknumber | bigintLeague rank of the row's personal fouls drawn for the season and split.
pts_ranknumber | bigintLeague rank of the row's points scored for the season and split.
plus_minus_ranknumber | bigintLeague rank of the row's plus-minus point differential while on the floor for the season and split.
nba_fantasy_pts_ranknumber | bigintLeague rank of the row's NBA fantasy points (league scoring formula) for the season and split.
dd2_ranknumber | bigintLeague rank of the row's double-doubles for the season and split.
td3_ranknumber | bigintLeague rank of the row's triple-doubles for the season and split.
wnba_fantasy_pts_ranknumber | bigintLeague rank of the row's WNBA fantasy points (league scoring formula) for the season and split.
pct_drebnumberShare of the team's defensive rebounds accounted for by the player while on the floor, as a decimal.
pct_stlnumberShare of the team's steals accounted for by the player while on the floor, as a decimal.
pct_blknumberShare of the team's blocked shots accounted for by the player while on the floor, as a decimal.
opp_pts_off_tovnumberOpponent points scored off opponent turnovers allowed over the split.
opp_pts_2nd_chancenumberOpponent second-chance points allowed over the split.
opp_pts_fbnumberOpponent fast-break points allowed over the split.
opp_pts_paintnumberOpponent points in the paint allowed over the split.
def_wsnumberDefensive win shares credited to the player (NBA Stats defense dashboard metric).
def_ws_rawnumberUnscaled (raw) defensive win shares value carried alongside def_ws by the NBA Stats API.
pct_dreb_ranknumber | bigintLeague rank of the row's share of the team's defensive rebounds accounted for by the player while on the floor for the season and split.
pct_stl_ranknumber | bigintLeague rank of the row's share of the team's steals accounted for by the player while on the floor for the season and split.
pct_blk_ranknumber | bigintLeague rank of the row's share of the team's blocked shots accounted for by the player while on the floor for the season and split.
opp_pts_off_tov_ranknumber | bigintLeague rank of the row's opponent points scored off opponent turnovers for the season and split.
opp_pts_2nd_chance_ranknumber | bigintLeague rank of the row's opponent second-chance points for the season and split.
opp_pts_fb_ranknumber | bigintLeague rank of the row's opponent fast-break points for the season and split.
opp_pts_paint_ranknumber | bigintLeague rank of the row's opponent points in the paint for the season and split.
def_ws_ranknumber | bigintLeague rank of the row's defensive win shares (NBA Stats defense dashboard metric) for the season and split.
pts_off_tovnumberPoints scored off opponent turnovers over the split.
pts_2nd_chancenumberSecond-chance points over the split.
pts_fbnumberFast-break points over the split.
pts_paintnumberPoints in the paint over the split.
pts_off_tov_ranknumber | bigintLeague rank of the row's points scored off opponent turnovers for the season and split.
pts_2nd_chance_ranknumber | bigintLeague rank of the row's second-chance points for the season and split.
pts_fb_ranknumber | bigintLeague rank of the row's fast-break points for the season and split.
pts_paint_ranknumber | bigintLeague rank of the row's points in the paint for the season and split.
pct_fga_2ptnumberShare of field goal attempts taken as two-pointers, as a decimal.
pct_fga_3ptnumberShare of field goal attempts taken as three-pointers, as a decimal.
pct_pts_2ptnumberShare of points scored on two-point field goals, as a decimal.
pct_pts_2pt_mrnumberShare of points scored on mid-range two-pointers, as a decimal.
pct_pts_3ptnumberShare of points scored on three-pointers, as a decimal.
pct_pts_fbnumberShare of points scored on fast breaks, as a decimal.
pct_pts_ftnumberShare of points scored at the free throw line, as a decimal.
pct_pts_off_tovnumberShare of points scored off opponent turnovers, as a decimal.
pct_pts_paintnumberShare of points scored in the paint, as a decimal.
pct_ast_2pmnumberPercentage of made two-pointers that were assisted, as a decimal.
pct_uast_2pmnumberPercentage of made two-pointers that were unassisted, as a decimal.
pct_ast_3pmnumberPercentage of made three-pointers that were assisted, as a decimal.
pct_uast_3pmnumberPercentage of made three-pointers that were unassisted, as a decimal.
pct_ast_fgmnumberPercentage of made field goals that were assisted, as a decimal.
pct_uast_fgmnumberPercentage of made field goals that were unassisted, as a decimal.
pct_fga_2pt_ranknumber | bigintLeague rank of the row's share of field goal attempts taken as two-pointers for the season and split.
pct_fga_3pt_ranknumber | bigintLeague rank of the row's share of field goal attempts taken as three-pointers for the season and split.
pct_pts_2pt_ranknumber | bigintLeague rank of the row's share of points scored on two-point field goals for the season and split.
pct_pts_2pt_mr_ranknumber | bigintLeague rank of the row's share of points scored on mid-range two-pointers for the season and split.
pct_pts_3pt_ranknumber | bigintLeague rank of the row's share of points scored on three-pointers for the season and split.
pct_pts_fb_ranknumber | bigintLeague rank of the row's share of points scored on fast breaks for the season and split.
pct_pts_ft_ranknumber | bigintLeague rank of the row's share of points scored at the free throw line for the season and split.
pct_pts_off_tov_ranknumber | bigintLeague rank of the row's share of points scored off opponent turnovers for the season and split.
pct_pts_paint_ranknumber | bigintLeague rank of the row's share of points scored in the paint for the season and split.
pct_ast_2pm_ranknumber | bigintLeague rank of the row's percentage of made two-pointers that were assisted for the season and split.
pct_uast_2pm_ranknumber | bigintLeague rank of the row's percentage of made two-pointers that were unassisted for the season and split.
pct_ast_3pm_ranknumber | bigintLeague rank of the row's percentage of made three-pointers that were assisted for the season and split.
pct_uast_3pm_ranknumber | bigintLeague rank of the row's percentage of made three-pointers that were unassisted for the season and split.
pct_ast_fgm_ranknumber | bigintLeague rank of the row's percentage of made field goals that were assisted for the season and split.
pct_uast_fgm_ranknumber | bigintLeague rank of the row's percentage of made field goals that were unassisted for the season and split.
pct_fgmnumberShare of the team's field goals made accounted for by the player while on the floor, as a decimal.
pct_fganumberShare of the team's field goals attempted accounted for by the player while on the floor, as a decimal.
pct_fg3mnumberShare of the team's three-point field goals made accounted for by the player while on the floor, as a decimal.
pct_fg3anumberShare of the team's three-point field goals attempted accounted for by the player while on the floor, as a decimal.
pct_ftmnumberShare of the team's free throws made accounted for by the player while on the floor, as a decimal.
pct_ftanumberShare of the team's free throws attempted accounted for by the player while on the floor, as a decimal.
pct_orebnumberShare of the team's offensive rebounds accounted for by the player while on the floor, as a decimal.
pct_rebnumberShare of the team's total rebounds accounted for by the player while on the floor, as a decimal.
pct_astnumberShare of the team's assists accounted for by the player while on the floor, as a decimal.
pct_tovnumberShare of the team's turnovers accounted for by the player while on the floor, as a decimal.
pct_blkanumberShare of the team's shot attempts blocked by opponents (blocks against) accounted for by the player while on the floor, as a decimal.
pct_pfnumberShare of the team's personal fouls committed accounted for by the player while on the floor, as a decimal.
pct_pfdnumberShare of the team's personal fouls drawn accounted for by the player while on the floor, as a decimal.
pct_ptsnumberShare of the team's points scored accounted for by the player while on the floor, as a decimal.
pct_fgm_ranknumber | bigintLeague rank of the row's share of the team's field goals made accounted for by the player while on the floor for the season and split.
pct_fga_ranknumber | bigintLeague rank of the row's share of the team's field goals attempted accounted for by the player while on the floor for the season and split.
pct_fg3m_ranknumber | bigintLeague rank of the row's share of the team's three-point field goals made accounted for by the player while on the floor for the season and split.
pct_fg3a_ranknumber | bigintLeague rank of the row's share of the team's three-point field goals attempted accounted for by the player while on the floor for the season and split.
pct_ftm_ranknumber | bigintLeague rank of the row's share of the team's free throws made accounted for by the player while on the floor for the season and split.
pct_fta_ranknumber | bigintLeague rank of the row's share of the team's free throws attempted accounted for by the player while on the floor for the season and split.
pct_oreb_ranknumber | bigintLeague rank of the row's share of the team's offensive rebounds accounted for by the player while on the floor for the season and split.
pct_reb_ranknumber | bigintLeague rank of the row's share of the team's total rebounds accounted for by the player while on the floor for the season and split.
pct_ast_ranknumber | bigintLeague rank of the row's share of the team's assists accounted for by the player while on the floor for the season and split.
pct_tov_ranknumber | bigintLeague rank of the row's share of the team's turnovers accounted for by the player while on the floor for the season and split.
pct_blka_ranknumber | bigintLeague rank of the row's share of the team's shot attempts blocked by opponents (blocks against) accounted for by the player while on the floor for the season and split.
pct_pf_ranknumber | bigintLeague rank of the row's share of the team's personal fouls committed accounted for by the player while on the floor for the season and split.
pct_pfd_ranknumber | bigintLeague rank of the row's share of the team's personal fouls drawn accounted for by the player while on the floor for the season and split.
pct_pts_ranknumber | bigintLeague rank of the row's share of the team's points scored accounted for by the player while on the floor for the season and split.

loadNbaStatsPossessionsV3​

Deprecated

Use loadNbaStatsPossessions — same seasons convention. Emits a one-time DeprecationWarning.

Release: nba_stats_possessions · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/nba_stats_possessions/nba_possessions_{season + 1}.parquet

Pass the season's START year (e.g. 1996 for the 1996-97 season); the asset is keyed by the END year and the loader translates internally, so a season column in the rows carries that END year.

optiontyperequireddescription
seasonsnumber | number[]yesseason(s) to load (>= 1996)
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaStatsPossessionsV3({ seasons: 2025 });
// snake_case alias (py/R parity): sdv.nba.load_nba_stats_possessions_v3(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaStatsPossessionsRow (exported from the package root).

col_nametypedescription
game_idstringUnique game identifier.
periodnumber | bigintPeriod of the game (1-4 quarters; 5+ for OT).
possession_numbernumber | bigintPossession number.
offense_team_idstringUnique identifier for offense team.
defense_team_idstringNBA Stats team id of the defending team for the possession.
start_order_indexnumber | bigintorder_index of the play-by-play event that starts the possession.
end_order_indexnumber | bigintorder_index of the play-by-play event that ends the possession.
start_seconds_remainingnumberSeconds remaining in the period when the possession started.
end_seconds_remainingnumberSeconds remaining in the period when the possession ended.
pointsnumber | bigintPoints the offense scored in the possession (its score step), plus any score it made while on defense since its previous possession (a technical free throw, say), credited here with its shots.
is_second_chancebooleanWhether the row is a second-chance continuation following an offensive rebound.
number_in_periodnumber | bigintSequential possession number for the offense within the period.
possession_start_typestringHow the possession began (e.g. off a made shot, defensive rebound, turnover, or period start).
count_as_possessionbooleanWhether the row counts as a true possession for per-possession rate stats.
fg2anumber | bigintTwo-point field goal attempts during the possession.
fg2mnumber | bigintTwo-point field goals made during the possession.
fg3anumber | bigintThree-point field goal attempts.
fg3mnumber | bigintThree-point field goals made.
ftanumber | bigintFree throw attempts.
ftmnumber | bigintFree throws made.
orebnumber | bigintOffensive rebounds.
drebnumber | bigintDefensive rebounds.
tovnumber | bigintTurnovers.
off_player_1number | bigintNBA Stats player id of offensive on-court player 1 of 5 for the possession.
off_player_2number | bigintNBA Stats player id of offensive on-court player 2 of 5 for the possession.
off_player_3number | bigintNBA Stats player id of offensive on-court player 3 of 5 for the possession.
off_player_4number | bigintNBA Stats player id of offensive on-court player 4 of 5 for the possession.
off_player_5number | bigintNBA Stats player id of offensive on-court player 5 of 5 for the possession.
def_player_1number | bigintNBA Stats player id of defensive on-court player 1 of 5 for the possession.
def_player_2number | bigintNBA Stats player id of defensive on-court player 2 of 5 for the possession.
def_player_3number | bigintNBA Stats player id of defensive on-court player 3 of 5 for the possession.
def_player_4number | bigintNBA Stats player id of defensive on-court player 4 of 5 for the possession.
def_player_5number | bigintNBA Stats player id of defensive on-court player 5 of 5 for the possession.
lineup_sourcestringProvenance of the on-court lineup identification for the row (how the five-man units were resolved).
seasonnumber | bigintSeason year.

loadNbaStatsRosters​

Release: nba_stats_rosters · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/nba_stats_rosters/rosters_{season + 1}.parquet

Pass the season's START year (e.g. 1996 for the 1996-97 season); the asset is keyed by the END year and the loader translates internally, so a season column in the rows carries that END year.

optiontyperequireddescription
seasonsnumber | number[]yesseason(s) to load (>= 1996)
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaStatsRosters({ seasons: 2025 });
// snake_case alias (py/R parity): sdv.nba.load_nba_stats_rosters(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaStatsRostersRow (exported from the package root).

col_nametypedescription
team_idstringUnique team identifier.
seasonnumberSeason year.
league_idstringLeague identifier ('10' = WNBA).
playerstringPlayer name.
nicknamestringTeam or athlete nickname.
player_slugstringURL-safe player identifier.
numstringJersey number worn by the player.
positionstringListed roster position (G, F, C, etc.).
heightstringPlayer height (string e.g. '6-2' or inches).
weightstringPlayer weight in pounds.
birth_datestringDate of birth (YYYY-MM-DD).
agenumberPlayer age (in years).
expstringYears of NBA playing experience entering the season ('R' = rookie).
schoolstringPlayer school / pre-draft team.
player_idstringUnique player identifier.
how_acquiredstringHow the team acquired the player (e.g. draft, trade, free agency).
supplemental_statusnumber | bigintNumeric supplemental roster-status code from the stats.nba.com roster feed.
season_typestringSeason type (1=pre-season, 2=regular season, 3=postseason, 4=off-season for ESPN; or string label for WNBA Stats).

loadNbaStatsShots​

Release: nba_stats_shots · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/nba_stats_shots/shots_{season + 1}.parquet

Pass the season's START year (e.g. 1996 for the 1996-97 season); the asset is keyed by the END year and the loader translates internally, so a season column in the rows carries that END year.

optiontyperequireddescription
seasonsnumber | number[]yesseason(s) to load (>= 1996)
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaStatsShots({ seasons: 2025 });
// snake_case alias (py/R parity): sdv.nba.load_nba_stats_shots(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaStatsShotsRow (exported from the package root).

col_nametypedescription
game_idstringUnique game identifier.
seasonnumberSeason year.
season_type_idstringSeason-type digit: the 3rd character of game_id (and the leading digit of season_id). 1 = preseason, 2 = regular season, 3 = All-Star, 4 = playoffs, 5 = play-in, 6 = NBA Cup final, 9 = international.
periodnumber | bigintPeriod of the game (1-4 quarters; 5+ for OT).
clockstringGame clock value.
team_idstringUnique team identifier.
team_tricodestringThree-letter team code (e.g. 'LAS' / 'NYL').
person_idstringUnique player identifier (V3 endpoints).
player_namestringPlayer name.
action_typestringAction type label (e.g. 'Made Shot', 'Substitution').
sub_typestringAction sub-type label.
shot_resultstringShot result ('Made' / 'Missed').
shot_valuenumber | bigintPoint value of the shot (2 or 3).
shot_distancenumber | bigintShot distance from the basket, in feet.
x_legacynumber | bigintV2-format X coordinate (preserved for V3-to-V2 compatibility).
y_legacynumber | bigintV2-format Y coordinate (preserved for V3-to-V2 compatibility).
descriptionstringLong-form description text.
score_homestringScore home.
score_awaystringScore away.

loadNbaStatsStandings​

Release: nba_stats_standings · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/nba_stats_standings/standings_{season + 1}.parquet

Pass the season's START year (e.g. 1996 for the 1996-97 season); the asset is keyed by the END year and the loader translates internally, so a season column in the rows carries that END year.

optiontyperequireddescription
seasonsnumber | number[]yesseason(s) to load (>= 1996)
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaStatsStandings({ seasons: 2025 });
// snake_case alias (py/R parity): sdv.nba.load_nba_stats_standings(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaStatsStandingsRow (exported from the package root).

col_nametypedescription
league_idstringLeague identifier ('10' = WNBA).
season_idstringUnique season identifier.
team_idstringUnique team identifier.
team_citystringTeam city or region (e.g. 'Las Vegas').
team_namestringFull team display name (e.g. 'Las Vegas Aces').
team_slugstringURL-safe team identifier (e.g. 'lasvegas-aces' / 'aces').
conferencestringConference name.
conference_recordstringConference win-loss record.
playoff_ranknumber | bigintLeague/season rank for playoff.
clinch_indicatorstringPlayoff clinch indicator (e.g. 'x' clinched playoff, 'e' eliminated).
divisionstringTeam division.
division_recordstringWin-loss record against division opponents.
division_ranknumber | bigintTeam's rank within its division.
winsnumber | bigintTotal wins.
lossesnumber | bigintTotal losses.
win_pctnumberWin percentage (0-1 decimal).
league_ranknumber | bigintTeam's rank in the overall league standings.
recordstringOverall win-loss record.
homestringHome.
roadstringRoad.
l10stringLast-ten record.
last10_homestringWin-loss record over the team's last 10 home games.
last10_roadstringWin-loss record over the team's last 10 road games.
otstringOt.
three_pts_or_lessstringWin-loss record in games decided by three points or fewer.
ten_pts_or_morestringWin-loss record in games decided by ten points or more.
long_home_streaknumber | bigintLongest home streak of the season (positive counts wins, negative losses).
str_long_home_streakstringLongest home streak of the season as display text (e.g. "W 5").
long_road_streaknumber | bigintLongest road streak of the season (positive counts wins, negative losses).
str_long_road_streakstringLongest road streak of the season as display text (e.g. "W 5").
long_win_streaknumber | bigintLongest winning streak of the season, in games.
long_loss_streaknumber | bigintLongest losing streak of the season, in games.
current_home_streaknumber | bigintCurrent home streak (positive counts wins, negative losses).
str_current_home_streakstringCurrent home streak as display text (e.g. "L 2").
current_road_streaknumber | bigintCurrent road streak (positive counts wins, negative losses).
str_current_road_streakstringCurrent road streak as display text (e.g. "W 3").
current_streaknumber | bigintCurrent overall streak (positive counts wins, negative losses).
str_current_streakstringCurrent overall streak as display text (e.g. "W 4").
conference_games_backnumberGames behind the conference leader.
division_games_backnumberGames behind the division leader.
clinched_conference_titlenumber | bigintFlag (1/0) for whether the team has clinched the conference title.
clinched_division_titlenumber | bigintFlag (1/0) for whether the team has clinched its division.
clinched_playoff_birthnumber | bigintFlag (1/0) for whether the team has clinched a playoff berth.
clinched_play_innumber | bigintFlag (1/0) for whether the team has clinched a play-in tournament spot.
eliminated_conferencenumber | bigintFlag (1/0) for whether the team is eliminated from conference contention.
eliminated_divisionnumber | bigintFlag (1/0) for whether the team is eliminated from division contention.
ahead_at_halfstringWin-loss record when leading at halftime.
behind_at_halfstringWin-loss record when trailing at halftime.
tied_at_halfstringWin-loss record when tied at halftime.
ahead_at_thirdstringWin-loss record when leading after three quarters.
behind_at_thirdstringWin-loss record when trailing after three quarters.
tied_at_thirdstringWin-loss record when tied after three quarters.
score100_ptsstringWin-loss record when scoring 100 or more points.
opp_score100_ptsstringWin-loss record when the opponent scores 100 or more points.
opp_over500stringWin-loss record against teams with winning (over .500) records.
lead_in_fgpctstringWin-loss record when posting the higher field goal percentage.
lead_in_rebstringWin-loss record when out-rebounding the opponent.
fewer_turnoversstringWin-loss record when committing fewer turnovers than the opponent.
points_pgnumberPoints pg.
opp_points_pgnumberOpponent points pg.
diff_points_pgnumberDiff points pg.
vs_eaststringWin-loss record against Eastern Conference opponents.
vs_atlanticstringWin-loss record against Atlantic Division opponents.
vs_centralstringWin-loss record against Central Division opponents.
vs_southeaststringWin-loss record against Southeast Division opponents.
vs_weststringWin-loss record against Western Conference opponents.
vs_northweststringWin-loss record against Northwest Division opponents.
vs_pacificstringWin-loss record against Pacific Division opponents.
vs_southweststringWin-loss record against Southwest Division opponents.
janstringWin-loss record in games played in January.
febstringWin-loss record in games played in February.
marstringWin-loss record in games played in March.
aprstringWin-loss record in games played in April.
mayunknownWin-loss record in games played in May.
jununknownWin-loss record in games played in June.
julunknownWin-loss record in games played in July.
augunknownWin-loss record in games played in August.
sepunknownWin-loss record in games played in September.
octstringWin-loss record in games played in October.
novstringWin-loss record in games played in November.
decstringWin-loss record in games played in December.
score_80_plusstringWin-loss record when scoring 80 or more points.
opp_score_80_plusstringWin-loss record when the opponent scores 80 or more points.
score_below_80stringWin-loss record when scoring fewer than 80 points.
opp_score_below_80stringWin-loss record when holding the opponent below 80 points.
total_pointsnumber | bigintTotal points scored by the team over the season to date.
opp_total_pointsnumber | bigintTotal points allowed by the team over the season to date.
diff_total_pointsnumber | bigintSeason point differential (points scored minus points allowed).
league_games_backnumberGames behind the overall league leader.
playoff_seedingnumber | bigintTeam's current playoff seed.
clinched_post_seasonnumber | bigintFlag (1/0) for whether the team has clinched any postseason berth.
neutralstringNeutral.
seasonnumberSeason year.
season_typestringSeason type (1=pre-season, 2=regular season, 3=postseason, 4=off-season for ESPN; or string label for WNBA Stats).

loadNbaStatsTeamBoxscores​

Release: nba_stats_team_boxscores · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/nba_stats_team_boxscores/team_boxscores_{season + 1}.parquet

Pass the season's START year (e.g. 1996 for the 1996-97 season); the asset is keyed by the END year and the loader translates internally, so a season column in the rows carries that END year.

optiontyperequireddescription
seasonsnumber | number[]yesseason(s) to load (>= 1996)
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaStatsTeamBoxscores({ seasons: 2025 });
// snake_case alias (py/R parity): sdv.nba.load_nba_stats_team_boxscores(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaStatsTeamBoxscoresRow (exported from the package root).

col_nametypedescription
team_idstringUnique team identifier.
team_citystringCity/market of the team ("Indiana"); pair with team_name for the full club name.
team_namestringFull team display name (e.g. 'Las Vegas Aces').
team_tricodestringThree-letter team code (e.g. 'LAS' / 'NYL').
team_slugstringURL slug of the team ("pacers").
sidestringSide label (e.g. 'home', 'away', or 'overUnder').
minutesstringMinutes played, formatted MM:SS (V3 PT-duration parsed) or decimal minutes (V2).
field_goals_madenumber | bigintField goals made (2-pt + 3-pt).
field_goals_attemptednumber | bigintField goal attempts (2-pt + 3-pt).
field_goals_percentagenumberField goal percentage (0-1 decimal).
three_pointers_madenumber | bigintThree-point field goals made.
three_pointers_attemptednumber | bigintThree-point field goal attempts.
three_pointers_percentagenumberThree-point field goal percentage (0-1 decimal).
free_throws_madenumber | bigintFree throws made.
free_throws_attemptednumber | bigintFree throw attempts.
free_throws_percentagenumberFree throw percentage (0-1 decimal).
rebounds_offensivenumber | bigintOffensive rebounds.
rebounds_defensivenumber | bigintDefensive rebounds.
rebounds_totalnumber | bigintTotal rebounds.
assistsnumber | bigintTotal assists.
stealsnumber | bigintTotal steals.
blocksnumber | bigintTotal blocks.
turnoversnumber | bigintTotal turnovers.
fouls_personalnumber | bigintPersonal fouls.
pointsnumber | bigintPoints scored.
plus_minus_pointsnumberPlus/minus point differential while on court.
game_idstringUnique game identifier.
seasonnumberSeason year.
season_type_idstringSeason-type digit: the 3rd character of game_id (and the leading digit of season_id). 1 = preseason, 2 = regular season, 3 = All-Star, 4 = playoffs, 5 = play-in, 6 = NBA Cup final, 9 = international.

loadNbaStatsTeamSeasonStats​

Release: nba_stats_team_season_stats · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/nba_stats_team_season_stats/team_season_stats_{season + 1}.parquet

Pass the season's START year (e.g. 1996 for the 1996-97 season); the asset is keyed by the END year and the loader translates internally, so a season column in the rows carries that END year.

optiontyperequireddescription
seasonsnumber | number[]yesseason(s) to load (>= 1996)
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaStatsTeamSeasonStats({ seasons: 2025 });
// snake_case alias (py/R parity): sdv.nba.load_nba_stats_team_season_stats(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaStatsTeamSeasonStatsRow (exported from the package root).

col_nametypedescription
team_idstringUnique team identifier.
team_namestringFull team display name (e.g. 'Las Vegas Aces').
gpnumber | bigintGames played.
wnumber | bigintWins.
lnumber | bigintLosses.
w_pctnumberWins percentage (0-1 decimal).
minnumberMinutes played.
e_off_ratingnumberEstimated offensive rating (NBA Stats estimated-metrics family) over the split.
off_ratingnumberOffensive rating (points scored per 100 possessions) over the split.
e_def_ratingnumberEstimated defensive rating (NBA Stats estimated-metrics family) over the split.
def_ratingnumberDefensive rating (points allowed per 100 possessions) over the split.
e_net_ratingnumberEstimated net rating (NBA Stats estimated-metrics family) over the split.
net_ratingnumberNet rating (off rating - def rating).
ast_pctnumberAssist percentage.
ast_tonumberAssist-to-turnover ratio over the split.
ast_rationumberAssist ratio (assists per 100 possessions used) over the split.
oreb_pctnumberOffensive rebound percentage over the split, as a decimal.
dreb_pctnumberDefensive rebound percentage over the split, as a decimal.
reb_pctnumberTotal rebound percentage over the split, as a decimal.
tm_tov_pctnumberTeam turnover percentage (turnovers per 100 possessions) over the split, as a decimal.
efg_pctnumberEffective field goal percentage over the split, as a decimal.
ts_pctnumberTrue shooting percentage (0-1).
e_pacenumberEstimated pace (NBA Stats estimated-metrics family) over the split.
pacenumberPossessions per 48 minutes.
pace_per40numberPace per40.
possnumber | bigintPoss.
pienumberPlayer Impact Estimate (0-1).
gp_ranknumber | bigintLeague rank of the row's games played for the season and split.
w_ranknumber | bigintLeague rank of the row's wins for the season and split.
l_ranknumber | bigintLeague rank of the row's losses for the season and split.
w_pct_ranknumber | bigintLeague rank of the row's win percentage for the season and split.
min_ranknumber | bigintLeague rank of the row's minutes played for the season and split.
off_rating_ranknumber | bigintLeague rank of the row's offensive rating (points scored per 100 possessions) for the season and split.
def_rating_ranknumber | bigintLeague rank of the row's defensive rating (points allowed per 100 possessions) for the season and split.
net_rating_ranknumber | bigintLeague rank of the row's net rating (offensive minus defensive rating) for the season and split.
ast_pct_ranknumber | bigintLeague rank of the row's assist percentage (share of teammate field goals assisted while on the floor) for the season and split.
ast_to_ranknumber | bigintLeague rank of the row's assist-to-turnover ratio for the season and split.
ast_ratio_ranknumber | bigintLeague rank of the row's assist ratio (assists per 100 possessions used) for the season and split.
oreb_pct_ranknumber | bigintLeague rank of the row's offensive rebound percentage for the season and split.
dreb_pct_ranknumber | bigintLeague rank of the row's defensive rebound percentage for the season and split.
reb_pct_ranknumber | bigintLeague rank of the row's total rebound percentage for the season and split.
tm_tov_pct_ranknumber | bigintLeague rank of the row's team turnover percentage (turnovers per 100 possessions) for the season and split.
efg_pct_ranknumber | bigintLeague rank of the row's effective field goal percentage for the season and split.
ts_pct_ranknumber | bigintLeague rank of the row's true shooting percentage for the season and split.
pace_ranknumber | bigintLeague rank of the row's pace (possessions per 48 minutes) for the season and split.
pie_ranknumber | bigintLeague rank of the row's Player Impact Estimate (PIE, the NBA Stats catch-all impact metric) for the season and split.
seasonnumberSeason year.
season_typestringSeason type (1=pre-season, 2=regular season, 3=postseason, 4=off-season for ESPN; or string label for WNBA Stats).
measure_typestringNBA Stats measure type the row was pulled from (e.g. Base, Advanced, Misc, Scoring, Opponent, Usage, Defense).
per_modestringNBA Stats per-mode of the row values (e.g. Totals, PerGame, Per100Possessions).
fgmnumberField goals made.
fganumberField goal attempts.
fg_pctnumberField goal percentage (0-1).
fg3mnumberThree-point field goals made.
fg3anumberThree-point field goal attempts.
fg3_pctnumberThree-point field goal percentage (0-1).
ftmnumberFree throws made.
ftanumberFree throw attempts.
ft_pctnumberFree throw percentage (0-1).
orebnumberOffensive rebounds.
drebnumberDefensive rebounds.
rebnumberRebounds per game.
astnumberAssists.
tovnumberTurnovers.
stlnumberSteals.
blknumberBlocks.
blkanumberShot attempts blocked by opponents (blocks against).
pfnumberPersonal fouls.
pfdnumberPersonal fouls drawn.
ptsnumberPoints scored.
plus_minusnumberPlus/minus point differential while on court.
fgm_ranknumber | bigintLeague rank of the row's field goals made for the season and split.
fga_ranknumber | bigintLeague rank of the row's field goals attempted for the season and split.
fg_pct_ranknumber | bigintLeague rank of the row's field goal percentage for the season and split.
fg3m_ranknumber | bigintLeague rank of the row's three-point field goals made for the season and split.
fg3a_ranknumber | bigintLeague rank of the row's three-point field goals attempted for the season and split.
fg3_pct_ranknumber | bigintLeague rank of the row's three-point field goal percentage for the season and split.
ftm_ranknumber | bigintLeague rank of the row's free throws made for the season and split.
fta_ranknumber | bigintLeague rank of the row's free throws attempted for the season and split.
ft_pct_ranknumber | bigintLeague rank of the row's free throw percentage for the season and split.
oreb_ranknumber | bigintLeague rank of the row's offensive rebounds for the season and split.
dreb_ranknumber | bigintLeague rank of the row's defensive rebounds for the season and split.
reb_ranknumber | bigintLeague rank of the row's total rebounds for the season and split.
ast_ranknumber | bigintLeague rank of the row's assists for the season and split.
tov_ranknumber | bigintLeague rank of the row's turnovers for the season and split.
stl_ranknumber | bigintLeague rank of the row's steals for the season and split.
blk_ranknumber | bigintLeague rank of the row's blocked shots for the season and split.
blka_ranknumber | bigintLeague rank of the row's shot attempts blocked by opponents (blocks against) for the season and split.
pf_ranknumber | bigintLeague rank of the row's personal fouls committed for the season and split.
pfd_ranknumber | bigintLeague rank of the row's personal fouls drawn for the season and split.
pts_ranknumber | bigintLeague rank of the row's points scored for the season and split.
plus_minus_ranknumber | bigintLeague rank of the row's plus-minus point differential while on the floor for the season and split.
opp_pts_off_tovnumberOpponent points scored off opponent turnovers allowed over the split.
opp_pts_2nd_chancenumberOpponent second-chance points allowed over the split.
opp_pts_fbnumberOpponent fast-break points allowed over the split.
opp_pts_paintnumberOpponent points in the paint allowed over the split.
opp_pts_off_tov_ranknumber | bigintLeague rank of the row's opponent points scored off opponent turnovers for the season and split.
opp_pts_2nd_chance_ranknumber | bigintLeague rank of the row's opponent second-chance points for the season and split.
opp_pts_fb_ranknumber | bigintLeague rank of the row's opponent fast-break points for the season and split.
opp_pts_paint_ranknumber | bigintLeague rank of the row's opponent points in the paint for the season and split.
pts_off_tovnumberPoints scored off opponent turnovers over the split.
pts_2nd_chancenumberSecond-chance points over the split.
pts_fbnumberFast-break points over the split.
pts_paintnumberPoints in the paint over the split.
pts_off_tov_ranknumber | bigintLeague rank of the row's points scored off opponent turnovers for the season and split.
pts_2nd_chance_ranknumber | bigintLeague rank of the row's second-chance points for the season and split.
pts_fb_ranknumber | bigintLeague rank of the row's fast-break points for the season and split.
pts_paint_ranknumber | bigintLeague rank of the row's points in the paint for the season and split.
opp_fgmnumberOpponent field goals made allowed over the split.
opp_fganumberOpponent field goals attempted allowed over the split.
opp_fg_pctnumberOpponent field goal percentage allowed over the split.
opp_fg3mnumberOpponent three-point field goals made allowed over the split.
opp_fg3anumberOpponent three-point field goals attempted allowed over the split.
opp_fg3_pctnumberOpponent three-point field goal percentage allowed over the split.
opp_ftmnumberOpponent free throws made allowed over the split.
opp_ftanumberOpponent free throws attempted allowed over the split.
opp_ft_pctnumberOpponent free throw percentage allowed over the split.
opp_orebnumberOpponent offensive rebounds allowed over the split.
opp_drebnumberOpponent defensive rebounds allowed over the split.
opp_rebnumberOpponent total rebounds allowed over the split.
opp_astnumberOpponent assists allowed over the split.
opp_tovnumberOpponent turnovers allowed over the split.
opp_stlnumberOpponent steals allowed over the split.
opp_blknumberOpponent blocked shots allowed over the split.
opp_blkanumberOpponent shot attempts blocked by opponents (blocks against) allowed over the split.
opp_pfnumberOpponent personal fouls committed allowed over the split.
opp_pfdnumberOpponent personal fouls drawn allowed over the split.
opp_ptsnumberOpponent points.
opp_fgm_ranknumber | bigintLeague rank of the row's opponent field goals made for the season and split.
opp_fga_ranknumber | bigintLeague rank of the row's opponent field goals attempted for the season and split.
opp_fg_pct_ranknumber | bigintLeague rank of the row's opponent field goal percentage for the season and split.
opp_fg3m_ranknumber | bigintLeague rank of the row's opponent three-point field goals made for the season and split.
opp_fg3a_ranknumber | bigintLeague rank of the row's opponent three-point field goals attempted for the season and split.
opp_fg3_pct_ranknumber | bigintLeague rank of the row's opponent three-point field goal percentage for the season and split.
opp_ftm_ranknumber | bigintLeague rank of the row's opponent free throws made for the season and split.
opp_fta_ranknumber | bigintLeague rank of the row's opponent free throws attempted for the season and split.
opp_ft_pct_ranknumber | bigintLeague rank of the row's opponent free throw percentage for the season and split.
opp_oreb_ranknumber | bigintLeague rank of the row's opponent offensive rebounds for the season and split.
opp_dreb_ranknumber | bigintLeague rank of the row's opponent defensive rebounds for the season and split.
opp_reb_ranknumber | bigintLeague rank of the row's opponent total rebounds for the season and split.
opp_ast_ranknumber | bigintLeague rank of the row's opponent assists for the season and split.
opp_tov_ranknumber | bigintLeague rank of the row's opponent turnovers for the season and split.
opp_stl_ranknumber | bigintLeague rank of the row's opponent steals for the season and split.
opp_blk_ranknumber | bigintLeague rank of the row's opponent blocked shots for the season and split.
opp_blka_ranknumber | bigintLeague rank of the row's opponent shot attempts blocked by opponents (blocks against) for the season and split.
opp_pf_ranknumber | bigintLeague rank of the row's opponent personal fouls committed for the season and split.
opp_pfd_ranknumber | bigintLeague rank of the row's opponent personal fouls drawn for the season and split.
opp_pts_ranknumber | bigintLeague rank of the row's opponent points scored for the season and split.
pct_fga_2ptnumberShare of field goal attempts taken as two-pointers, as a decimal.
pct_fga_3ptnumberShare of field goal attempts taken as three-pointers, as a decimal.
pct_pts_2ptnumberShare of points scored on two-point field goals, as a decimal.
pct_pts_2pt_mrnumberShare of points scored on mid-range two-pointers, as a decimal.
pct_pts_3ptnumberShare of points scored on three-pointers, as a decimal.
pct_pts_fbnumberShare of points scored on fast breaks, as a decimal.
pct_pts_ftnumberShare of points scored at the free throw line, as a decimal.
pct_pts_off_tovnumberShare of points scored off opponent turnovers, as a decimal.
pct_pts_paintnumberShare of points scored in the paint, as a decimal.
pct_ast_2pmnumberPercentage of made two-pointers that were assisted, as a decimal.
pct_uast_2pmnumberPercentage of made two-pointers that were unassisted, as a decimal.
pct_ast_3pmnumberPercentage of made three-pointers that were assisted, as a decimal.
pct_uast_3pmnumberPercentage of made three-pointers that were unassisted, as a decimal.
pct_ast_fgmnumberPercentage of made field goals that were assisted, as a decimal.
pct_uast_fgmnumberPercentage of made field goals that were unassisted, as a decimal.
pct_fga_2pt_ranknumber | bigintLeague rank of the row's share of field goal attempts taken as two-pointers for the season and split.
pct_fga_3pt_ranknumber | bigintLeague rank of the row's share of field goal attempts taken as three-pointers for the season and split.
pct_pts_2pt_ranknumber | bigintLeague rank of the row's share of points scored on two-point field goals for the season and split.
pct_pts_2pt_mr_ranknumber | bigintLeague rank of the row's share of points scored on mid-range two-pointers for the season and split.
pct_pts_3pt_ranknumber | bigintLeague rank of the row's share of points scored on three-pointers for the season and split.
pct_pts_fb_ranknumber | bigintLeague rank of the row's share of points scored on fast breaks for the season and split.
pct_pts_ft_ranknumber | bigintLeague rank of the row's share of points scored at the free throw line for the season and split.
pct_pts_off_tov_ranknumber | bigintLeague rank of the row's share of points scored off opponent turnovers for the season and split.
pct_pts_paint_ranknumber | bigintLeague rank of the row's share of points scored in the paint for the season and split.
pct_ast_2pm_ranknumber | bigintLeague rank of the row's percentage of made two-pointers that were assisted for the season and split.
pct_uast_2pm_ranknumber | bigintLeague rank of the row's percentage of made two-pointers that were unassisted for the season and split.
pct_ast_3pm_ranknumber | bigintLeague rank of the row's percentage of made three-pointers that were assisted for the season and split.
pct_uast_3pm_ranknumber | bigintLeague rank of the row's percentage of made three-pointers that were unassisted for the season and split.
pct_ast_fgm_ranknumber | bigintLeague rank of the row's percentage of made field goals that were assisted for the season and split.
pct_uast_fgm_ranknumber | bigintLeague rank of the row's percentage of made field goals that were unassisted for the season and split.

loadNbaPlayerCrosswalk​

Release: nba_crosswalk · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/nba_crosswalk/nba_player_crosswalk_{season}.parquet

optiontyperequireddescription
seasonsnumber | number[]yesseason(s) to load (>= 2002)
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaPlayerCrosswalk({ seasons: 2026 });
// snake_case alias (py/R parity): sdv.nba.load_nba_player_crosswalk(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaPlayerCrosswalkRow (exported from the package root).

col_nametypedescription
seasonnumberSeason year.
espn_team_idstringESPN team id (canonical key).
team_abbreviationstringShort team abbreviation (e.g. 'LAS').
player_namestringPlayer name.
espn_athlete_idstringESPN athlete id.
espn_full_namestringESPN full name.
espn_jerseystringESPN jersey number.
espn_positionstringESPN position abbreviation.
nba_player_idstringNBA Stats player id side of the ESPN-to-NBA player crosswalk.
nba_player_namestringPlayer name as listed by the NBA Stats API.
nba_jersey_numstringPlayer's jersey number as listed by the NBA Stats API.
nba_positionstringPlayer's position as listed by the NBA Stats API.
fox_athlete_idstringFox athlete id (NA if unmatched).
fox_playerstringFox player name (NA if unmatched).
fox_jerseystringFox jersey number (NA if unmatched).
fox_position_groupstringFox position group label (NA if unmatched).
yahoo_player_idstringYahoo player id (NA placeholder).
yahoo_player_namestringYahoo player name (NA placeholder).
match_methodstringCombination of matched sources, e.g. "fox+bart" / "fox_only" / "bart_only" / "espn_only".
match_confidencenumberJaro-Winkler score or 1 for exact (NA if none).
match_keysstringNA (reserved for future use).

loadNbaScheduleCrosswalk​

Release: nba_crosswalk · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/nba_crosswalk/nba_schedule_crosswalk_{season}.parquet

optiontyperequireddescription
seasonsnumber | number[]yesseason(s) to load (>= 2002)
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaScheduleCrosswalk({ seasons: 2026 });
// snake_case alias (py/R parity): sdv.nba.load_nba_schedule_crosswalk(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaScheduleCrosswalkRow (exported from the package root).

col_nametypedescription
seasonnumberSeason year.
season_typestringSeason type (1=pre-season, 2=regular season, 3=postseason, 4=off-season for ESPN; or string label for WNBA Stats).
game_dateDateGame date (YYYY-MM-DD).
home_espn_team_idstringESPN home team id (NA for bart-only rows).
away_espn_team_idstringESPN away team id (NA for bart-only rows).
espn_game_idstringESPN game id (NA for bart-only rows).
nba_game_idstringNBA Stats 10-character game id matched to the ESPN game in the crosswalk.
nba_game_codestringNBA game code (date and matchup string) from the NBA Stats API.
nba_home_team_idstringNBA Stats team id of the home team.
nba_away_team_idstringNBA Stats team id of the away team.
fox_game_idstringFox Sports game id (string), filled when the crosswalk is built with fox=True from Fox's per-date scoreboard, matched on the US-Eastern date and the two teams in either home/away order; null otherwise and where Fox lists no matching game.
fox_home_team_idstringFox Sports team id of the side Fox calls home in fox_game_id, filled with fox=True; at a neutral site it can be the reverse of home_espn_team_id. Null when fox_game_id is.
fox_away_team_idstringFox Sports team id of the side Fox calls away in fox_game_id, filled with fox=True; at a neutral site it can be the reverse of away_espn_team_id. Null when fox_game_id is.
yahoo_game_idstringYahoo Sports game id, nba.g.<YYYYMMDD US-Eastern date><home ESPN team id, 2 digits> (Yahoo's NBA team ids equal ESPN's), from the 2014 season on; null for earlier seasons.
match_methodstringCombination of matched sources, e.g. "fox+bart" / "fox_only" / "bart_only" / "espn_only".
match_confidencenumberJaro-Winkler score or 1 for exact (NA if none).

loadNbaTeamCrosswalk​

Release: nba_crosswalk · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/nba_crosswalk/nba_team_crosswalk_{season}.parquet

optiontyperequireddescription
seasonsnumber | number[]yesseason(s) to load (>= 2002)
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaTeamCrosswalk({ seasons: 2026 });
// snake_case alias (py/R parity): sdv.nba.load_nba_team_crosswalk(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaTeamCrosswalkRow (exported from the package root).

col_nametypedescription
seasonnumberSeason year.
espn_team_idstringESPN team id (canonical key).
espn_abbreviationstringESPN abbreviation.
espn_display_namestringESPN display name (school + mascot).
espn_short_namestringESPN short name.
espn_locationstringESPN school/location only.
espn_mascotstringESPN mascot/nickname.
nba_team_idstringNBA Stats team id side of the ESPN-to-NBA team crosswalk.
nba_team_abbreviationstringTeam abbreviation as listed by the NBA Stats API.
nba_team_namestringFull NBA team name from the NBA Stats API, city followed by nickname (e.g. 'Boston Celtics').
nba_team_citystringTeam city as listed by the NBA Stats API.
nba_team_slugstringURL-friendly slug for the team's name on NBA Stats.
nba_conferencestringTeam's conference as listed by the NBA Stats API.
nba_divisionstringTeam's division as listed by the NBA Stats API.
fox_team_idstringFox Bifrost team id (NA if unmatched).
fox_team_namestringFox team name (NA if unmatched).
yahoo_team_idstringYahoo Sports team id, nba.t.<espn_team_id> (Yahoo's NBA ids equal ESPN's), from the 2014 season on; null for earlier seasons.
yahoo_team_abbreviationstringYahoo abbreviation (NA placeholder).
yahoo_team_namestringYahoo team name (NA placeholder).
match_methodstringCombination of matched sources, e.g. "fox+bart" / "fox_only" / "bart_only" / "espn_only".
match_confidencenumberJaro-Winkler score or 1 for exact (NA if none).

loadNbaPlayerCore​

Release: espn_nba_player_core · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/espn_nba_player_core/player_core_{season}.parquet

optiontyperequireddescription
seasonsnumber | number[]yesseason(s) to load (>= 1994)
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaPlayerCore({ seasons: 2025 });
// snake_case alias (py/R parity): sdv.nba.load_nba_player_core(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaPlayerCoreRow (exported from the package root).

col_nametypedescription
seasonnumberSeason year.
athlete_idstringUnique athlete identifier (ESPN).
guidstringStable cross-league team GUID.
uidstringESPN UID string.
slugstringURL-safe identifier.
typestringRecord type / category.
first_namestringPlayer's first name.
last_namestringPlayer's last name.
full_namestringPlayer's full name.
display_namestringDisplay name.
short_namestringShort display name.
heightnumberPlayer height (string e.g. '6-2' or inches).
display_heightstringPlayer height in display format (e.g. '6-2').
weightnumberPlayer weight in pounds.
display_weightstringPlayer weight in display format (e.g. '180 lbs').
agenumberPlayer age (in years).
date_of_birthstringDate of birth (YYYY-MM-DD).
birth_citystringBirth city.
birth_statestringBirth state / region.
birth_countrystring
jerseystringJersey number worn by the player.
position_idstringUnique position identifier.
position_namestringListed roster position ('Guard', 'Forward', 'Center').
position_abbreviationstringPosition abbreviation ('G' / 'F' / 'C').
position_display_namestringPosition display name.
college_idstringUnique identifier for college.
current_team_idstring
headshot_hrefstringHeadshot image URL.
experience_yearsnumberExperience years.
status_idstringStatus identifier.
status_namestringStatus label.
status_typestringStatus type.
draft_yearnumberDraft year (4-digit).
draft_roundnumberRound of the draft selection.
draft_selectionnumberDraft selection.
activebooleanTRUE if the row represents an active record (player / team / season).

loadNbaPlayerImpact​

Release: nba_player_impact · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/nba_player_impact/nba_player_impact_{season}.parquet

Coverage

The BOBAN (Bayesian Online Box-Augmented Net) columns boban_filtered_skill, boban_projected_rating and boban_projected_sd were renamed from darko_; the loader returns both names for one release, so move to boban_ now (darko_* is deprecated and will be removed).

optiontyperequireddescription
seasonsnumber | number[]yesseason(s) to load (>= 1996)
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaPlayerImpact({ seasons: 2024 });
// snake_case alias (py/R parity): sdv.nba.load_nba_player_impact(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaPlayerImpactRow (exported from the package root).

col_nametypedescription
player_idstringUnique player identifier.
player_namestringPlayer name.
team_idstringUnique team identifier.
team_abbreviationstringShort team abbreviation (e.g. 'LAS').
team_namestringFull team display name (e.g. 'Las Vegas Aces').
teamsstringNested list of member-team membership spans.
seasonnumber | bigintSeason year.
season_typestringSeason type (1=pre-season, 2=regular season, 3=postseason, 4=off-season for ESPN; or string label for WNBA Stats).
o_rapmnumberOffensive regularized adjusted plus-minus per 100 possessions from the single-season ridge fit over possession-level lineup indicators, at the fixed NBA penalty described under rapm; positive means the player raised his team's scoring rate while on offense.
d_rapmnumberDefensive regularized adjusted plus-minus per 100 possessions, negated from the raw points-allowed coefficient so a positive value marks a defender who suppresses opponent scoring; fit at the fixed NBA penalty described under rapm.
rapmnumberTotal regularized adjusted plus-minus per 100 possessions, exactly the sum of o_rapm and d_rapm. The ridge penalty is one fixed NBA value (2,683, sportsdataverse.nba.nba_rapm.RAPM_PENALTY) in every season and season type, so the scale compares across seasons (regular-season sd 1.53 to 1.99 per 100 over players with 400 possessions, 1997-2026; a shorter season, such as the 1999 lockout (1.53), shrinks harder, so part of a spread difference is sample size). Seasons published before November 2026 were fit at a per-season RidgeCV pick (1,931 or 5,179 in the regular season, up to 37,276 in the playoffs), and their spread jumps with it (sd 1.15 to 2.34): do not compare those values across seasons or with values fit at the fixed penalty.
off_possnumber | bigintNumber of possessions the player was on the floor on offense, the count of design-matrix rows carrying his offensive indicator and therefore the offensive-side sample size behind o_rapm.
def_possnumber | bigintNumber of possessions the player was on the floor on defense, the sample size behind d_rapm; it tracks off_poss almost exactly because substitutions rarely split an offense-defense pair.
o_adj_rapmnumberOffensive prior-informed RAPM per 100 possessions: the ridge fit shrunk toward the player's box-score (SPM) prior. In releases fit with sportsdataverse-py #1137 or later it is relative to an average player, up to any offset the SPM prior carries (possession-weighted league mean near 0: 2025-26 +0.32). A player with no prior (a rookie, or no previous-season SPM) is shrunk toward 0 instead. Releases fit before #1137, whatever their build date, fit no intercept, so the league's points per possession leaked into every coefficient: o_adj_rapm averaged about +12.03 and d_adj_rapm -10.84 (2025-26), a thinly-used player kept less of that offset, and only their sum, adj_rapm, read on a common scale.
d_adj_rapmnumberDefensive prior-informed RAPM per 100 possessions, positive for a defender who lowers opponent scoring. In releases fit with sportsdataverse-py #1137 or later it is relative to an average player, up to any offset the SPM prior carries (possession-weighted league mean near 0: 2025-26 +0.06). Releases fit before #1137, whatever their build date, averaged about -10.84 here (see o_adj_rapm).
adj_rapmnumberTotal prior-informed RAPM per 100 possessions, exactly the sum of o_adj_rapm and d_adj_rapm. Releases fit with sportsdataverse-py #1137 or later fit the ridge with an unpenalized intercept; releases fit before it (whatever their build date) fit none, which leaned this total toward heavy-minute players. Releases fit with #1150 or later also use one ridge penalty per league (NBA 3,728, WNBA 2,683) and the SPM prior doubled (nba_adj_rapm.ADJ_RAPM_PENALTY / ADJ_RAPM_PRIOR_SCALE); earlier ones took a per-season RidgeCV pick, whose spread collapsed in some seasons (WNBA 2004 sd 0.64 against plain RAPM's 1.47).
ospmnumberOffensive statistical plus-minus per 100 possessions: the player's per-100 box-score feature vector scored through ridge coefficients trained on that season's o_rapm target.
dspmnumberDefensive statistical plus-minus per 100 possessions from the same box-score feature vector scored through coefficients trained on the d_rapm target.
spmnumberTotal statistical plus-minus per 100 possessions, exactly the sum of ospm and dspm.
minnumberMinutes played.
gpnumber | bigintGames played.
obpmnumberOffensive box plus-minus per 100 possessions for this season and season type only, team-adjusted on its own so a team's minute-weighted obpm is its offensive rating minus the league's, divided by 5 (a traded player counted with his first team; league minute-weighted mean near 0). Releases fit before sportsdataverse-py #1148 put the whole team adjustment here (regular-season league minute-weighted mean about -2.9).
dbpmnumberDefensive box plus-minus per 100 possessions for this season and season type only: bpm minus obpm, so a team's minute-weighted dbpm is (the league's offensive rating minus its defensive rating) / 5, positive for a good defense (league minute-weighted mean near 0). Releases fit before sportsdataverse-py #1148 left it unadjusted (regular-season league minute-weighted mean about +2.9, mostly positive).
bpmnumberBox plus-minus (BPM 2.0) per 100 possessions for this season and season type only, not a career value: the box-score estimate of total impact. The team adjustment makes a team's minute-weighted bpm its net rating per 100 divided by 5 (the five %Min slots sum to the net rating; a traded player's minutes count with the first team he appears for in the logs). Releases fit before sportsdataverse-py #1148 targeted the full net rating, five times BPM 2.0's team mean.
warnumberWins above replacement, computed as (rapm minus a replacement level of -2.0 per 100) times total possessions divided by 100, divided by a points-per-win constant calibrated each season by regressing team wins on full-season point margin.
boban_filtered_skillnumberBOBAN Kalman-filtered skill estimate at the end of the player's observed multi-season RAPM panel, i.e. his current-form rating after aging drift and possession-weighted observation noise.
boban_projected_ratingnumberOne-season-ahead BOBAN forecast, the filtered skill plus the empirical aging-curve drift at the player's last observed age; both season_type rows of a player-season carry the same value because the projection is not playoff-specific.
boban_projected_sdnumberStandard deviation of the one-season-ahead BOBAN forecast, the square root of the filtered state variance plus the Kalman process variance; it sits at roughly 10.19 for a player with only one season in the panel, whose diffuse prior variance was never updated.
darko_filtered_skillnumberDeprecated alias of boban_filtered_skill, kept for one release; use that column.
darko_projected_ratingnumberDeprecated alias of boban_projected_rating, kept for one release; use that column.
darko_projected_sdnumberDeprecated alias of boban_projected_sd, kept for one release; use that column.
o_rapm_nnumber | bigintPossessions o_rapm is measured over: off_poss, his offensive possessions. o_rapm_rank and o_rapm_pct are set only when it reaches 1,700, the NBA sample at which a player's on-court rating of the same side is half signal (sportsdataverse._common.rate_gates.ON_OFF_MIN_N); a regularized or box rating is less noisy, so the gate is conservative.
d_rapm_nnumber | bigintPossessions d_rapm is measured over: def_poss, his defensive possessions. d_rapm_rank and d_rapm_pct are set only when it reaches 1,600, the NBA sample at which a player's on-court rating of the same side is half signal (sportsdataverse._common.rate_gates.ON_OFF_MIN_N); a regularized or box rating is less noisy, so the gate is conservative.
rapm_nnumber | bigintPossessions rapm is measured over: off_poss + def_poss, his possessions at both ends. rapm_rank and rapm_pct are set only when it reaches 3,200, the NBA sample at which a player's on-court rating of the same side is half signal (sportsdataverse._common.rate_gates.ON_OFF_MIN_N); a regularized or box rating is less noisy, so the gate is conservative.
o_adj_rapm_nnumber | bigintPossessions o_adj_rapm is measured over: off_poss, his offensive possessions. o_adj_rapm_rank and o_adj_rapm_pct are set only when it reaches 1,700, the NBA sample at which a player's on-court rating of the same side is half signal (sportsdataverse._common.rate_gates.ON_OFF_MIN_N); a regularized or box rating is less noisy, so the gate is conservative.
d_adj_rapm_nnumber | bigintPossessions d_adj_rapm is measured over: def_poss, his defensive possessions. d_adj_rapm_rank and d_adj_rapm_pct are set only when it reaches 1,600, the NBA sample at which a player's on-court rating of the same side is half signal (sportsdataverse._common.rate_gates.ON_OFF_MIN_N); a regularized or box rating is less noisy, so the gate is conservative.
adj_rapm_nnumber | bigintPossessions adj_rapm is measured over: off_poss + def_poss, his possessions at both ends. adj_rapm_rank and adj_rapm_pct are set only when it reaches 3,200, the NBA sample at which a player's on-court rating of the same side is half signal (sportsdataverse._common.rate_gates.ON_OFF_MIN_N); a regularized or box rating is less noisy, so the gate is conservative.
ospm_nnumber | bigintPossessions ospm is measured over: off_poss, his offensive possessions. ospm_rank and ospm_pct are set only when it reaches 1,700, the NBA sample at which a player's on-court rating of the same side is half signal (sportsdataverse._common.rate_gates.ON_OFF_MIN_N); a regularized or box rating is less noisy, so the gate is conservative.
dspm_nnumber | bigintPossessions dspm is measured over: def_poss, his defensive possessions. dspm_rank and dspm_pct are set only when it reaches 1,600, the NBA sample at which a player's on-court rating of the same side is half signal (sportsdataverse._common.rate_gates.ON_OFF_MIN_N); a regularized or box rating is less noisy, so the gate is conservative.
spm_nnumber | bigintPossessions spm is measured over: off_poss + def_poss, his possessions at both ends. spm_rank and spm_pct are set only when it reaches 3,200, the NBA sample at which a player's on-court rating of the same side is half signal (sportsdataverse._common.rate_gates.ON_OFF_MIN_N); a regularized or box rating is less noisy, so the gate is conservative.
obpm_nnumber | bigintPossessions obpm is measured over: off_poss, his offensive possessions. obpm_rank and obpm_pct are set only when it reaches 1,700, the NBA sample at which a player's on-court rating of the same side is half signal (sportsdataverse._common.rate_gates.ON_OFF_MIN_N); a regularized or box rating is less noisy, so the gate is conservative.
dbpm_nnumber | bigintPossessions dbpm is measured over: def_poss, his defensive possessions. dbpm_rank and dbpm_pct are set only when it reaches 1,600, the NBA sample at which a player's on-court rating of the same side is half signal (sportsdataverse._common.rate_gates.ON_OFF_MIN_N); a regularized or box rating is less noisy, so the gate is conservative.
bpm_nnumber | bigintPossessions bpm is measured over: off_poss + def_poss, his possessions at both ends. bpm_rank and bpm_pct are set only when it reaches 3,200, the NBA sample at which a player's on-court rating of the same side is half signal (sportsdataverse._common.rate_gates.ON_OFF_MIN_N); a regularized or box rating is less noisy, so the gate is conservative.
war_nnumber | bigintPossessions behind war: off_poss + def_poss. war is a count, not a rate, so its rank has no sample gate.
boban_filtered_skill_nnumber | bigintPossessions behind boban_filtered_skill: off_poss + def_poss, this season's possessions at both ends. The boban filtered skill is a multi-season Kalman estimate with no single-side sample, so it takes the net on-court gate, the closest existing one: boban_filtered_skill_rank and boban_filtered_skill_pct are set only when this reaches 3,200. Conservative, since the filter also carries earlier seasons.
boban_projected_rating_nnumber | bigintPossessions behind boban_projected_rating: off_poss + def_poss, this season's possessions at both ends. The boban next-season projection is a multi-season Kalman estimate with no single-side sample, so it takes the net on-court gate, the closest existing one: boban_projected_rating_rank and boban_projected_rating_pct are set only when this reaches 3,200. Conservative, since the filter also carries earlier seasons.
boban_projected_sd_nnumber | bigintPossessions behind boban_projected_sd: off_poss + def_poss, this season's possessions at both ends. The boban projection's standard deviation is a multi-season Kalman estimate with no single-side sample, so it takes the net on-court gate, the closest existing one: boban_projected_sd_rank and boban_projected_sd_pct are set only when this reaches 3,200. Conservative, since the filter also carries earlier seasons.
o_rapm_ranknumberRank of o_rapm among the season and season type's ranked players, 1 = best (higher is better), ties averaged; null when o_rapm_n is under 1,700, and null when its season or season_type is null, in a group of fewer than 10 ranked players, or when every ranked value in the group is equal. Regular-season gates: a playoff run rarely reaches one.
d_rapm_ranknumberRank of d_rapm among the season and season type's ranked players, 1 = best (higher is better), ties averaged; null when d_rapm_n is under 1,600, and null when its season or season_type is null, in a group of fewer than 10 ranked players, or when every ranked value in the group is equal. Regular-season gates: a playoff run rarely reaches one.
rapm_ranknumberRank of rapm among the season and season type's ranked players, 1 = best (higher is better), ties averaged; null when rapm_n is under 3,200, and null when its season or season_type is null, in a group of fewer than 10 ranked players, or when every ranked value in the group is equal. Regular-season gates: a playoff run rarely reaches one.
o_adj_rapm_ranknumberRank of o_adj_rapm among the season and season type's ranked players, 1 = best (higher is better), ties averaged; null when o_adj_rapm_n is under 1,700, and null when its season or season_type is null, in a group of fewer than 10 ranked players, or when every ranked value in the group is equal. Regular-season gates: a playoff run rarely reaches one.
d_adj_rapm_ranknumberRank of d_adj_rapm among the season and season type's ranked players, 1 = best (higher is better), ties averaged; null when d_adj_rapm_n is under 1,600, and null when its season or season_type is null, in a group of fewer than 10 ranked players, or when every ranked value in the group is equal. Regular-season gates: a playoff run rarely reaches one.
adj_rapm_ranknumberRank of adj_rapm among the season and season type's ranked players, 1 = best (higher is better), ties averaged; null when adj_rapm_n is under 3,200, and null when its season or season_type is null, in a group of fewer than 10 ranked players, or when every ranked value in the group is equal. Regular-season gates: a playoff run rarely reaches one.
ospm_ranknumberRank of ospm among the season and season type's ranked players, 1 = best (higher is better), ties averaged; null when ospm_n is under 1,700, and null when its season or season_type is null, in a group of fewer than 10 ranked players, or when every ranked value in the group is equal. Regular-season gates: a playoff run rarely reaches one.
dspm_ranknumberRank of dspm among the season and season type's ranked players, 1 = best (higher is better), ties averaged; null when dspm_n is under 1,600, and null when its season or season_type is null, in a group of fewer than 10 ranked players, or when every ranked value in the group is equal. Regular-season gates: a playoff run rarely reaches one.
spm_ranknumberRank of spm among the season and season type's ranked players, 1 = best (higher is better), ties averaged; null when spm_n is under 3,200, and null when its season or season_type is null, in a group of fewer than 10 ranked players, or when every ranked value in the group is equal. Regular-season gates: a playoff run rarely reaches one.
obpm_ranknumberRank of obpm among the season and season type's ranked players, 1 = best (higher is better), ties averaged; null when obpm_n is under 1,700, and null when its season or season_type is null, in a group of fewer than 10 ranked players, or when every ranked value in the group is equal. Regular-season gates: a playoff run rarely reaches one.
dbpm_ranknumberRank of dbpm among the season and season type's ranked players, 1 = best (higher is better), ties averaged; null when dbpm_n is under 1,600, and null when its season or season_type is null, in a group of fewer than 10 ranked players, or when every ranked value in the group is equal. Regular-season gates: a playoff run rarely reaches one.
bpm_ranknumberRank of bpm among the season and season type's ranked players, 1 = best (higher is better), ties averaged; null when bpm_n is under 3,200, and null when its season or season_type is null, in a group of fewer than 10 ranked players, or when every ranked value in the group is equal. Regular-season gates: a playoff run rarely reaches one.
war_ranknumberRank of war among the season and season type's ranked players, 1 = best (higher is better), ties averaged; ungated (every player with a finite war qualifies), null when its season or season_type is null, in a group of fewer than 10 ranked players, or when every ranked value in the group is equal.
boban_filtered_skill_ranknumberRank of boban_filtered_skill among the season and season type's ranked players, 1 = best (higher is better), ties averaged; null when boban_filtered_skill_n is under 3,200, when season or season_type is null, in a group of fewer than 10 ranked players, or when every ranked value is equal.
boban_projected_rating_ranknumberRank of boban_projected_rating among the season and season type's ranked players, 1 = best (higher is better), ties averaged; null when boban_projected_rating_n is under 3,200, when season or season_type is null, in a group of fewer than 10 ranked players, or when every ranked value is equal.
boban_projected_sd_ranknumberRank of boban_projected_sd among the season and season type's ranked players, ASCENDING: 1 = the smallest sd, the most certain projection; ties averaged; null when boban_projected_sd_n is under 3,200, when season or season_type is null, in a group of fewer than 10 ranked players, or when every ranked value is equal.
o_rapm_pctnumberPercentile of o_rapm_rank, 100 * (n + 1 - rank) / (n + 1) over the n ranked players of the season and season type; null where the rank is.
d_rapm_pctnumberPercentile of d_rapm_rank, 100 * (n + 1 - rank) / (n + 1) over the n ranked players of the season and season type; null where the rank is.
rapm_pctnumberPercentile of rapm_rank, 100 * (n + 1 - rank) / (n + 1) over the n ranked players of the season and season type; null where the rank is.
o_adj_rapm_pctnumberPercentile of o_adj_rapm_rank, 100 * (n + 1 - rank) / (n + 1) over the n ranked players of the season and season type; null where the rank is.
d_adj_rapm_pctnumberPercentile of d_adj_rapm_rank, 100 * (n + 1 - rank) / (n + 1) over the n ranked players of the season and season type; null where the rank is.
adj_rapm_pctnumberPercentile of adj_rapm_rank, 100 * (n + 1 - rank) / (n + 1) over the n ranked players of the season and season type; null where the rank is.
ospm_pctnumberPercentile of ospm_rank, 100 * (n + 1 - rank) / (n + 1) over the n ranked players of the season and season type; null where the rank is.
dspm_pctnumberPercentile of dspm_rank, 100 * (n + 1 - rank) / (n + 1) over the n ranked players of the season and season type; null where the rank is.
spm_pctnumberPercentile of spm_rank, 100 * (n + 1 - rank) / (n + 1) over the n ranked players of the season and season type; null where the rank is.
obpm_pctnumberPercentile of obpm_rank, 100 * (n + 1 - rank) / (n + 1) over the n ranked players of the season and season type; null where the rank is.
dbpm_pctnumberPercentile of dbpm_rank, 100 * (n + 1 - rank) / (n + 1) over the n ranked players of the season and season type; null where the rank is.
bpm_pctnumberPercentile of bpm_rank, 100 * (n + 1 - rank) / (n + 1) over the n ranked players of the season and season type; null where the rank is.
war_pctnumberPercentile of war_rank, 100 * (n + 1 - rank) / (n + 1) over the n ranked players of the season and season type; null where the rank is.
boban_filtered_skill_pctnumberPercentile of boban_filtered_skill_rank, 100 * (n + 1 - rank) / (n + 1) over the n ranked players of the season and season type; null where the rank is.
boban_projected_rating_pctnumberPercentile of boban_projected_rating_rank, 100 * (n + 1 - rank) / (n + 1) over the n ranked players of the season and season type; null where the rank is.
boban_projected_sd_pctnumberPercentile of boban_projected_sd_rank, 100 * (n + 1 - rank) / (n + 1): high means a more certain projection (a smaller sd), not a larger one; null where the rank is.

loadNbaGroups​

Release: nba_groups · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/nba_groups/nba_groups.parquet

Coverage

One season-less file: one row per group lineage (the league, subdivisions, conferences, divisions) with the first and last season it had members. group_id is SDV's own id (e.g. nba:atlantic) and names a lineage: a rename that keeps continuity keeps the id, a new body gets a new one, and notes records each call. Seasons are the ENDING year (2025 = the 2024-25 season).

optiontyperequireddescription
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaGroups();
// snake_case alias (py/R parity): sdv.nba.load_nba_groups(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaGroupsRow (exported from the package root).

col_nametypedescription
leaguestringLeague code of the table ("nba"); the prefix of every group_id in it.
group_idstringSDV group id, {league}:{slug}. It names a lineage: renames that keep continuity keep the id, and a new body (a new conference, or a merger the sources treat as new) gets a new one.
levelstringHierarchy level of the group: "league", "subdivision", "conference" or "division".
first_seasonnumberFirst season in which the group had at least one member (ENDING year: 2025 = the 2024-25 season).
last_seasonnumberLast season in which the group had at least one member (ENDING year: 2025 = the 2024-25 season).
notesstringBuilder notes on the group: the lineage decisions behind its group_id and any source caveats.

loadNbaGroupSeasons​

Release: nba_groups · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/nba_groups/nba_group_seasons.parquet

Coverage

One season-less file: one row per group per season it existed, with its name, short name, abbreviation and parent group AS OF that season (never today's label applied to the past) and its member count. season is the ENDING year (2025 = the 2024-25 season).

optiontyperequireddescription
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaGroupSeasons();
// snake_case alias (py/R parity): sdv.nba.load_nba_group_seasons(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaGroupSeasonsRow (exported from the package root).

col_nametypedescription
leaguestringLeague code of the table ("nba"); the prefix of every group_id in it.
group_idstringSDV group id, {league}:{slug}. It names a lineage: renames that keep continuity keep the id, and a new body (a new conference, or a merger the sources treat as new) gets a new one.
seasonnumberSeason the row describes (ENDING year: 2025 = the 2024-25 season).
levelstringHierarchy level of the group: "league", "subdivision", "conference" or "division".
namestringFull name of the group as of that season -- the label in use then, not today's name.
short_namestringShort display name of the group as of that season.
abbreviationstringAbbreviation of the group as of that season.
parent_group_idstringgroup_id one level up as of that season (division -> conference -> subdivision -> league); null at the top level or where no higher group applied that season.
n_teamsnumberNumber of member teams in the group that season.

loadNbaGroupAliases​

Release: nba_groups · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/nba_groups/nba_group_aliases.parquet

Coverage

One season-less file: every name, abbreviation, slug and source id that a source (espn, nba_stats, sdv) uses for a group, each with the seasons it is valid for (valid_from / valid_to, inclusive; null = unbounded). Match a source's conference or division label here to reach group_id.

optiontyperequireddescription
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaGroupAliases();
// snake_case alias (py/R parity): sdv.nba.load_nba_group_aliases(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaGroupAliasesRow (exported from the package root).

col_nametypedescription
leaguestringLeague code of the table ("nba"); the prefix of every group_id in it.
group_idstringSDV group id, {league}:{slug}. It names a lineage: renames that keep continuity keep the id, and a new body (a new conference, or a merger the sources treat as new) gets a new one.
sourcestringSource that uses this label or id (in this table: espn, nba_stats, sdv); "sdv" marks SDV's own labels.
source_idstringThe source's own id for the group (ESPN group id, NCAA conf_id, CFBD id, MLB division id) when it has one; null otherwise.
name_kindstringKind of label in value: "name", "short_name", "abbreviation", "slug" or "code".
valuestringThe label exactly as the source writes it; match a source's conference or division label against it to reach group_id.
valid_fromnumberFirst season the alias is valid for, inclusive (ENDING year: 2025 = the 2024-25 season); null = unbounded.
valid_tonumberLast season the alias is valid for, inclusive (ENDING year: 2025 = the 2024-25 season); null = unbounded (still in use).

loadNbaTeamGroupSeasons​

Release: nba_groups · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/nba_groups/nba_team_group_seasons_{season}.parquet

Coverage

One row per team per season: the SDV subdivision, conference and division group ids the team belonged to that season (null where a level does not apply), the team name as of that season, where the membership came from, and whether a second source agreed (null when only one source covers the season). team_id is a string: the ESPN team id; team_id_source names the id space. season is the ENDING year (2025 = the 2024-25 season); seasons 1971-2027.

optiontyperequireddescription
seasonsnumber | number[]yesseason(s) to load (>= 1971)
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaTeamGroupSeasons({ seasons: 2024 });
// snake_case alias (py/R parity): sdv.nba.load_nba_team_group_seasons(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaTeamGroupSeasonsRow (exported from the package root).

col_nametypedescription
leaguestringLeague code of the table ("nba"); the prefix of every group_id in it.
seasonnumberSeason of the membership (ENDING year: 2025 = the 2024-25 season).
team_idstringTeam id as a string: the ESPN team id where ESPN covers the team, otherwise the league's own id; team_id_source says which.
team_id_sourcestringId space of team_id (in this table: espn).
team_namestringTeam name as of that season, not today's.
subdivision_idstringSDV group_id of the team's subdivision that season (e.g. FBS / FCS, Division I); null where the league has no subdivision level.
conference_idstringSDV group_id of the team's conference that season; null where the team had no conference (an independent, or a season played without conferences).
division_idstringSDV group_id of the team's division that season; null where the level does not apply.
sourcestringSource the membership was taken from -- the most reliable per-season source for that era.
sources_agreebooleanWhether a second source agreed on the membership; null when only one source covers the season.
notesstringBuilder notes on the team-season, such as a source disagreement or which of several listed memberships was kept.

loadNbaTeamAliases​

Release: nba_groups · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/nba_groups/nba_team_aliases.parquet

Coverage

One season-less file: every id and name another source (cbs, fox, nba_stats, yahoo) uses for a team, keyed on the ESPN team id (team_id, a string; team_id_source = espn), with the seasons the source shows it on that team (valid_from / valid_to, inclusive; both null for an undated curated map). id_kind says whether source_id is the source's id for this sport's team or one id for every sport of a school; a name-only row has a null source_id. A source id means something only within its source (Fox 25 is not ESPN 25), so filter on source before joining on source_id. Seasons are the ENDING year (2025 = the 2024-25 season).

optiontyperequireddescription
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaTeamAliases();
// snake_case alias (py/R parity): sdv.nba.load_nba_team_aliases(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaTeamAliasesRow (exported from the package root).

col_nametypedescription
leaguestringLeague code of the table ("nba").
team_idstringESPN team id the alias points to, as a string (e.g. "150"); the same id as team_id in load_nba_team_group_seasons. Join on it to reach ESPN team data.
team_id_sourcestringId space of team_id; always "espn" in this table.
sourcestringProvider that uses this id or name (in this table: cbs, fox, nba_stats, yahoo).
source_idstringThe source's own id for the team (e.g. Fox 25, Yahoo nba.t.1); null for a name-only row. It means something only within its source (Fox 25 is Ohio State, ESPN 25 is California), so filter on source first.
id_kindstring"team" when source_id is the source's id for this sport's team, or "school" when it is one id for every sport of a school (the stats.ncaa.org org id, 247Sports institutionKey, On3 orgKey).
name_kindstringKind of label in value: "name" or "abbreviation"; null when the row carries only an id. An id with both a name and an abbreviation is two rows.
valuestringThe source's own spelling of the team name or abbreviation; null when the row carries only an id.
valid_fromnumberFirst season the source shows this id or name on the team, inclusive (ENDING year: 2025 = the 2024-25 season); null, with valid_to, for an undated curated map.
valid_tonumberLast season the source shows this id or name on the team, inclusive (ENDING year: 2025 = the 2024-25 season); null, with valid_from, for an undated curated map. A gap in the source splits the window into two rows.
match_methodstringHow the row was tied to the ESPN team: "deterministic" (a formula or shared key, e.g. Yahoo {league}.t.{espn id}; CFBD ids are ESPN ids), "crosswalk" (carried from a published crosswalk leg), "curated" (a cited map), or for a hand-captured team directory row "alias", "exact_name" (normalized name) or "dict" (curated name dictionary).
evidencestringThe release asset or sportsdataverse/sdv-reference-data file the row came from.

loadNbaPlayerAliases​

Release: nba_groups · asset https://github.com/sportsdataverse/sportsdataverse-data/releases/download/nba_groups/nba_player_aliases.parquet

Coverage

One season-less file: every player id another source (fox, nba_stats, yahoo) uses for an ESPN athlete (athlete_id, a string; athlete_id_source = espn), with the ESPN team the match was made on (team_id) and the seasons it held (valid_from / valid_to, inclusive). match_method and match_confidence carry the matching leg's own method and score. Filter on source before joining on source_player_id. Seasons are the ENDING year (2025 = the 2024-25 season).

optiontyperequireddescription
columnsstring[]noread only these columns
format"rows" | "columns"norow objects (default) or column arrays
maxCellsnumbernosize guard; default scales with the heap, Infinity disables
timeoutMsnumbernodownload timeout in ms (default 300000)
const rows = await sdv.nba.loadNbaPlayerAliases();
// snake_case alias (py/R parity): sdv.nba.load_nba_player_aliases(...)

Returns (one row per record; every column optional and nullable):

Row type: LoadNbaPlayerAliasesRow (exported from the package root).

col_nametypedescription
leaguestringLeague code of the table ("nba").
athlete_idstringESPN athlete id the provider id points to, as a string (e.g. "4433134"). Join on it to reach ESPN player data; the ESPN player loaders (load_nba_player_core and its siblings) carry it as Int32, so cast first with pl.col("athlete_id").cast(pl.Int32).
athlete_id_sourcestringId space of athlete_id; always "espn" in this table.
sourcestringProvider whose player id this is (in this table: fox, nba_stats, yahoo).
source_player_idstringThe source's own player id (e.g. Yahoo nba.p.5637, Fox 76284). It means something only within its source, so filter on source first.
team_idstringESPN team id (a string) the athlete was on in these seasons: the team block the match was made in, an ESPN team-season of load_nba_team_group_seasons.
source_team_idstringThe source's own team id when the matching leg gives one (the season's stats.ncaa.org team id; the Fox / Yahoo / CBS team for the NFL); null otherwise.
valid_fromnumberFirst season the id is on that team, inclusive (ENDING year: 2025 = the 2024-25 season).
valid_tonumberLast season the id is on that team, inclusive (ENDING year: 2025 = the 2024-25 season). A gap or a change of match method splits the window into two rows.
match_methodstringThe matching leg's own method (exact_name, fuzzy_jw, the NCAA tiers name_jersey / name / last_jersey / jaro_winkler, CFB rosters name); crosswalk for a leg that names none; propagated when a pair matched in another season was carried to this provider id; override for a hand-verified override.
match_confidencenumberThe leg's own match confidence where it gives one (the lowest in the window); the NCAA tiers score 1.0 / 0.95 / 0.9 / 0.85; null otherwise.
evidencestringThe release asset or sportsdataverse/sdv-reference-data file the row came from.

Generated by tools/codegen/generate.mjs from tools/codegen/endpoints/releases.yaml (vendored from sdv-py) — see How this library is built.