NBA Function Index
sportsdataverse.nba package
sportsdataverse.nba.load_nba_pbp(seasons: List[int])
Load NBA play by play data going back to 2002
Example:
nba_df = sportsdataverse.nba.load_nba_pbp(seasons=range(2002,2022))
Args:
seasons (list): Used to define different seasons. 2002 is the earliest available season.
Returns:
pd.DataFrame: Pandas dataframe containing the
play-by-plays available for the requested seasons.
Raises:
ValueError: If season is less than 2002.
sportsdataverse.nba.load_nba_player_boxscore(seasons: List[int])
Load NBA player boxscore data
Example:
nba_df = sportsdataverse.nba.load_nba_player_boxscore(seasons=range(2002,2022))
Args:
seasons (list): Used to define different seasons. 2002 is the earliest available season.
Returns:
pd.DataFrame: Pandas dataframe containing the
player boxscores available for the requested seasons.
Raises:
ValueError: If season is less than 2002.
sportsdataverse.nba.load_nba_schedule(seasons: List[int])
Load NBA schedule data
Example:
nba_df = sportsdataverse.nba.load_nba_schedule(seasons=range(2002,2022))
Args:
seasons (list): Used to define different seasons. 2002 is the earliest available season.
Returns:
pd.DataFrame: Pandas dataframe containing the
schedule for the requested seasons.
Raises:
ValueError: If season is less than 2002.
sportsdataverse.nba.load_nba_team_boxscore(seasons: List[int])
Load NBA team boxscore data
Example:
nba_df = sportsdataverse.nba.load_nba_team_boxscore(seasons=range(2002,2022))
Args:
seasons (list): Used to define different seasons. 2002 is the earliest available season.
Returns:
pd.DataFrame: Pandas dataframe containing the
team boxscores available for the requested seasons.
Raises:
ValueError: If season is less than 2002.
sportsdataverse.nba.espn_nba_pbp(game_id: int)
nba_pbp() - Pull the game by id - Data from API endpoints - nba/playbyplay, nba/summary
Args:
game_id (int): Unique game_id, can be obtained from nba_schedule().
Returns:
Dict: Dictionary of game data with keys - “gameId”, “plays”, “winprobability”, “boxscore”, “header”, “broadcasts”,
“videos”, “playByPlaySource”, “standings”, “leaders”, “seasonseries”, “timeouts”, “pickcenter”, “againstTheSpread”,
“odds”, “predictor”, “espnWP”, “gameInfo”, “season”
Example:
nba_df = sportsdataverse.nba.espn_nba_pbp(game_id=401307514)
sportsdataverse.nba.espn_nba_calendar(season=None)
nba_calendar - look up the NBA calendar for a given season from ESPN
Args:
season (int): Used to define different seasons. 2002 is the earliest available season.
Returns:
pd.DataFrame: Pandas dataframe containing
calendar dates for the requested season.
Raises:
ValueError: If season is less than 2002.
sportsdataverse.nba.espn_nba_schedule(dates=None, season_type=None)
nba_schedule - look up the NBA schedule for a given date from ESPN
Args:
dates (int): Used to define different seasons. 2002 is the earliest available season.
season_type (int): season type, 1 for pre-season, 2 for regular season, 3 for post-season, 4 for all-star, 5 for off-season
Returns:
pd.DataFrame: Pandas dataframe containing
schedule events for the requested season.