def get_team_roster_by_year(team_id,
year,
raw=False,
overwrite=False,
verbose=False):
'''
Retrieves roster for a given team in a given season.
:param team_id: NHL assigned id for the team
:param year: season start year (e.g. 2018-2019 season would be 2018)
:param overwrite: overwrites cache if True
:param verbose: shows logs if True
:returns: dataframe of the roster for a given team in a given season
'''
base = 'https://statsapi.web.nhl.com/api/v1/'
url = f'{base}teams/{team_id}/roster?season={year}{year+1}'
response = make_request(url, overwrite=overwrite, verbose=verbose)
if raw: return response['roster']
df = pd.DataFrame.from_records(response['roster'])
# Some df cleanup for easier analysis
df['playerId'] = df['person'].apply(lambda x: x['id'])
df['fullName'] = df['person'].apply(lambda x: x['fullName'])
df['positionName'] = df['position'].apply(lambda x: x['name'])
df['positionCode'] = df['position'].apply(lambda x: x['code'])
df['positionType'] = df['position'].apply(lambda x: x['type'])
df.drop(['person', 'position'], axis=1, inplace=True)
df.insert(loc=0, column='teamId', value=team_id)
return df
def get_all_games_for_team_by_year(team_id,
year,
raw=False,
overwrite=False,
verbose=False):
'''
Retrieves game information for a given team in a given season.
:param team_id: NHL assigned id for the team
:param year: season start year (e.g. 2018-2019 season would be 2018)
:param raw: returns raw json response from web if True
:param overwrite: overwrites cache if True
:param verbose: shows logs if True
:returns: dataframe of game information for a given team in a given season.
'''
base = 'https://statsapi.web.nhl.com/api/v1/'
url = f'{base}schedule?teamId={team_id}&season={year}{year+1}'
response = make_request(url, overwrite=overwrite, verbose=verbose)
if raw: return response['dates']
df = pd.DataFrame.from_records(response['dates']).loc[:, ['date', 'games']]
df['gameId'] = df['games'].apply(lambda x: x[0]['gamePk'])
df['gameType'] = df['games'].apply(lambda x: x[0]['gameType'])
df['season'] = df['games'].apply(lambda x: x[0]['season'])
df['homeId'] = df['games'].apply(
lambda x: x[0]['teams']['home']['team']['id'])
df['awayId'] = df['games'].apply(
lambda x: x[0]['teams']['away']['team']['id'])
df['homeScore'] = df['games'].apply(
lambda x: x[0]['teams']['home']['score'])
df['awayScore'] = df['games'].apply(
lambda x: x[0]['teams']['away']['score'])
df['homeRecordWin'] = df['games'].apply(
lambda x: x[0]['teams']['home']['leagueRecord']['wins'])
df['homeRecordLoss'] = df['games'].apply(
lambda x: x[0]['teams']['home']['leagueRecord']['losses'])
df['homeRecordOT'] = df['games'].apply(lambda x : x[0]['teams']['home']['leagueRecord']['ot'] \
if x[0]['gameType'] == 'R' and year >= 2005 else 0)
df['awayRecordWin'] = df['games'].apply(
lambda x: x[0]['teams']['away']['leagueRecord']['wins'])
df['awayRecordLoss'] = df['games'].apply(
lambda x: x[0]['teams']['away']['leagueRecord']['losses'])
df['awayRecordOT'] = df['games'].apply(lambda x : x[0]['teams']['away']['leagueRecord']['ot'] \
if x[0]['gameType'] == 'R' and year >= 2005 else 0)
df['homeRecordTie'] = df['games'].apply(lambda x : x[0]['teams']['home']['leagueRecord']['ties'] \
if x[0]['gameType'] == 'R' and year < 2005 else 0)
df['awayRecordTie'] = df['games'].apply(lambda x : x[0]['teams']['away']['leagueRecord']['ties'] \
if x[0]['gameType'] == 'R' and year < 2005 else 0)
df['link'] = df['games'].apply(lambda x: x[0]['link'])
df.drop(['games'], axis=1, inplace=True)
df.insert(loc=0, column='teamId', value=team_id)
return df
def get_teams_by_year(year, raw=False, overwrite=False, verbose=False):
'''
Retrieves all teams that played in a given season.
:param year: season start year (e.g. 2018-2019 season would be 2018)
:param raw: returns raw json response from web if True
:param overwrite: overwrites cache if True
:param verbose: shows logs if True
:returns: dataframe of all teams that played in a given season
'''
base = 'https://statsapi.web.nhl.com/api/v1/'
url = f'{base}teams?season={year}{year+1}'
response = make_request(url, overwrite=overwrite, verbose=verbose)
if raw: return response['teams']
df = pd.DataFrame.from_records(response['teams'])
df = df.rename(columns={'id': 'teamId'})
return df
def get_team_information(team_id, raw=False, overwrite=False, verbose=False):
'''
Retrieves information about a given team.
:param team_id: NHL assigned id for the team
:param raw: returns raw json response from web if True
:param overwrite: overwrites cache if True
:param verbose: shows logs if True
:returns: dataframe of information about a given team
'''
base = 'https://statsapi.web.nhl.com/api/v1/'
url = f'{base}teams/{team_id}'
response = make_request(url, overwrite=overwrite, verbose=verbose)
if raw: return response['teams']
df = pd.DataFrame.from_records(response['teams'])
df = df.rename(columns={'id': 'teamId'})
return df
def get_player_game_level_statistics_by_year(player_id, year, raw=False, overwrite=False, verbose=False):
'''
Retrieves game level statistics for a given player for a given season.
:param player_id: NHL assigned id for the player
:param year: season start year (e.g. 2018-2019 season would be 2018)
:param raw: returns raw json response from web if True
:param overwrite: overwrites cache if True
:param verbose: shows logs if True
:returns: dataframe of game level statistics about a given player
'''
base = 'https://statsapi.web.nhl.com/api/v1/'
url = f'{base}people/{player_id}/stats?stats=gameLog&season={year}{year+1}'
response = make_request(url, overwrite=overwrite, verbose=verbose)
if 'stats' not in response.keys() or len(response['stats'][0]['splits']) == 0:
raise PlayerNotFoundException(f'No game level statistics available for player \
{player_id} for year {year}')
if raw: return response['stats'][0]['splits']
df = pd.DataFrame.from_records(response['stats'][0]['splits'])
# Only extract useful fields
df['gameId'] = df['game'].apply(lambda x : x['gamePk'])
df['teamId'] = df['team'].apply(lambda x : x['id'])
df['opponentTeamId'] = df['opponent'].apply(lambda x : x['id'])
# Get consistent stat types
stat_types = list(get_player_statistics_by_year(player_id, year, \
overwrite=overwrite, verbose=verbose).keys())
stat_types = [stat for stat in stat_types if 'PerGame' not in stat \
and stat != 'playerId' and stat != 'season']
for stat_type in stat_types:
df[stat_type] = df['stat'].apply(lambda x : x[stat_type] if \
stat_type in x.keys() else np.nan)
df.insert(loc=0, column='playerId', value=player_id)
df.drop(['game', 'team', 'opponent', 'stat'], axis=1, inplace=True)
return df
def get_team_statistics_rankings_by_year(team_id,
year,
raw=False,
overwrite=False,
verbose=False):
'''
Retrieves statistics rankings for a given team in a given season.
:param team_id: NHL assigned id for the team
:param year: season start year (e.g. 2018-2019 season would be 2018)
:param raw: returns raw json response from web if True
:param overwrite: overwrites cache if True
:param verbose: shows logs if True
:returns: dataframe of statistics rankings for a given team in a given season
'''
base = 'https://statsapi.web.nhl.com/api/v1/'
url = f'{base}teams/{team_id}/stats?season={year}{year+1}'
response = make_request(url, overwrite=overwrite, verbose=verbose)
if raw: return response['stats'][0]['splits'][0]['stat']
df = pd.DataFrame.from_records(response['stats'][1]['splits'][0]['stat'],
index=[0])
df.insert(loc=0, column='teamId', value=team_id)
return df
def get_player_statistics_by_year(player_id, year, raw=False, overwrite=False, verbose=False):
'''
Retrieves statistics for a given player for a given season.
:param player_id: NHL assigned id for the player
:param year: season start year (e.g. 2018-2019 season would be 2018)
:param raw: returns raw json response from web if True
:param overwrite: overwrites cache if True
:param verbose: shows logs if True
:returns: dataframe of statistics about a given player
'''
base = 'https://statsapi.web.nhl.com/api/v1/'
url = f'{base}people/{player_id}/stats?stats=statsSingleSeason&season={year}{year+1}'
response = make_request(url, overwrite=overwrite, verbose=verbose)
if 'stats' not in response.keys() or len(response['stats'][0]['splits']) == 0:
raise PlayerNotFoundException(f'No statistics available for player \
{player_id} for year {year}')
if raw: return response['stats'][0]['splits'][0]['stat']
df = pd.DataFrame.from_records(response['stats'][0]['splits'][0]['stat'], index=[0])
df.insert(loc=0, column='playerId', value=player_id)
df.insert(loc=0, column='season', value=f'{year}{year+1}')
return df