def get_all_pitcher_fps():
url = "https://www.baseball-reference.com/leagues/MLB/{}-standard-pitching.shtml".format(CUR_SEASON)
page = br.get_page(url)
pitchers = br.build_df(br.get_table_by_class(page, 'stats_table'), 1, ['Name', 'Tm', 'Lg'], ['Rk', 'Age'])
url = "https://www.baseball-reference.com/leagues/MLB/{}-reliever-pitching.shtml".format(CUR_SEASON)
relief_page = br.get_page(url)
relievers = br.build_df(br.get_table_by_class(relief_page, 'stats_table'), 1, ['Name', 'Tm'], ['Rk', 'Age'])
url = "https://www.baseball-reference.com/leagues/MLB/{}-starter-pitching.shtml".format(CUR_SEASON)
qs_page = br.get_page(url)
qs_table = br.build_df(br.get_table_by_class(qs_page, 'stats_table'), 1, ['Name', 'Tm'], ['Rk', 'Age'])
url = "https://www.baseball-reference.com/leagues/MLB/{}-basesituation-pitching.shtml".format(CUR_SEASON)
sb_page = br.get_page(url)
sb_table = br.build_df(br.get_table_by_class(sb_page, 'stats_table'), 1, ['Name', 'Tm'], ['Rk', 'Age'])
df2 = pd.merge(pitchers, relievers[['Hold', 'BSv', 'Name', 'Tm']], on=['Name', 'Tm'], how='left').fillna(0)
df3 = pd.merge(df2, qs_table[['QS', 'Name', 'Tm']], on=['Name', 'Tm'], how='left').fillna(0)
df4 = pd.merge(df3, sb_table[['SB', 'Name', 'Tm']], on=['Name', 'Tm'], how='left').fillna(0)
#print(list(df3))
players = df3['Name'].astype('str')
games = df3['G'].astype('float64')
ip = df3['IP'].astype('float64')
w = df3['W'].astype('float64')
l = df3['L'].astype('float64')
cg = df3['CG'].astype('float64')
sv = df3['SV'].astype('float64')
h = df3['H'].astype('float64')
er = df3['ER'].astype('float64')
walks = df3['BB'].astype('float64')
hb = df3['HBP'].astype('float64')
k = df3['SO'].astype('float64')
hld = df3['Hold'].astype('float64')
bsv = df3['BSv'].astype('float64')
#handling half innings
half_ip = []
inns = []
for i in ip:
inn = int(i)
half_i = str(i)[-1]
inns.append(inn)
half_ip.append(float(half_i))
innings = pd.Series(inns)
rem = pd.Series(half_ip)
qs = df3['QS'].astype('float64')
sb = df4['SB'].astype('float64')
fps = (innings * 1.0) + (rem * 0.33) + (w * 9) - (l * 6) + (cg * 7) + (sv * 8) - (h * 0.25) - er - (walks * 0.5) - (hb * 0.5) + k + (hld * 7.5) - (bsv * 3) + (qs * 5) - (sb * 0.25)
fps_g = fps / games
df = build_fp_table(players, fps, fps_g)
return df
def get_finished_df(season, source):
"""Function that takes in a team stat dataset from Fangraphs or Baseball Prospectus and joins in some baseball reference data, namely W-L% and postseason finish"""
if source == 'fg':
war_df = get_fg_team_stats(str(season))
pre_df = teamname_to_abbr(war_df)
start_year = FG_START_YEAR
elif source == 'bp':
warp_df = get_bp_team_stats(str(season))
pre_df = bp.abbr_to_master(warp_df)
start_year = BP_START_YEAR
#print(df)
standings = bref.get_page(
'https://www.baseball-reference.com/leagues/MLB/{}-standings.shtml'.
format(str(season)))
standings_table = bref.get_table_by_id(standings,
'expanded_standings_overall')
standings_df = bref.build_df(standings_table, 0, [
'Tm', 'Lg', 'Strk', 'pythWL', 'vEast', 'vCent', 'vWest', 'Inter',
'Home', 'Road', 'ExInn', '1Run', 'vRHP', 'vLHP', 'last10', 'last20',
'last30', 'gte.500', 'lt.500'
], [])
bref_df = bref.abbr_to_master(standings_df, season)
df = pre_df.merge(bref_df[['Master Team', 'W-L%']],
on='Master Team',
how='left')
#print(df)
if season < int(CUR_SEASON):
if season != 1904 and season != 1994:
champ = get_ws_champ(season, start_year)['Master Team'].values[0]
#print(champ)
champ_row = df['Master Team'] == champ
idx = df.index[champ_row].tolist()[0]
df.loc[idx, 'FINISH'] = 'CHAMPION'
return df
def load_bref_team_sp():
url = "https://www.baseball-reference.com/leagues/MLB/{}-starter-pitching.shtml".format(
CUR_SEASON)
page = bref.get_page(url)
df = bref.build_df(bref.get_table_by_class(page, 'stats_table'), 0, ['Tm'],
['']).sort_values(by=['GmScA'], ascending=False)
refresh_table('MLB_TEAM', 'bref_team_sp', df)
def get_all_batter_fps():
url = "https://www.baseball-reference.com/leagues/MLB/{}-standard-batting.shtml".format(CUR_SEASON)
page = br.get_page(url)
batters = br.build_df(br.get_table_by_class(page, 'stats_table'), 1, ['Name', 'Tm', 'Lg', 'Pos\xa0Summary'], ['Rk', 'Age'])
url = "https://www.baseball-reference.com/leagues/MLB/{}-standard-fielding.shtml".format(CUR_SEASON)
field_page = br.get_page(url)
fielders = br.build_df(br.get_table_by_class(field_page, 'stats_table'), 1, ['Name', 'Tm', 'Lg', 'Pos\xa0Summary'], ['Rk', 'Age'])
url = "https://www.baseball-reference.com/leagues/MLB/{}-specialpos_of-fielding.shtml".format(CUR_SEASON)
of_page = br.get_page(url)
ofs = br.build_df(br.get_table_by_class(of_page, 'stats_table'), 1, ['Name', 'Tm', 'Lg'], ['Rk', 'Age'])
df2 = pd.merge(batters, ofs[['A', 'Name', 'Tm']], on=['Name', 'Tm'], how='left').fillna(0)
df2.rename(columns={'A':'OFA'}, inplace=True)
df3 = pd.merge(df2, fielders[['A', 'E', 'Name', 'Tm']], on=['Name', 'Tm'], how='left').fillna(0)
#batting and fielding data
players = df3['Name'].astype('str')
games = df3['G'].astype('float64')
r = df3['R'].astype('float64')
double = df3['2B'].astype('float64')
triple = df3['3B'].astype('float64')
homer = df3['HR'].astype('float64')
single = df3['H'].astype('float64') - double - triple - homer
rbi = df3['RBI'].astype('float64')
sb = df3['SB'].astype('float64')
cs = df3['CS'].astype('float64')
bb = df3['BB'].astype('float64')
hbp = df3['HBP'].astype('float64')
so = df3['SO'].astype('float64')
e = df3['E'].astype('float64')
ofa = df3['OFA'].astype('float64')
a = df3['A'].astype('float64')
fps = r + (single * 1.0) + (double * 2) + (triple * 3) + (homer * 4) + rbi + (sb * 1.75) - (cs * 0.5) + (bb * 0.75) + (hbp * 0.5) - (so * .1) - e + (ofa * 1) + (a * 0.05)
fps_g = fps / games
df = build_fp_table(players, fps, fps_g)
return df
def get_ws_champs(start_year):
page = bref.get_page('https://www.baseball-reference.com/postseason/')
table = bref.get_table_by_id(page, 'postseason_series')
df = pd.DataFrame(columns=['Year', 'Team'])
years = []
champs = []
rows = table[0].find_all('tr')
for row in rows:
cells = row.find_all(['th', 'td'])
cells = [cell.text.replace('*', '').strip().lower() for cell in cells]
if "world series" in cells[0]:
years.append(cells[0].split()[0])
champs.append(cells[2][:cells[2].index("(")])
df['Year'] = years
df['Year'] = df['Year'].astype(int)
df['Team'] = champs
return df[df['Year'] >= start_year]