Create a gist now

Instantly share code, notes, and snippets.

What would you like to do?
Generate a full season's worth of batting Marcel projections from past years' stats
## Generate a full season's worth of batting Marcel projections from past years' stats
from createTuple import createTuple ## gist: 778481
from writeMatrixCSV import writeMatrixCSV ## gist: 778484
def makeBatTable(r):
for stat in ['AB', 'H', 'D', 'T', 'HR', 'SO', 'BB', 'SF', 'HP', 'CI']:
if stat in r: pass
else: r[stat] = 0
if r['AB'] == 0:
r['SLG'] = 0
r['AVG'] = 0
slg = 0
else:
avg = float(r['H'])/float(r['AB'])
r['AVG'] = round(avg, 3)
slg = float(r['H']+r['D']+(2*r['T'])+(3*r['HR']))/float(r['AB'])
r['SLG'] = round(slg, 3)
if (r['AB']+r['BB']+r['SF']+r['HP']) == 0:
r['OBP'] = 0
r['OPS'] = 0
r['wOBA'] = 0
else:
pa = float(r['AB']+r['BB']+r['SF']+r['HP']+r['CI'])
obp = float(r['H']+r['BB']+r['HP']+r['CI'])/pa
r['OBP'] = round(obp, 3)
sing = int(r['H']) - int(r['HR']) - int(r['T']) - int(r['D'])
num = (.72*int(r['BB'])) + (.75*int(r['HP'])) + (.9*sing) + (1.24*int(r['D'])) + (1.56*int(r['T'])) + (1.95*int(r['HR']))
den = int(r['BB']) + int(r['AB']) + int(r['HP'])
woba = num/den
r['wOBA'] = round(woba, 3)
return r
def marcelBattingSeason(yr):
# yr = year being projected, input as int
yr = str(yr)
yr1 = str(int(yr) - 1)
yr2 = str(int(yr) - 2)
yr3 = str(int(yr) - 3)
## get list of pitchers; determine which batters are really
## 'batters' and throw out pitchers with at-bats
projectBatters = []
for yr in [yr3, yr2, yr]:
yearPitchers = {}
for p in pitchers:
pitchID = p[0]
if p[1] == yr:
yearPitchers[pitchID] = int(p[12])
else: pass
for b in batters:
batID = b[0]
if b[1] == yr: pass
else: continue
abString = b[7]
if abString == '': continue
else: batAb = int(abString)
if batID in yearPitchers:
if yearPitchers[batID] > batAb: continue
else: pass
else: pass
if batID in projectBatters: pass
else: projectBatters.append(batID)
## find league average for previous year
yearPitchers = {}
for p in pitchers:
pitchID = p[0]
if p[1] == yr1:
yearPitchers[pitchID] = int(p[12])
else: pass
leagueAverage = {}
for b in batters:
batID = b[0]
if b[1] == yr1: pass
else: continue
abString = b[7]
if abString == '': continue
else: batAb = int(abString)
if batID in yearPitchers:
if yearPitchers[batID] > batAb: continue
else: pass
else: pass
if batID in projectBatters: pass
else: projectBatters.append(batID)
for stat in batHeaders:
col = batHeaders[stat]
try: playerStat = int(b[col])
except: continue
else: pass
if stat in leagueAverage:
leagueAverage[stat] += playerStat
else:
leagueAverage[stat] = playerStat
for stat in ['HP', 'SF', 'SH']:
if stat in leagueAverage: pass
else: leagueAverage[stat] = 0
totalPa = leagueAverage['AB'] + leagueAverage['BB'] + leagueAverage['HP'] + leagueAverage['SF'] + leagueAverage['SH']
regression = {}
for stat in leagueAverage:
regression[stat] =(1200.0/totalPa)*leagueAverage[stat]
rawProjections = {}
## calculate projections for each player
for b in projectBatters:
components = {}
y2pa = 0
y1pa = 0
for stat in batHeaders:
components[stat] = 0
for row in batters:
if row[0] == b: pass
else: continue
if row[1] == yr3:
for stat in batHeaders:
try: playerStat = int(row[batHeaders[stat]])
except: continue
components[stat] += 3*playerStat
elif row[1] == yr2:
for stat in batHeaders:
try: playerStat = int(row[batHeaders[stat]])
except: continue
components[stat] += 4*playerStat
for stat in ['AB', 'BB', 'HP', 'SF', 'SH']:
try: y2pa += int(row[batHeaders[stat]])
except: continue
elif row[1] == yr1:
for stat in batHeaders:
try: playerStat = int(row[batHeaders[stat]])
except: continue
components[stat] += 5*playerStat
for stat in ['AB', 'BB', 'HP', 'SF', 'SH']:
try: y1pa += int(row[batHeaders[stat]])
except: continue
else: continue
## add regression component
for stat in regression:
components[stat] += regression[stat]
## get projected PA
projPa = 0.5*y1pa + 0.1*y2pa + 200
## prorate into projected PA
compPa = components['AB'] + components['BB'] + components['HP'] + components['SF'] + components['SH']
prorateProj = {}
for stat in components:
prorateProj[stat] = (projPa/compPa)*components[stat]
prorateProj['PA'] = projPa
try: age = int(yr) - int(birthYear[b])
except: age = 29 ## in case birthyear is missing or corrupted
## age adjust
if age > 29:
ageAdj = 1/(1 + ((age - 29)*0.003))
elif age < 29:
ageAdj = 1 + ((29 - age)*0.006)
else:
ageAdj = 1
finalProj = {}
for stat in prorateProj:
if stat in ['PA', 'AB']:
finalProj[stat] = prorateProj[stat]
elif stat in ['R', 'H', 'D', 'T', 'HR', 'RBI', 'SB', 'BB', 'IBB', 'HP', 'SH', 'SF']:
finalProj[stat] = prorateProj[stat]*ageAdj
else:
finalProj[stat] = prorateProj[stat]/ageAdj
## reliability
reliab = 1 - (1200.0/compPa)
finalProj['rel'] = round(reliab, 2)
finalProj['Age'] = age
## add to master dict
rawProjections[b] = finalProj
## re-baseline
projTotal = {}
for pl in rawProjections:
for stat in rawProjections[pl]:
if stat in projTotal:
projTotal[stat] += rawProjections[pl][stat]
else:
projTotal[stat] = rawProjections[pl][stat]
projTotalPa = projTotal['PA']
projRatios = {}
for stat in ['AB', 'R', 'H', 'D', 'T', 'HR', 'RBI', 'SB', 'CS', 'BB', 'SO', 'IBB', 'HP', 'SH', 'SF', 'GDP']:
projRatios[stat] = projTotal[stat]/projTotalPa
trueRatios = {}
for stat in ['AB', 'R', 'H', 'D', 'T', 'HR', 'RBI', 'SB', 'CS', 'BB', 'SO', 'IBB', 'HP', 'SH', 'SF', 'GDP']:
try: trueRatios[stat] = leagueAverage[stat]/float(totalPa)
except: trueRatios[stat] = 0
marcels = {}
for pl in rawProjections:
marcels[pl] = {}
for stat in rawProjections[pl]:
if stat in ['AB', 'R', 'H', 'D', 'T', 'HR', 'RBI', 'SB', 'CS', 'BB', 'SO', 'IBB', 'HP', 'SH', 'SF', 'GDP']:
if projRatios[stat] == 0:
marcels[pl][stat] = rawProjections[pl][stat]
else:
marcels[pl][stat] = round((trueRatios[stat]/projRatios[stat])*rawProjections[pl][stat], 0)
elif stat == 'PA':
marcels[pl][stat] = round(rawProjections[pl][stat], 0)
else:
marcels[pl][stat] = rawProjections[pl][stat]
header = ['bdbID', 'First', 'Last', 'Year', 'age', 'rel', 'wOBA', 'AVG', 'OBP', 'SLG',
'PA', 'AB', 'R', 'H', 'D', 'T', 'HR', 'RBI',
'SB', 'CS', 'BB', 'SO', 'IBB', 'HP', 'SH', 'SF', 'GDP']
marcelSheet = [header]
for pl in marcels:
row = [pl]
row += firstlast[pl]
row.append(yr)
marcels[pl] = makeBatTable(marcels[pl])
for stat in ['Age', 'rel', 'wOBA', 'AVG', 'OBP', 'SLG',
'PA', 'AB', 'R', 'H', 'D', 'T', 'HR', 'RBI',
'SB', 'CS', 'BB', 'SO', 'IBB', 'HP', 'SH', 'SF', 'GDP']:
row.append(marcels[pl][stat])
marcelSheet.append(row)
filename = 'marcel_batters_' + yr + '.csv'
writeMatrixCSV(marcelSheet, filename)
batHeaders = {'AB': 7,
'R': 8,
'H': 9,
'D': 10,
'T': 11,
'HR': 12,
'RBI': 13,
'SB': 14,
'CS': 15,
'BB': 16,
'SO': 17,
'IBB': 18,
'HP': 19,
'SH': 20,
'SF': 21,
'GDP': 22
}
pitchers = createTuple('bdb_pitching.csv')
## this is the pitcher seasons sheet from the lahman db. headers:
## playerID,yearID,stint,teamID,lgID,W,L,G,GS,CG,SHO,SV,Ipouts,H,ER,HR,BB,SO,Baopp,ERA,IBB,WP,HP,BK,BFP,GF,R
batters = createTuple('bdb_batting.csv')
## this is the batting seasons sheet from the lahman db. headers:
## playerID,yearID,stint,teamID,lgID,G,G_batting,AB,R,H,D,T,HR,RBI,SB,CS,BB,SO,IBB,HP,SH,SF,GIDP,G_old
## master db for birthYear
master = createTuple('bdb_master.csv')
## master biographical data sheet from lahman db. headers:
## lahmanID,playerID,managerID,hofID,birthYear,birthMonth,birthDay,birthCountry,birthState,birthCity,deathYear,deathMonth,deathDay,deathCountry,deathState,deathCity,nameFirst,nameLast,nameNote,nameGiven,nameNick,weight,height,bats,throws,debut,finalGame,college,lahman40ID,lahman
birthYear = {}
for pl in master:
birthYear[pl[1]] = pl[4]
firstlast = {}
for pl in master:
firstlast[pl[1]] = [pl[16], pl[17]]
## sample usage
marcelBattingSeason(2005)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment