Skip to content

Commit b2f48f9

Browse files
ElToro13ElToro13
authored
Create WorldCup2018_Simulation
1 parent 2db462f commit b2f48f9

File tree

1 file changed

+226
-0
lines changed

1 file changed

+226
-0
lines changed

‎FootballPredictor/WorldCup2018/WorldCup2018_Simulation‎

Lines changed: 226 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,226 @@
1+
import urllib2
2+
from bs4 import BeautifulSoup
3+
import requests
4+
import pandas as pd
5+
import random
6+
import math
7+
wct = ['Argentina', 'Australia', 'Belgium', 'Brazil', 'Colombia', 'Costa Rica', 'Croatia', 'Denmark', 'Egypt', 'England', 'France', 'Germany',
8+
'Iceland', 'Iran', 'Japan', 'Mexico', 'Morocco', 'Nigeria', 'Panama', 'Peru', 'Poland', 'Portugal', 'Russia', 'Saudi Arabia', 'Senegal',
9+
'Serbia', 'South Korea', 'Spain', 'Sweden', 'Switzerland', 'Tunisia', 'Uruguay']
10+
11+
12+
df = pd.read_csv('WorldCup2018.csv')
13+
#dates = [str(i)+'/'+str(j) if j>9 else str(i)+'/0'+str(j) for i in range(2016,2019) for j in range(1,13)]
14+
15+
#req = urllib2.Request('http://www.betexplorer.com/soccer/world/friendly-international/mutual-matches/?home=I9l9aqLq&away=f9OppQjp&where=0', None, headers)
16+
def H2H():
17+
link1 = 'http://www.betexplorer.com/soccer/world/friendly-international/mutual-matches/?home='
18+
link2 = '&away='
19+
link3 = '&where=0'
20+
teamID = {'Russia': 'hrgrswHh', 'Saudi Arabia': 'biSY8ox4', 'Egypt':'bejDn7NN', 'Uruguay':'xMk44orG', 'Spain':'bLyo6mco', 'Portugal':'WvJrjFVN', 'Morocco': 'IDKYO3R8',
21+
'Iran': 'xrRx85iA', 'France':'QkGeVG1n', 'Australia':'xSrf6qMM', 'Peru':'Uend67D3','Denmark':'0KUdxQVi','Argentina':'f9OppQjp', 'Nigeria':'EBE2Xb3l','Croatia':'K8aznggo',
22+
'Iceland':'6TsAIrGN','Brazil':'I9l9aqLq','Switzerland':'rHJ2vy1B', 'Costa Rica':'C4ePE2kq', 'Serbia':'8Kl6iq0i', 'Germany':'ptQide1O', 'Mexico':'O6iHcNkd', 'Sweden': 'OQyqbHWB',
23+
'South Korea':'K6Gs7P6G', 'Belgium':'GbB957na', 'Panama':'OWKqbCfi', 'Tunisia': 'QqZVYk95', 'England':'j9N9ZNFA', 'Poland':'2HzmcynI','Senegal':'hOIsJLJr', 'Colombia':'G02s4PCS', 'Japan':'ULXPdOUj'}
24+
25+
headers = { 'User-Agent' : 'Mozilla/5.0' }
26+
req = urllib2.Request('https://www.bbc.co.uk/sport/football/international-friendlies/scores-fixtures/2018-05', None, headers)
27+
page = urllib2.urlopen(req).read()
28+
#page = urllib2.urlopen(link1 + teamID['Germany'] + link2 + teamID['Brazil'])
29+
#page = urllib2.urlopen('https://stackoverflow.com/questions/44854334/no-handlers-could-be-found-for-logger-bs4-dammit')
30+
soup = BeautifulSoup(page, 'html.parser')
31+
scoH = soup.findAll('span',{'class':'sp-c-fixture__number sp-c-fixture__number--home sp-c-fixture__number--ft'})
32+
scoA = soup.findAll('span',{'class':'sp-c-fixture__number sp-c-fixture__number--away sp-c-fixture__number--ft'})
33+
print(len(scoH))
34+
print(len(scoA))
35+
Ht = soup.findAll('span',{'class':'gs-u-display-none gs-u-display-block@m qa-full-team-name sp-c-fixture__team-name-trunc'})
36+
del Ht[0]
37+
del Ht[0]
38+
print(len(Ht))
39+
for i in range(0,len(Ht),2):
40+
try:
41+
team1, team2 = Poisson(Ht[i].text, Ht[i+1].text)
42+
print(scoH[i].text, scoA[i].text)
43+
print(int(team1.index(max(team1))),int(team2.index(max(team2))))
44+
except:
45+
pass
46+
#print(Ht[-8].text.decode('UTF-8'), Ht[-6].text.decode('UTF-8'), Ht[-5].text.decode('UTF-8'),Ht[-4].text.decode('UTF-8'))
47+
## r = (float(Ht[-6].text))/float(Ht[-7].text)
48+
## return 1-r, r
49+
50+
#print(Ht[-16].text.decode('UTF-8'), Ht[-14].text.decode('UTF-8'), Ht[-13].text.decode('UTF-8'),Ht[-12].text.decode('UTF-8'))
51+
## except IndexError:
52+
## #print('This two team have never played against each other')
53+
## return 0.0, 0.0
54+
## except KeyError as e:
55+
## print('Name of {} does not match in our Database'.format(e))
56+
57+
58+
def Cal():
59+
group = {'A':'Russia,Egypt,Saudi Arabia,Uruguay', 'B':'Spain,Portugal,Morocco,Iran','C':'France,Australia,Peru,Denmark', 'D':'Argentina,Iceland,Croatia,Nigeria',
60+
'E':'Brazil,Switzerland,Costa Rica,Serbia','F':'Germany,Mexico,Sweden,South Korea','G':'Belgium,Panama,Tunisia,England', 'H':'Poland,Senegal,Colombia,Japan'}
61+
## group = {'A':'Brazil,Croatia,Cameroon,Mexico', 'B':'Spain,Netherlands,Chile,Australia','C':'Colombia,Greece,Ivory Coast,Japan', 'D':'Uruguay,Costa Rica,England,Italy',
62+
## 'E':'Switzerland,France,Ecuador,Honduras','F':'Argentina,Nigeria,Bosnian-Herzegovina,Iran','G':'Germany,Portugal,Ghana,USA', 'H':'Belgium,Russia,Algeria,South Korea'}
63+
## group = {'A':'South Africa,Mexico,Uruguay,France', 'B':'Argentina,Nigeria,South Korea,Greece','C':'England,USA,Algeria,Slovenia', 'D':'Germany,Australia,Serbia,Ghana',
64+
## 'E':'Netherlands,Denmark,Japan,Cameroon','F':'Italy,Paraguay,New Zealand,Slovakia','G':'Brazil,Portugal,North Korea,Ivory Coast', 'H':'Spain,Switzerland,Honduras,Chile'}
65+
66+
final= {}
67+
for letter in ['A','B','C','D','E','F','G','H']:
68+
names = group[letter].split(',')
69+
points = {}
70+
gd = {}
71+
for n in names:
72+
points[n]=0
73+
gd[n]=0
74+
n=0
75+
for name in names:
76+
n+=1
77+
v = names[n:]
78+
for oppo in v:
79+
team1, team2 = Poisson(name,oppo)
80+
print(int(team1.index(max(team1))),int(team2.index(max(team2))))
81+
gd[name]+= int(team1.index(max(team1))) - int(team2.index(max(team2)))
82+
gd[oppo]+= int(team2.index(max(team2))) - int(team1.index(max(team1)))
83+
if int(team1.index(max(team1))) > int(team2.index(max(team2))):
84+
points[name] +=3
85+
#points[oppo]-=1
86+
elif int(team1.index(max(team1))) < int(team2.index(max(team2))):
87+
points[oppo] +=3
88+
#points[name]-=1
89+
else:
90+
points[oppo] -=float(df[oppo][12])
91+
points[name] -=float(df[name][12])
92+
print points
93+
point = {}
94+
po =[]
95+
for team in names:
96+
point[points[team] + gd[team]] = team
97+
po.append(points[team] + gd[team])
98+
po = sorted(po, reverse=True)
99+
final[letter+'1'] = point[po[0]]
100+
final[letter+'2'] = point[po[1]]
101+
102+
return final
103+
104+
105+
#{'F1': 'Mexico', 'F2': 'Germany', 'G2': 'England', 'G1': 'Belgium', 'H2': 'Senegal',
106+
#'E1': 'Brazil', 'H1': 'Colombia', 'A1': 'Uruguay', 'A2': 'Russia', 'B1': 'Spain',
107+
#'B2': 'Portugal', 'C2': 'Peru', 'C1': 'France', 'E2': 'Serbia', 'D2': 'Croatia', 'D1': 'Argentina'}
108+
def R16(teams={}):
109+
r16 = {}
110+
print(teams)
111+
groupList = [A+B for A in ['A','B','C','D','E','F','G','H'] for B in ['1','2']]
112+
qL = [a+str(b) for a in ['Q'] for b in range(1,9)]
113+
for n in range(0,16,4):
114+
print(n)
115+
team1, team2 = Poisson(teams[groupList[n]], teams[groupList[n+3]])
116+
if team1.index(max(team1)) > team2.index(max(team2)):
117+
r16[qL[n/2]] = teams[groupList[n]]
118+
elif team1.index(max(team1)) < team2.index(max(team2)):
119+
r16[qL[n/2]] = teams[groupList[n+3]]
120+
else:
121+
if float(df[teams[groupList[n]]][12])< float(df[teams[groupList[n+3]]][12]):
122+
r16[qL[n/2]] = teams[groupList[n+3]]
123+
else:
124+
r16[qL[n/2]] = teams[groupList[n]]
125+
126+
team1, team2 = Poisson(teams[groupList[n+1]], teams[groupList[n+2]])
127+
if team1.index(max(team1)) > team2.index(max(team2)):
128+
r16[qL[n/2+1]] = teams[groupList[n+1]]
129+
elif team1.index(max(team1)) < team2.index(max(team2)):
130+
r16[qL[n/2+1]] = teams[groupList[n+2]]
131+
else:
132+
if float(df[teams[groupList[n+1]]][12])< float(df[teams[groupList[n+2]]][12]):
133+
r16[qL[n/2+1]] = teams[groupList[n+2]]
134+
else:
135+
r16[qL[n/2+1]] = teams[groupList[n+1]]
136+
print(r16)
137+
return r16
138+
139+
140+
def R8(team={}):
141+
r8={}
142+
qL = [a+str(b) for a in ['Q'] for b in range(1,9)]
143+
sL = [a+str(b) for a in ['S'] for b in range(1,5)]
144+
for i in range(0,5,4):
145+
team1,team2= Poisson(team[qL[i]], team[qL[i+2]])
146+
if team1.index(max(team1)) > team2.index(max(team2)):
147+
r8[sL[i/2]] = team[qL[i]]
148+
elif team1.index(max(team1)) < team2.index(max(team2)):
149+
r8[sL[i/2]] = team[qL[i+2]]
150+
else:
151+
if float(df[team[qL[i]]][12])< float(df[team[qL[i+2]]][12]):
152+
r8[sL[i/2]] = team[qL[i+2]]
153+
else:
154+
r8[sL[i/2]] = team[qL[i]]
155+
team1,team2= Poisson(team[qL[i+1]], team[qL[i+3]])
156+
if team1.index(max(team1)) > team2.index(max(team2)):
157+
r8[sL[i/2+1]] = team[qL[i+1]]
158+
elif team1.index(max(team1)) < team2.index(max(team2)):
159+
r8[sL[i/2+1]] = team[qL[i+3]]
160+
else:
161+
if float(df[team[qL[i+1]]][12])< float(df[team[qL[i+3]]][12]):
162+
r8[sL[i/2+1]] = team[qL[i+3]]
163+
else:
164+
r8[sL[i/2+1]] = team[qL[i+1]]
165+
print(r8)
166+
return r8
167+
168+
def R4(team={}):
169+
r4={}
170+
i=0
171+
sL = [a+str(b) for a in ['S'] for b in range(1,5)]
172+
fL = [a+str(b) for a in ['F'] for b in range(1,3)]
173+
team1,team2= Poisson(team[sL[i]], team[sL[i+2]])
174+
if team1.index(max(team1)) > team2.index(max(team2)):
175+
r4[fL[i/2]] = team[sL[i]]
176+
elif team1.index(max(team1)) < team2.index(max(team2)):
177+
r4[fL[i/2]] = team[sL[i+2]]
178+
else:
179+
if float(df[team[sL[i]]][12])< float(df[team[sL[i+2]]][12]):
180+
r4[fL[i/2]] = team[sL[i+2]]
181+
else:
182+
r4[fL[i/2]] = team[sL[i]]
183+
team1,team2= Poisson(team[sL[i+1]], team[sL[i+3]])
184+
if team1.index(max(team1)) > team2.index(max(team2)):
185+
r4[fL[i/2+1]] = team[sL[i+1]]
186+
elif team1.index(max(team1)) < team2.index(max(team2)):
187+
r4[fL[i/2+1]] = team[sL[i+3]]
188+
else:
189+
if float(df[team[sL[i+1]]][12])< float(df[team[sL[i+3]]][12]):
190+
r4[fL[i/2+1]] = team[sL[i+3]]
191+
else:
192+
r4[fL[i/2+1]] = team[sL[i+1]]
193+
print(r4)
194+
return r4
195+
196+
def Finals(team={}):
197+
final={}
198+
team1,team2= Poisson(team['F1'], team['F2'])
199+
if team1.index(max(team1)) > team2.index(max(team2)):
200+
final['WinningTeam'] = team['F1']
201+
elif team1.index(max(team1)) < team2.index(max(team2)):
202+
final['WinningTeam'] = team['F2']
203+
else:
204+
if float(df[team['F1']][12])< float(df[team['F2']][12]):
205+
final['WinningTeam'] = team['F2']
206+
else:
207+
final['WinningTeam'] = team['F1']
208+
print(final)
209+
return('Made by: Yash Soni')
210+
211+
212+
def Poisson(HT='',AT=''):
213+
print(HT,AT)
214+
HTS = 0
215+
ATS = 0
216+
#print h2h_H, h2h_A
217+
ATS = ((float(df[AT][12])+(float(df[AT][13])))*(float(df[AT][2])/(float(df[AT][0]))))+random.uniform(0.1,float(df[AT][13]))
218+
HTS = ((float(df[HT][12])+(float(df[HT][13])))*(float(df[HT][2])/(float(df[HT][0]))))+random.uniform(0.1,float(df[HT][13]))
219+
220+
221+
print(HTS,ATS)
222+
223+
return [(((math.exp(-HTS))*(math.pow(HTS,i))))/(math.factorial(i)) for i in range(0,5)], [(((math.exp(-ATS))*(math.pow(ATS,i))))/(math.factorial(i)) for i in range(0,5)]
224+
225+
226+
print(Finals(R4(R8(R16(Cal())))))

0 commit comments

Comments
 (0)