Perform ‘Exploratory Data Analysis’ on dataset ‘Indian Premier League’
● As a sports analysts, find out the most successful teams, players and factors contributing win or loss of a team.
● Suggest teams or players a company should endorse for its products.
● You can choose any of the tool of your choice (Python/R/Tableau/PowerBI/Excel/SAP/SAS)
In [79]:
import pandas as pd
import numpy as np
import seaborn as sn
import plotly.express as px
import plotly.graph_objects as go
import plotly.offline as py
from plotly.subplots import make_subplots
from plotly import tools
import matplotlib.pyplot as plt
%matplotlib inline
In [2]:
deliveries = pd.read_csv('deliveries.csv')
matches = pd.read_csv('matches.csv')
In [3]:
pd.set_option('display.max_columns',None)
In [4]:
deliveries.head(3)
Out[4]:
In [5]:
deliveries.shape,matches.shape
Out[5]:
In [6]:
matches.head(3)
Out[6]:
In [7]:
matches.columns
Out[7]:
In [8]:
temp_match = matches[['season','team1', 'team2','toss_winner','toss_decision','winner', 'win_by_runs',
'win_by_wickets', 'player_of_match', 'venue']]
temp_match.head()
Out[8]:
In [9]:
temp_match['season'].unique()
Out[9]:
In [10]:
temp_match = temp_match.sort_values('season',ascending=True).reset_index()
temp_match = temp_match.drop(['index'],axis = 1)
temp_match.head()
Out[10]:
In [11]:
matches['season'].unique()
Out[11]:
In [12]:
df_2008 = temp_match[(temp_match['season']==2008)]
df_2009 = temp_match[(temp_match['season']==2009)]
df_2010 = temp_match[(temp_match['season']==2010)]
df_2011 = temp_match[(temp_match['season']==2011)]
df_2012 = temp_match[(temp_match['season']==2012)]
df_2013 = temp_match[(temp_match['season']==2013)]
df_2014 = temp_match[(temp_match['season']==2014)]
df_2015 = temp_match[(temp_match['season']==2015)]
df_2016 = temp_match[(temp_match['season']==2016)]
df_2017 = temp_match[(temp_match['season']==2017)]
df_2018 = temp_match[(temp_match['season']==2018)]
In [13]:
df_list = [df_2008,df_2009,df_2010,df_2011,df_2012,df_2013,df_2014,df_2015,df_2016,df_2017,df_2018]
In [14]:
count = 2008
for i in range(len(df_list)):
print(f' Dataframe : df_{str(count)} has a shape of {df_list[i].shape}')
count+=1
In [15]:
df_2008.columns
Out[15]:
Teams which Won the toss and also the Match¶
In [16]:
def returndf(df):
temp = df[(df['toss_winner'])==(df['winner'])]
return temp
In [17]:
df_2008_toss = returndf(df_2008)
df_2009_toss = returndf(df_2009)
df_2010_toss = returndf(df_2010)
df_2011_toss = returndf(df_2011)
df_2012_toss = returndf(df_2012)
df_2013_toss = returndf(df_2013)
df_2014_toss = returndf(df_2014)
df_2015_toss = returndf(df_2015)
df_2016_toss = returndf(df_2016)
df_2017_toss = returndf(df_2017)
df_2018_toss = returndf(df_2018)
In [18]:
df_2008_toss.head()
Out[18]:
In [19]:
toss_list = [df_2008_toss,df_2009_toss,df_2010_toss,df_2011_toss,df_2012_toss,df_2013_toss,df_2014_toss,df_2015_toss,
df_2016_toss,df_2017_toss,df_2018_toss]
shape_list = []
count = 2008
for i in range(len(toss_list)):
print(f' Dataframe df_{str(count)}_toss has a shape of : {toss_list[i].shape}')
shape_list.append(toss_list[i].shape[0])
count+=1
In [20]:
shape_list,len(shape_list)
Out[20]:
In [21]:
temp = pd.DataFrame()
temp['Season'] = [i for i in range(2008,2019)]
temp['Wins'] = np.array(shape_list)
temp.head()
Out[21]:
In [22]:
sn.barplot(data=temp,x = 'Season',y = 'Wins',palette='plasma')
plt.title('Teams which won the Toss and Match as well')
plt.xlabel('Season')
plt.ylabel('Wins')
Out[22]:
In [23]:
px.bar(data_frame=temp,x = 'Season',y = 'Wins',color = 'Wins',hover_name='Season',title='Toss Won Match Won :)')
In [24]:
df_2008.head()
Out[24]:
In [25]:
def returndict(df):
values = df['winner'].values
values = list(values)
df_val = dict(zip(values,map(values.count,values)))
key_list = []
val_list = []
for key,val in df_val.items():
key_list.append(key)
val_list.append(val)
df = pd.DataFrame()
df['Team'] = key_list
df['Wins'] = val_list
return df
win_2008 = returndict(df_2008)
win_2009 = returndict(df_2009)
win_2010 = returndict(df_2010)
win_2011 = returndict(df_2011)
win_2012 = returndict(df_2012)
win_2013 = returndict(df_2013)
win_2014 = returndict(df_2014)
win_2015 = returndict(df_2015)
win_2016 = returndict(df_2016)
win_2017 = returndict(df_2017)
win_2018 = returndict(df_2018)
In [26]:
fig = make_subplots(rows=4,cols=3,subplot_titles=("Successful Teams of Season 2008",
"Successful Teams of Season 2009",
"Successful Teams of Season 2010",
"Successful Teams of Season 2011",
"Successful Teams of Season 2012",
"Successful Teams of Season 2013",
"Successful Teams of Season 2014",
"Successful Teams of Season 2015",
"Successful Teams of Season 2016",
"Successful Teams of Season 2017",
"Successful Teams of Season 2018"),shared_yaxes=True)
fig.add_trace(go.Bar(x = win_2008['Team'],y = win_2008['Wins'],marker = dict(color = win_2008['Wins'],coloraxis = "coloraxis")),row = 1,col = 1)
fig.add_trace(go.Bar(x = win_2009['Team'],y = win_2009['Wins'],marker = dict(color = win_2009['Wins'],coloraxis = "coloraxis")),row = 1,col = 2)
fig.add_trace(go.Bar(x = win_2010['Team'],y = win_2010['Wins'],marker = dict(color = win_2010['Wins'],coloraxis = "coloraxis")),row = 1,col = 3)
fig.add_trace(go.Bar(x = win_2011['Team'],y = win_2011['Wins'],marker = dict(color = win_2011['Wins'],coloraxis = "coloraxis")),row = 2,col = 1)
fig.add_trace(go.Bar(x = win_2012['Team'],y = win_2012['Wins'],marker = dict(color = win_2012['Wins'],coloraxis = "coloraxis")),row = 2,col = 2)
fig.add_trace(go.Bar(x = win_2013['Team'],y = win_2013['Wins'],marker = dict(color = win_2013['Wins'],coloraxis = "coloraxis")),row = 2,col = 3)
fig.add_trace(go.Bar(x = win_2014['Team'],y = win_2014['Wins'],marker = dict(color = win_2014['Wins'],coloraxis = "coloraxis")),row = 3,col = 1)
fig.add_trace(go.Bar(x = win_2015['Team'],y = win_2015['Wins'],marker = dict(color = win_2015['Wins'],coloraxis = "coloraxis")),row = 3,col = 2)
fig.add_trace(go.Bar(x = win_2016['Team'],y = win_2016['Wins'],marker = dict(color = win_2016['Wins'],coloraxis = "coloraxis")),row = 3,col = 3)
fig.add_trace(go.Bar(x = win_2017['Team'],y = win_2017['Wins'],marker = dict(color = win_2017['Wins'],coloraxis = "coloraxis")),row = 4,col = 1)
fig.add_trace(go.Bar(x = win_2018['Team'],y = win_2018['Wins'],marker = dict(color = win_2018['Wins'],coloraxis = "coloraxis")),row = 4,col = 2)
fig.update_layout(height =1500,width = 1400,title_text = "Successful Teams in IPL from season 2008-2018")
fig.show()
In [27]:
win_2008_3 = win_2008.sort_values('Wins',ascending=False).head(3)
win_2009_3 = win_2009.sort_values('Wins',ascending=False).head(3)
win_2010_3 = win_2010.sort_values('Wins',ascending=False).head(3)
win_2011_3 = win_2011.sort_values('Wins',ascending=False).head(3)
win_2012_3 = win_2012.sort_values('Wins',ascending=False).head(3)
win_2013_3 = win_2013.sort_values('Wins',ascending=False).head(3)
win_2014_3 = win_2014.sort_values('Wins',ascending=False).head(3)
win_2015_3 = win_2015.sort_values('Wins',ascending=False).head(3)
win_2016_3 = win_2016.sort_values('Wins',ascending=False).head(3)
win_2017_3 = win_2017.sort_values('Wins',ascending=False).head(3)
win_2018_3 = win_2018.sort_values('Wins',ascending=False).head(3)
In [28]:
fig = make_subplots(rows=6,cols=2,subplot_titles=("Successful Teams of Season 2008",
"Successful Teams of Season 2009",
"Successful Teams of Season 2010",
"Successful Teams of Season 2011",
"Successful Teams of Season 2012",
"Successful Teams of Season 2013",
"Successful Teams of Season 2014",
"Successful Teams of Season 2015",
"Successful Teams of Season 2016",
"Successful Teams of Season 2017",
"Successful Teams of Season 2018"),shared_yaxes=True)
fig.add_trace(go.Bar(x = win_2008_3['Team'],y = win_2008_3['Wins'],marker = dict(color = win_2008_3['Wins'],coloraxis = "coloraxis")),row = 1,col = 1)
fig.add_trace(go.Bar(x = win_2009_3['Team'],y = win_2009_3['Wins'],marker = dict(color = win_2009_3['Wins'],coloraxis = "coloraxis")),row = 1,col = 2)
fig.add_trace(go.Bar(x = win_2010_3['Team'],y = win_2010_3['Wins'],marker = dict(color = win_2010_3['Wins'],coloraxis = "coloraxis")),row = 2,col = 1)
fig.add_trace(go.Bar(x = win_2011_3['Team'],y = win_2011_3['Wins'],marker = dict(color = win_2011_3['Wins'],coloraxis = "coloraxis")),row = 2,col = 2)
fig.add_trace(go.Bar(x = win_2012_3['Team'],y = win_2012_3['Wins'],marker = dict(color = win_2012_3['Wins'],coloraxis = "coloraxis")),row = 3,col = 1)
fig.add_trace(go.Bar(x = win_2013_3['Team'],y = win_2013_3['Wins'],marker = dict(color = win_2013_3['Wins'],coloraxis = "coloraxis")),row = 3,col = 2)
fig.add_trace(go.Bar(x = win_2014_3['Team'],y = win_2014_3['Wins'],marker = dict(color = win_2014_3['Wins'],coloraxis = "coloraxis")),row = 4,col = 1)
fig.add_trace(go.Bar(x = win_2015_3['Team'],y = win_2015_3['Wins'],marker = dict(color = win_2015_3['Wins'],coloraxis = "coloraxis")),row = 4,col = 2)
fig.add_trace(go.Bar(x = win_2016_3['Team'],y = win_2016_3['Wins'],marker = dict(color = win_2016_3['Wins'],coloraxis = "coloraxis")),row = 5,col = 1)
fig.add_trace(go.Bar(x = win_2017_3['Team'],y = win_2017_3['Wins'],marker = dict(color = win_2017_3['Wins'],coloraxis = "coloraxis")),row = 5,col = 2)
fig.add_trace(go.Bar(x = win_2018_3['Team'],y = win_2018_3['Wins'],marker = dict(color = win_2018_3['Wins'],coloraxis = "coloraxis")),row = 6,col = 1)
fig.update_layout(height =1500,width = 900,title_text = "Successful Teams in IPL from season 2008-2018")
fig.show()
For the season 2008,Rajasthan Royals,Kings XI Punjab,Chennai SuperKings,were the top 3 Performing Teams
For the season 2009,Delhi DareDevils,Deccan Chargers,Royal Challengers Banglore,were the top 3 Performing Teams
For the season 2010,Mumbai Indians,Chennai SuperKings,Royal Challengers Banglore,were the top 3 Performing Teams
For the season 2011,Mumbai Indians,Chennai SuperKings,Royal Challengers Banglore,were the top 3 Performing Teams
For the season 2012,Kolkata Knight Riders,Delhi DareDevils,Mumbai Indians,were the top 3 Performing Teams
For the season 2013,Mumbai Indians,Chennai SuperKings,Rajasthan Royals,were the top 3 Performing Teams
For the season 2014,Kings XI Punjab,Kolkata Knight Riders,Chennai SuperKings,were the top 3 Performing Teams
For the season 2015,Chennai Super Kings,Mumbai Indians,Royal Challengers Banglore,were the top 3 Performing Teams
For the season 2016,Surises Hyderabad,Gujarat Lions,Royal Challengers Banglore,were the top 3 Performing Teams
For the season 2017,Mumbai Indians,Rising Pune Supergaints,Kolkata Knight Riders,were the top 3 Performing Teams
For the season 2018,Chennai SuperKings,Sunrises Hyderabad,Kolkata Knight Riders,were the top 3 Performing Teams
We can observe the trend from the following Graph,we can observe that Mumbai Indians,Chennai Super Kings,Royal Challengers Banglore,Kolkata Knight Riders were the most Performing Teams/Successful Teams out of all teams,as these teams were in the top-3 positions in the Points Table,hence these teams have participated most number of times in the World Champions T20 League from India,So these teams can be a selection choice if a company wants to endorse its products!,as these are the most loved teams in the Indian Premier League almost for all seasons
Top Performing Batsmans¶
In [29]:
deliveries.head()
Out[29]:
In [30]:
a = list(deliveries['batting_team'].unique())
a
Out[30]:
In [31]:
df_csk = deliveries[(deliveries['batting_team']=='Chennai Super Kings')&(deliveries['total_runs']!=0)&(deliveries['total_runs']!=5)&(deliveries['total_runs']!=7)]
df_mi = deliveries[(deliveries['batting_team']=='Mumbai Indians')&(deliveries['total_runs']!=0)&(deliveries['total_runs']!=5)&(deliveries['total_runs']!=7)]
df_rcb = deliveries[(deliveries['batting_team']=='Royal Challengers Bangalore')&(deliveries['total_runs']!=0)&(deliveries['total_runs']!=5)&(deliveries['total_runs']!=7)]
df_kkr = deliveries[(deliveries['batting_team']=='Kolkata Knight Riders')&(deliveries['total_runs']!=0)&(deliveries['total_runs']!=5)&(deliveries['total_runs']!=7)]
df_rr = deliveries[(deliveries['batting_team']=='Rajasthan Royals')&(deliveries['total_runs']!=0)&(deliveries['total_runs']!=5)&(deliveries['total_runs']!=7)]
df_kxip = deliveries[(deliveries['batting_team']=='Kings XI Punjab')&(deliveries['total_runs']!=0)&(deliveries['total_runs']!=5)&(deliveries['total_runs']!=7)]
df_dd = deliveries[(deliveries['batting_team']=='Delhi Daredevils') | (deliveries['batting_team']=='Delhi Capitals')&(deliveries['total_runs']!=0)&(deliveries['total_runs']!=5)&(deliveries['total_runs']!=7)]
df_srh = deliveries[(deliveries['batting_team']=='Sunrisers Hyderabad')|(deliveries['batting_team']=='Deccan Chargers')&(deliveries['total_runs']!=0)&(deliveries['total_runs']!=5)&(deliveries['total_runs']!=7)]
In [ ]:
In [32]:
df_csk.head()
Out[32]:
In [33]:
csk_bat = list(df_csk['batsman'].unique())
mi_bat = list(df_mi['batsman'].unique())
rcb_bat = list(df_rcb['batsman'].unique())
kkr_bat = list(df_kkr['batsman'].unique())
rr_bat = list(df_rr['batsman'].unique())
kxip_bat = list(df_kxip['batsman'].unique())
dd_bat = list(df_dd['batsman'].unique())
srh_bat = list(df_srh['batsman'].unique())
In [34]:
def getmap(df,player):
temp = df.groupby(df['batsman']==player)[['total_runs']].sum()
runs = temp['total_runs'][1]
return runs
In [35]:
df_list = [df_csk,df_mi,df_rcb,df_kkr,df_rr,df_kxip,df_dd,df_srh]
bat_list = [csk_bat,mi_bat,rcb_bat,kkr_bat,rr_bat,kxip_bat,dd_bat,srh_bat]
teams = ['csk','mi','rcb','kkr','rr','kxip','dd','srh']
year = [i for i in range(2008,2019)]
count = 2008
main_map = {}
for team,batsman,team_name in zip(df_list,bat_list,teams):
mapper = {}
for player in batsman:
mapper[player] = getmap(team,player)
main_map[team_name] = mapper
count+=1
In [36]:
def gettopbatdf(team):
temp = main_map[team]
temp = dict(sorted(temp.items(),key=lambda x:x[1],reverse=True))
temp = dict(list(temp.items())[:10])
top_bat_df = pd.DataFrame()
name = []
runs = []
for key,val in temp.items():
name.append(key)
runs.append(val)
top_bat_df['Player'] = name
top_bat_df['Runs'] = runs
return top_bat_df
In [37]:
teams = ['csk','mi','rcb','kkr','rr','kxip','dd','srh']
In [38]:
top_bat_csk = gettopbatdf('csk')
top_bat_mi = gettopbatdf('mi')
top_bat_rcb = gettopbatdf('rcb')
top_bat_kkr = gettopbatdf('kkr')
top_bat_rr = gettopbatdf('rr')
top_bat_kxip = gettopbatdf('kxip')
top_bat_dd = gettopbatdf('dd')
top_bat_srh = gettopbatdf('srh')
In [39]:
top_bat_csk
Out[39]:
In [40]:
teams = ['csk','mi','rcb','kkr','rr','kxip','dd','srh']
fig = make_subplots(rows=4,cols=2,subplot_titles=('CSK','MI','RCB','KKR',
'RR','KXIP','DD','SRH'),start_cell = 'top-left',
specs=[[{'type':'polar'},{'type':'polar'}],
[{'type':'polar'},{'type':'polar'}],
[{'type':'polar'},{'type':'polar'}],
[{'type':'polar'},{'type':'polar'}]],shared_yaxes=True)
# fig = make_subplots(rows=4,cols=2,specs=[[{'type':'polar'},{'type':'polar'}],
# [{'type':'polar'},{'type':'polar'}],
# [{'type':'polar'},{'type':'polar'}],
# [{'type':'polar'},{'type':'polar'}]])
fig.add_trace(go.Barpolar(theta=list(top_bat_csk['Player'].unique()),r = list(top_bat_csk['Runs'].unique())),1,1)
fig.add_trace(go.Barpolar(theta=list(top_bat_mi['Player'].unique()),r = list(top_bat_mi['Runs'].unique())),1,2)
fig.add_trace(go.Barpolar(theta=list(top_bat_rcb['Player'].unique()),r = list(top_bat_rcb['Runs'].unique())),2,1)
fig.add_trace(go.Barpolar(theta=list(top_bat_kkr['Player'].unique()),r = list(top_bat_kkr['Runs'].unique())),2,2)
fig.add_trace(go.Barpolar(theta=list(top_bat_rr['Player'].unique()),r = list(top_bat_rr['Runs'].unique())),3,1)
fig.add_trace(go.Barpolar(theta=list(top_bat_kxip['Player'].unique()),r = list(top_bat_kxip['Runs'].unique())),3,2)
fig.add_trace(go.Barpolar(theta=list(top_bat_dd['Player'].unique()),r = list(top_bat_dd['Runs'].unique())),4,1)
fig.add_trace(go.Barpolar(theta=list(top_bat_srh['Player'].unique()),r = list(top_bat_srh['Runs'].unique())),4,2)
# fig.update_traces(hoverinfo="label+percent+name")
# fig.update_layout(
# title_text="Global Emissions 1990-2011",
# # Add annotations in the center of the donut pies.
# annotations=[dict(text='GHG', x=0.18, y=0.5, font_size=20, showarrow=False),
# dict(text='CO2', x=0.82, y=0.5, font_size=20, showarrow=False),
# dict(text='CO2', x=0.82, y=0.5, font_size=20, showarrow=False),
# dict(text='CO2', x=0.82, y=0.5, font_size=20, showarrow=False),
# dict(text='CO2', x=0.82, y=0.5, font_size=20, showarrow=False),
# dict(text='CO2', x=0.82, y=0.5, font_size=20, showarrow=False),
# dict(text='CO2', x=0.82, y=0.5, font_size=20, showarrow=False),
# dict(text='CO2', x=0.82, y=0.5, font_size=20, showarrow=False)])
fig.update_layout(height =1600,width = 1000,title_text = "Top Performing Batsmans From Each team")
# annot =list(fig.layout.annotations)
# fig.layout.annotations = annot
fig.show()
Top 5 Batsmans Stats¶
In [41]:
def gettopbatdf(team):
temp = main_map[team]
temp = dict(sorted(temp.items(),key=lambda x:x[1],reverse=True))
temp = dict(list(temp.items())[:5])
keys = []
for key,val in temp.items():
keys.append(key)
return keys
In [42]:
bat_csk = gettopbatdf('csk')
bat_mi = gettopbatdf('mi')
bat_rcb = gettopbatdf('rcb')
bat_kkr = gettopbatdf('kkr')
bat_rr = gettopbatdf('rr')
bat_kxip = gettopbatdf('kxip')
bat_dd = gettopbatdf('dd')
bat_srh = gettopbatdf('srh')
In [43]:
bat_csk
Out[43]:
In [44]:
def getruns(df,player):
temp = df[(df['batsman']==player)]
runs = list(temp['total_runs'])
mapper = dict(zip(runs,map(runs.count,runs)))
return mapper
In [45]:
df_list = [df_csk,df_mi,df_rcb,df_kkr,df_rr,df_kxip,df_dd,df_srh]
bats = [bat_csk,bat_mi,bat_rcb,bat_kkr,bat_rr,bat_kxip,bat_dd,bat_srh]
teams = ['csk','mi','rcb','kkr','rr','kxip','dd','srh']
final_mapper = {}
for i,j,k in zip(df_list,bats,teams):
mapper = {}
for player in j:
mapper[player] = getruns(i,player)
final_mapper[k] = mapper
In [46]:
def gettop5dataframe(team):
top_5 = dict(list(final_mapper[team].items()))
dummy = pd.DataFrame(top_5)
dummy = dummy.T
dummy = dummy.fillna(value = 0)
for i in dummy.columns:
if i==5 or i==8 or i==10 or i==7:
dummy = dummy.drop([i],axis = 1)
dummy = dummy.astype('int32')
dummy['Player'] = dummy.index
dummy = dummy.reset_index()
dummy = dummy[['Player',1,2,3,4,6]]
return dummy
In [47]:
teams = ['csk','mi','rcb','kkr','rr','kxip','dd','srh']
In [48]:
csk_ind = gettop5dataframe('csk')
mi_ind = gettop5dataframe('mi')
rcb_ind = gettop5dataframe('rcb')
kkr_ind = gettop5dataframe('kkr')
rr_ind = gettop5dataframe('rr')
kxip_ind = gettop5dataframe('kxip')
dd_ind = gettop5dataframe('dd')
srh_ind = gettop5dataframe('srh')
In [49]:
csk_ind
Out[49]:
In [50]:
def getnameandruns(df_i):
final = []
for i in range(df_i.shape[0]):
first = df_i.iloc[[i]]
temp = []
for i in first.columns:
a = []
a.append(list(first[i]))
a = sum(a,[])
temp.append(a)
final.append(temp)
final = sum(final,[])
final = sum(final,[])
name = [i for i in final if type(i)!=int]
runs = [i for i in final if type(i)==int]
final_runs = []
start = 0
end = 5
while end<=len(runs):
temp = runs[start:end]
final_runs.append(temp)
start = end
end+=5
return name,final_runs
In [51]:
teams = ['csk','mi','rcb','kkr','rr','kxip','dd','srh']
In [52]:
csk_name,csk_runs = getnameandruns(csk_ind)
mi_name,mi_runs = getnameandruns(mi_ind)
rcb_name,rcb_runs = getnameandruns(rcb_ind)
kkr_name,kkr_runs = getnameandruns(kkr_ind)
rr_name,rr_runs = getnameandruns(rr_ind)
kxip_name,kxip_runs = getnameandruns(kxip_ind)
dd_name,dd_runs = getnameandruns(dd_ind)
srh_name,srh_runs = getnameandruns(srh_ind)
In [53]:
srh_runs
Out[53]:
In [54]:
list(top_bat_csk['Runs'].unique())[:5]
Out[54]:
In [55]:
names = [csk_name,mi_name,rcb_name,kkr_name,rcb_name,kxip_name,dd_name,srh_name]
final = []
for i in range(len(names)):
final.append(names[i])
final
Out[55]:
In [56]:
def drawplots(player_runs,player_names,player_name,top_bat,text):
runs = [1,2,3,4,6]
runs = [str(i) for i in runs]
fig = make_subplots(rows=2,cols = 3,subplot_titles=tuple(player_names),
specs=[[{'type':'domain'},{'type':'domain'},{'type':'domain'}],
[{'type':'domain'},{'type':'domain'},{'type':'domain'}]],shared_yaxes=True)
fig.add_trace(go.Pie(labels=runs,values=player_runs[0],name = player_names[0]),1,1)
fig.add_trace(go.Pie(labels=runs,values=player_runs[1],name = player_names[1]),1,2)
fig.add_trace(go.Pie(labels=runs,values=player_runs[2],name = player_names[2]),1,3)
fig.add_trace(go.Pie(labels=runs,values=player_runs[3],name = player_names[3]),2,1)
fig.add_trace(go.Pie(labels=runs,values=player_runs[4],name = player_names[4]),2,2)
fig.add_trace(go.Pie(labels = player_name,values=list(top_bat['Runs'].unique())[:5]),2,3)
fig.update_traces(hole=.4, hoverinfo="label+percent+name")
fig.update_layout(height = 750,width = 900,title_text = text)
fig.show()
In [57]:
teams = ['csk','mi','rcb','kkr','rr','kxip','dd','srh']
top_bat = [top_bat_csk,top_bat_mi,top_bat_rcb,top_bat_kkr,top_bat_rr,top_bat_kxip,top_bat_dd,top_bat_srh]
player_name = [csk_name,mi_name,rcb_name,kkr_name,rr_name,kxip_name,dd_name,srh_name]
player_runs = [csk_runs,mi_runs,rcb_runs,kkr_runs,rr_runs,kxip_runs,dd_runs,srh_runs]
for i,j,k,l,m in zip(range(len(teams)),range(len(player_name)),range(len(player_runs)),range(len(top_bat)),range(len(final))):
drawplots(player_runs[k],player_name[j],final[m],top_bat[l],text = "Top 5 Batsmans From"+" "+str(teams[i]).upper())
print("-"*120)
Observation
From Chennai Super Kings,Suresh Raina,MS Dhoni,MK Hussey have very good Figures
From Mumbai Indians,L Malinga,Harbajan Singh,JJ Bumrah have very good Figures
From Royal Challengers Banglore,Y Chahal,R Vinay Kumar,Z Khan have very good Figures
From Kolkata Knight Riders,S Narine,P Chawla,A Russel have very good Figures
From Rajasthan Royals,SK Trivedi,S Watson,S Warne have very good Figures
From Kings XI Punjab,PP Chawla,Sandeep Sharma,Axar Patel have very good Figures
From Delhi DareDevils,A Mishra,U Yadav,M Morkel have very good Figures
From Surises Hyderabad,Bhuvneshwar Kumar,Rahid Khan,S Kaul have very good Figures
In [74]:
def getresults(df,player_list):
final = {}
for player in player_list:
temp = df[(df['batsman']==player)]
bowler = list(temp['bowler'])
bowler_map = dict(Counter(bowler))
bowler_map = dict(sorted(bowler_map.items(),key=lambda x:x[1],reverse=True))
bowler_map = dict(list(bowler_map.items())[:5])
final[player] = bowler_map
final_name = []
final_runs = []
for key,val in final.items():
player_runs = []
player_name = []
for key1,val1 in val.items():
player_name.append(key1)
player_runs.append(val1)
final_name.append(player_name)
final_runs.append(player_runs)
return final_name,final_runs
In [76]:
teams = ['csk','mi','rcb','kkr','rr','kxip','dd','srh']
In [75]:
csk_player,csk_run = getresults(df_csk,csk_name)
mi_player,mi_run = getresults(df_mi,mi_name)
rcb_player,rcb_run = getresults(df_rcb,rcb_name)
kkr_player,kkr_run = getresults(df_kkr,kkr_name)
rr_player,rr_run = getresults(df_rr,rr_name)
kxip_player,kxip_run = getresults(df_kxip,kxip_name)
dd_player,dd_run = getresults(df_dd,dd_name)
srh_player,srh_run = getresults(df_srh,srh_name)
In [95]:
srh_player[0]
Out[95]:
In [91]:
srh_run
Out[91]:
In [112]:
csk_player
Out[112]:
In [113]:
csk_run
Out[113]:
In [138]:
def drawplots(player_runs,player_names,batsman_name):
fig = make_subplots(rows=2,cols = 3,subplot_titles=tuple(batsman_name),
specs=[[{'type':'domain'},{'type':'domain'},{'type':'domain'}],
[{'type':'domain'},{'type':'domain'},{'type':'domain'}]],shared_yaxes=True)
fig.add_trace(go.Pie(labels=player_names[0],values=player_runs[0],name = batsman_name[0]),1,1)
fig.add_trace(go.Pie(labels=player_names[1],values=player_runs[1],name = batsman_name[1]),1,2)
fig.add_trace(go.Pie(labels=player_names[2],values=player_runs[2],name = batsman_name[2]),1,3)
fig.add_trace(go.Pie(labels=player_names[3],values=player_runs[3],name = batsman_name[3]),2,1)
fig.add_trace(go.Pie(labels=player_names[4],values=player_runs[4],name = batsman_name[4]),2,2)
# fig.add_trace(go.Pie(labels=player_names[4],values=player_runs[4]),2,3)
fig.update_traces(hole=.4, hoverinfo="label+percent+value")
fig.update_layout(height = 750,width = 900)
fig.show()
In [139]:
teams = ['csk','mi','rcb','kkr','rr','kxip','dd','srh']
top_bat = [top_bat_csk,top_bat_mi,top_bat_rcb,top_bat_kkr,top_bat_rr,top_bat_kxip,top_bat_dd,top_bat_srh]
player_name = [csk_player,mi_player,rcb_player,kkr_player,rr_player,kxip_player,dd_player,srh_player]
player_runs = [csk_run,mi_run,rcb_run,kkr_run,rr_run,kxip_run,dd_run,srh_run]
batsman_name = [csk_name,mi_name,rcb_name,kkr_name,rr_name,kxip_name,dd_name,srh_name]
for j,k,l in zip(player_name,player_runs,batsman_name):
drawplots(k,j,l)
print("-"*120)