The Indian Premier League (IPL) is a men's Twenty20 (T20) cricket league that is annually held in India and contested by ten city-based franchise teams. The BCCI founded the league in 2007. The competition is usually held in summer between March and May every year. The IPL is the most-popular cricket league in the world
This EDA is divided into 2 parts. We'll first look at team anlyisis and then move to individual player analysis. The source data is derived from Kaggle.
# Initial imports
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
sns.set()
%matplotlib inline
pd.set_option("display.max_columns", 100)
pd.set_option("display.max_rows", 1000)The original file from Kaggle did not have the match data for IPL 2023. New data was downloaded from a cricket data repository. Some data prep was needed before adding this data to the original file.
#Selecting only the files for IPL 2023 as the downloaded data had all seasons data.
import os
import csv
folder_path = r"C:\Users\naveen.sequeira\Downloads\ipl_csv2"
start_number = 1359475
end_number = 1370353
files = []
for number in range(start_number, end_number + 1):
file_name = str(number) + "_info.csv"
file_path = os.path.join(folder_path, file_name)
if os.path.isfile(file_path):
files.append(file_path)#The combined files were then concatenated vertically.
#But the combined files had an issue. They had team and umpire repeating twice.
#So I had to write some code to rename them as team1, team2, umpire1 and umpire2
lst = pd.DataFrame()
for file in files:
print(file)
match_id = file.split("\\")[-1].split("_")[0]
df = pd.read_csv(file, skiprows=1, usecols=[1,2], header=None, names=["info_", "data"])\
.query("info_ not in ['player', 'registry']")\
.append({"info_": "matchId", "data": match_id}, ignore_index=True)
duplicate_list = df.info_.duplicated(keep=False)
team_duplicates_count = 1
umpire_duplicates_count = 1
for index, is_duplicate in enumerate(duplicate_list):
if is_duplicate:
if df.loc[index, "info_"] == "team":
df.loc[index, "info_"] = f"team{team_duplicates_count}"
team_duplicates_count += 1
elif df.loc[index, "info_"] == "umpire":
df.loc[index, "info_"] = f"umpire{umpire_duplicates_count}"
umpire_duplicates_count += 1
df = df.set_index("info_")
lst = pd.concat([lst, df], axis=1)#Converting the dataframe from long format to wide format
lst = lst.T.reset_index(drop=True).rename_axis(None, axis=1)#Append 2023 data to the master data and save it
matches = matches.append(lst)
matches.to_csv("matches_updated_mens_ipl_2023.csv", index=False)The data for the EDA comes from two csv files. One file consists only match related information and another file contains ball by ball data. We'll explore team analysis using the match data
matches = pd.read_csv("matches_updated_mens_ipl_2023.csv")matches.head()| event | date | team1 | toss_winner | umpire2 | match_referee | season | team2 | winner_runs | balls_per_over | player_of_match | neutralvenue | venue | method | toss_decision | city | winner | umpire1 | match_number | eliminator | winner_wickets | matchId | gender | reserve_umpire | tv_umpire | outcome | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | Indian Premier League | 05-04-2017 00:00 | Sunrisers Hyderabad | Royal Challengers Bangalore | NJ Llong | J Srinath | 2017 | Royal Challengers Bangalore | 35.0 | 6 | Yuvraj Singh | NaN | Rajiv Gandhi International Stadium, Uppal | NaN | field | Hyderabad | Sunrisers Hyderabad | AY Dandekar | 1.0 | NaN | NaN | 1082591 | NaN | NaN | NaN | NaN |
| 1 | Indian Premier League | 06-04-2017 00:00 | Rising Pune Supergiant | Rising Pune Supergiant | S Ravi | M Nayyar | 2017 | Mumbai Indians | NaN | 6 | SPD Smith | NaN | Maharashtra Cricket Association Stadium | NaN | field | Pune | Rising Pune Supergiant | A Nand Kishore | 2.0 | NaN | 7.0 | 1082592 | NaN | NaN | NaN | NaN |
| 2 | Indian Premier League | 07-04-2017 00:00 | Gujarat Lions | Kolkata Knight Riders | CK Nandan | V Narayan Kutty | 2017 | Kolkata Knight Riders | NaN | 6 | CA Lynn | NaN | Saurashtra Cricket Association Stadium | NaN | field | Rajkot | Kolkata Knight Riders | Nitin Menon | 3.0 | NaN | 10.0 | 1082593 | NaN | NaN | NaN | NaN |
| 3 | Indian Premier League | 08-04-2017 00:00 | Kings XI Punjab | Kings XI Punjab | C Shamshuddin | Chinmay Sharma | 2017 | Rising Pune Supergiant | NaN | 6 | GJ Maxwell | NaN | Holkar Cricket Stadium | NaN | field | Indore | Kings XI Punjab | AK Chaudhary | 4.0 | NaN | 6.0 | 1082594 | NaN | NaN | NaN | NaN |
| 4 | Indian Premier League | 08-04-2017 00:00 | Royal Challengers Bangalore | Royal Challengers Bangalore | VK Sharma | J Srinath | 2017 | Delhi Daredevils | 15.0 | 6 | KM Jadhav | NaN | M.Chinnaswamy Stadium | NaN | bat | Bengaluru | Royal Challengers Bangalore | S Ravi | 5.0 | NaN | NaN | 1082595 | NaN | NaN | NaN | NaN |
matches.info()<class 'pandas.core.frame.DataFrame'>
RangeIndex: 1098 entries, 0 to 1097
Data columns (total 26 columns):
# Column Non-Null Count Dtype
--- ------ -------------- -----
0 event 1098 non-null object
1 date 1098 non-null object
2 team1 1098 non-null object
3 toss_winner 1098 non-null object
4 umpire2 1098 non-null object
5 match_referee 1098 non-null object
6 season 1098 non-null int64
7 team2 1098 non-null object
8 winner_runs 503 non-null float64
9 balls_per_over 1098 non-null int64
10 player_of_match 1092 non-null object
11 neutralvenue 77 non-null object
12 venue 1098 non-null object
13 method 20 non-null object
14 toss_decision 1098 non-null object
15 city 1047 non-null object
16 winner 1078 non-null object
17 umpire1 1098 non-null object
18 match_number 1032 non-null float64
19 eliminator 14 non-null object
20 winner_wickets 575 non-null float64
21 matchId 1098 non-null int64
22 gender 74 non-null object
23 reserve_umpire 74 non-null object
24 tv_umpire 74 non-null object
25 outcome 1 non-null object
dtypes: float64(3), int64(3), object(20)
memory usage: 223.2+ KB
matches.describe()| season | winner_runs | balls_per_over | match_number | winner_wickets | matchId | |
|---|---|---|---|---|---|---|
| count | 1098.000000 | 503.000000 | 1098.0 | 1032.000000 | 575.000000 | 1.098000e+03 |
| mean | 2016.080146 | 30.125249 | 6.0 | 31.535853 | 6.180870 | 9.018717e+05 |
| std | 4.855707 | 26.853784 | 0.0 | 18.416685 | 1.841824 | 3.623810e+05 |
| min | 2008.000000 | 1.000000 | 6.0 | 1.000000 | 1.000000 | 3.359820e+05 |
| 25% | 2012.000000 | 10.500000 | 6.0 | 16.000000 | 5.000000 | 5.483322e+05 |
| 50% | 2016.000000 | 22.000000 | 6.0 | 31.000000 | 6.000000 | 9.809640e+05 |
| 75% | 2021.000000 | 41.500000 | 6.0 | 46.000000 | 7.000000 | 1.254065e+06 |
| max | 2023.000000 | 146.000000 | 6.0 | 72.000000 | 10.000000 | 1.370353e+06 |
# Drop unwanted columns
unwanted_cols = ["outcome", "gender", "reserve_umpire","tv_umpire"]
matches = matches.drop(columns=unwanted_cols)#Change date columns to correct format
matches.date = matches.date.astype("datetime64")np.sort(matches.season.unique())array([2008, 2009, 2010, 2011, 2012, 2013, 2014, 2015, 2016, 2017, 2018,
2019, 2020, 2021, 2022, 2023], dtype=int64)
#Correct different names for Rising Pune Supergiant
# First find all columns where the wrong name exists and then change it
value_exists = matches.isin(["Rising Pune Supergiants"]).any()
value_exists[value_exists].index.tolist()
matches[['team1', 'toss_winner', 'team2', 'winner']] = matches[['team1', 'toss_winner', 'team2', 'winner']].replace(
"Rising Pune Supergiants", "Rising Pune Supergiant"
)#Adjust for changed team names
name_changes = {
"Delhi Daredevils": "Delhi Capitals",
"Kings XI Punjab": "Punjab Kings"
}
matches[['team1', 'toss_winner', 'team2', 'winner']] = matches[['team1', 'toss_winner', 'team2', 'winner']].replace(
name_changes
)# shorten team names for convenience
team_names = np.unique([matches.team1, matches.team2]).tolist()
short_names = ["".join([name[0] for name in name.split()]) for name in team_names]
shortened_names = dict(zip(team_names, short_names))
shortened_names["Deccan Chargers"] = "DCH"
shortened_names["Sunrisers Hyderabad"] = "SRH"
shortened_names{'Chennai Super Kings': 'CSK',
'Deccan Chargers': 'DCH',
'Delhi Capitals': 'DC',
'Gujarat Lions': 'GL',
'Gujarat Titans': 'GT',
'Kochi Tuskers Kerala': 'KTK',
'Kolkata Knight Riders': 'KKR',
'Lucknow Super Giants': 'LSG',
'Mumbai Indians': 'MI',
'Pune Warriors': 'PW',
'Punjab Kings': 'PK',
'Rajasthan Royals': 'RR',
'Rising Pune Supergiant': 'RPS',
'Royal Challengers Bangalore': 'RCB',
'Sunrisers Hyderabad': 'SRH'}
matches[['team1', 'toss_winner', 'team2', 'winner']] = matches[['team1', 'toss_winner', 'team2', 'winner']].replace(
shortened_names
)#defunct teams
defunct_teams = ["GL", "DCH", "KTK", "PW", "RPS"]pd.DataFrame({
"Matches":matches.groupby("team1").team1.count() + matches.groupby("team2").team2.count(),
"Matches_Won": matches.groupby("winner").winner.count()
})\
.assign(Win_pct=lambda x: (x.Matches_Won / x.Matches).round(2))\
.drop(defunct_teams)\
.sort_values("Win_pct", ascending=False)| Matches | Matches_Won | Win_pct | |
|---|---|---|---|
| GT | 50 | 34 | 0.68 |
| CSK | 240 | 141 | 0.59 |
| LSG | 45 | 25 | 0.56 |
| MI | 263 | 147 | 0.56 |
| KKR | 251 | 125 | 0.50 |
| RR | 220 | 108 | 0.49 |
| RCB | 254 | 121 | 0.48 |
| SRH | 180 | 82 | 0.46 |
| PK | 246 | 110 | 0.45 |
| DC | 252 | 110 | 0.44 |
(
matches.groupby("winner").winner.count() /
(matches.groupby("team1").team1.count() + matches.groupby("team2").team2.count())
)\
.drop(defunct_teams)\
.sort_values(ascending=False).plot(kind="bar")
plt.title("Win Percentage")
plt.xticks(rotation=40)
plt.xlabel("")
plt.savefig("win_pct.svg");
matches.toss_decision.value_counts()\
.plot(kind="pie", autopct='%1.1f%%')
plt.title("Teams winning toss choose:")
plt.axis("off")
plt.xlabel("");
pd.Series(
[matches[matches.toss_winner == matches.winner].winner.count(),
matches[matches.toss_winner != matches.winner].winner.count()],
index=["Won", "Lost"], name=""
).plot.pie(autopct='%1.1f%%', startangle=50, colors=[ "#F7B7A3","#FFF1C9"])
plt.title("Toss Factor (After winning toss - Win%)");
pd.Series(
[matches.query("toss_decision=='field' and winner_wickets == winner_wickets").winner.count(),
matches.query("toss_decision=='bat' and winner_runs == winner_runs").winner.count()],
index=["Chasing", "Bat First"], name="")\
.plot.pie(autopct='%1.1f%%', colors=["#726FB8", "#EA5F89"])
plt.title("Batting First vs Chasing (Win%)");
matches.player_of_match.value_counts().nlargest(10).sort_values().plot.barh(figsize=(10,6))
a = matches.player_of_match.value_counts().nlargest(10).sort_values()
for i, value in enumerate(a):
plt.text(value+0.08, i, str(round(value)), ha="left", va="center_baseline")
plt.xticks([])
plt.title("IPL - Most Man of the Match Awards")
plt.subplots_adjust(left=0.2);
matches.umpire1.value_counts().add(matches.umpire2.value_counts(), fill_value=0).nlargest(10).sort_values().plot.barh()
a = matches.umpire1.value_counts().add(matches.umpire2.value_counts(), fill_value=0).nlargest(10).sort_values()
plt.title("Most matches as Umpire")
plt.xticks([])
for i, value in enumerate(a):
plt.text(value+0.08, i, str(round(value)), ha="left", va="center_baseline");
matches.match_referee.value_counts().nlargest(10).sort_values().plot.barh()
plt.title("Most matches as Referee");
matches.city.value_counts().nlargest(10).sort_values().plot.barh(figsize=(10,6))
plt.title("City with most matches")
plt.subplots_adjust(left=0.2)
plt.savefig("city_mostMatches.svg");
points_table = pd.read_html("https://www.cricketwa.com/pointstable/indian-premier-league.aspx")[1]points_table = points_table.drop(["Rank", "AVG"], axis=1).replace("-", np.nan)points_table.iloc[:, 1:] = points_table.iloc[:, 1:].astype(float)points_table.to_excel("ipl_allSeasons_pointsTable.xlsx", index=False)# A Seaborn Heatmap was used create the ranking table
import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt
# Create the heatmap
plt.figure(figsize=(10, 6)) # Adjust the figure size as needed
sns.heatmap(points_table.set_index("Team").sort_index(axis=1),
annot=True, cmap='coolwarm', cbar=False,)
# Customize the chart
plt.ylabel(None)
plt.title('IPL Table Rankings (2008-2023)')
plt.subplots_adjust(left=0.22)
plt.savefig("team_rank.png")
players = pd.read_csv("IPL_ball_by_ball_updated.csv", parse_dates=["start_date"])players| match_id | season | start_date | venue | innings | ball | batting_team | bowling_team | striker | non_striker | bowler | runs_off_bat | extras | wides | noballs | byes | legbyes | penalty | wicket_type | player_dismissed | other_wicket_type | other_player_dismissed | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 335982 | 2008 | 2008-04-18 | M Chinnaswamy Stadium | 1 | 0.1 | Kolkata Knight Riders | Royal Challengers Bangalore | SC Ganguly | BB McCullum | P Kumar | 0 | 1 | NaN | NaN | NaN | 1.0 | NaN | NaN | NaN | NaN | NaN |
| 1 | 335982 | 2008 | 2008-04-18 | M Chinnaswamy Stadium | 1 | 0.2 | Kolkata Knight Riders | Royal Challengers Bangalore | BB McCullum | SC Ganguly | P Kumar | 0 | 0 | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN |
| 2 | 335982 | 2008 | 2008-04-18 | M Chinnaswamy Stadium | 1 | 0.3 | Kolkata Knight Riders | Royal Challengers Bangalore | BB McCullum | SC Ganguly | P Kumar | 0 | 1 | 1.0 | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN |
| 3 | 335982 | 2008 | 2008-04-18 | M Chinnaswamy Stadium | 1 | 0.4 | Kolkata Knight Riders | Royal Challengers Bangalore | BB McCullum | SC Ganguly | P Kumar | 0 | 0 | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN |
| 4 | 335982 | 2008 | 2008-04-18 | M Chinnaswamy Stadium | 1 | 0.5 | Kolkata Knight Riders | Royal Challengers Bangalore | BB McCullum | SC Ganguly | P Kumar | 0 | 0 | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN |
| ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... |
| 243810 | 1370353 | 2023 | 2023-05-29 | Narendra Modi Stadium, Ahmedabad | 2 | 14.2 | Chennai Super Kings | Gujarat Titans | S Dube | RA Jadeja | MM Sharma | 1 | 0 | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN |
| 243811 | 1370353 | 2023 | 2023-05-29 | Narendra Modi Stadium, Ahmedabad | 2 | 14.3 | Chennai Super Kings | Gujarat Titans | RA Jadeja | S Dube | MM Sharma | 1 | 0 | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN |
| 243812 | 1370353 | 2023 | 2023-05-29 | Narendra Modi Stadium, Ahmedabad | 2 | 14.4 | Chennai Super Kings | Gujarat Titans | S Dube | RA Jadeja | MM Sharma | 1 | 0 | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN |
| 243813 | 1370353 | 2023 | 2023-05-29 | Narendra Modi Stadium, Ahmedabad | 2 | 14.5 | Chennai Super Kings | Gujarat Titans | RA Jadeja | S Dube | MM Sharma | 6 | 0 | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN |
| 243814 | 1370353 | 2023 | 2023-05-29 | Narendra Modi Stadium, Ahmedabad | 2 | 14.6 | Chennai Super Kings | Gujarat Titans | RA Jadeja | S Dube | MM Sharma | 4 | 0 | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN |
243815 rows × 22 columns
players.info()<class 'pandas.core.frame.DataFrame'>
RangeIndex: 243815 entries, 0 to 243814
Data columns (total 22 columns):
# Column Non-Null Count Dtype
--- ------ -------------- -----
0 match_id 243815 non-null int64
1 season 243815 non-null int64
2 start_date 243815 non-null datetime64[ns]
3 venue 243815 non-null object
4 innings 243815 non-null int64
5 ball 243815 non-null float64
6 batting_team 243815 non-null object
7 bowling_team 243815 non-null object
8 striker 243815 non-null object
9 non_striker 243815 non-null object
10 bowler 243815 non-null object
11 runs_off_bat 243815 non-null int64
12 extras 243815 non-null int64
13 wides 7652 non-null float64
14 noballs 1018 non-null float64
15 byes 628 non-null float64
16 legbyes 3778 non-null float64
17 penalty 2 non-null float64
18 wicket_type 12067 non-null object
19 player_dismissed 12067 non-null object
20 other_wicket_type 0 non-null float64
21 other_player_dismissed 0 non-null float64
dtypes: datetime64[ns](1), float64(8), int64(5), object(8)
memory usage: 40.9+ MB
# Remove unnecessary columns
players = players.drop(["other_wicket_type", "other_player_dismissed"], axis=1)players.describe()| match_id | season | innings | ball | runs_off_bat | extras | wides | noballs | byes | legbyes | penalty | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| count | 2.438150e+05 | 243815.000000 | 243815.000000 | 243815.000000 | 243815.000000 | 243815.000000 | 7652.000000 | 1018.000000 | 628.000000 | 3778.000000 | 2.0 |
| mean | 8.707326e+05 | 2015.611119 | 1.483781 | 9.557662 | 1.252614 | 0.067174 | 1.204260 | 1.034381 | 1.867834 | 1.304129 | 5.0 |
| std | 3.532362e+05 | 4.650281 | 0.502855 | 5.686347 | 1.626764 | 0.341990 | 0.788406 | 0.325568 | 1.312454 | 0.847922 | 0.0 |
| min | 3.359820e+05 | 2008.000000 | 1.000000 | 0.100000 | 0.000000 | 0.000000 | 1.000000 | 1.000000 | 1.000000 | 1.000000 | 5.0 |
| 25% | 5.483160e+05 | 2012.000000 | 1.000000 | 4.500000 | 0.000000 | 0.000000 | 1.000000 | 1.000000 | 1.000000 | 1.000000 | 5.0 |
| 50% | 8.298190e+05 | 2015.000000 | 1.000000 | 9.400000 | 1.000000 | 0.000000 | 1.000000 | 1.000000 | 1.000000 | 1.000000 | 5.0 |
| 75% | 1.216507e+06 | 2020.000000 | 2.000000 | 14.400000 | 1.000000 | 0.000000 | 1.000000 | 1.000000 | 4.000000 | 1.000000 | 5.0 |
| max | 1.370353e+06 | 2023.000000 | 6.000000 | 19.900000 | 6.000000 | 7.000000 | 5.000000 | 5.000000 | 4.000000 | 5.000000 | 5.0 |
players.groupby("striker").runs_off_bat.sum().nlargest(10).sort_values()\
.plot.barh()
plt.ylabel(None);
players.groupby(["match_id","striker"], as_index=False).runs_off_bat.sum()\
.query("runs_off_bat >= 100")\
.groupby("striker").striker.count().nlargest(10).sort_values()\
.plot.barh()
plt.ylabel(None);
To omit outliers we'll use a qualification of min 200 balls faced for most of our remaining stats. We'll use this qualification to filter our results.
above_200_balls = players.groupby("striker").ball.count()\
[lambda x: x>=200].index.to_list()
qualifiers = lambda x: x.index.isin(above_200_balls)players.groupby("striker").runs_off_bat.sum()\
[qualifiers]\
.div(players.groupby("striker").match_id.nunique()
[qualifiers])\
.nlargest(10).sort_values()\
.plot.barh()
plt.ylabel(None);
# We have to exclude the wides and noballs to get the exact strike rate
players.groupby("striker").runs_off_bat.sum()\
[qualifiers]\
.div(players.query("wides != wides and noballs != noballs")
.groupby("striker").ball.count()
[qualifiers]).mul(100)\
.nlargest(10).sort_values()\
.plot.barh()
plt.ylabel(None);
avg = players.groupby("striker").runs_off_bat.sum()\
[qualifiers]\
.div(players.groupby("striker").match_id.nunique()
[qualifiers])
sr = players.groupby("striker").runs_off_bat.sum()\
[qualifiers]\
.div(players.query("wides != wides and noballs != noballs")
.groupby("striker").ball.count()
[qualifiers]).mul(100)
avg_sr = pd.DataFrame({"Average": avg, "StrikeRate": sr}).query("Average >= 20 and StrikeRate >= 145")
plt.figure(figsize=(9,5))
ax = sns.scatterplot(x="Average", y="StrikeRate", data=avg_sr)
for i, row in avg_sr.iterrows():
ax.text(row["Average"]+0.1, row["StrikeRate"]+0.1, i);
players.query("runs_off_bat == 6")\
.groupby("striker").runs_off_bat.count()\
.nlargest(10).sort_values()\
.plot.barh()
plt.ylabel(None);
players.query("runs_off_bat == 6")\
.groupby("striker").runs_off_bat.count()\
[qualifiers]\
.div(players.groupby("striker").match_id.nunique()
[qualifiers])\
.nlargest(10).sort_values()\
.plot.barh()
plt.ylabel(None);
players.query("runs_off_bat == 6")\
.groupby(["match_id", "striker"], as_index=False).runs_off_bat.count()\
.iloc[:, 1:].nlargest(10, "runs_off_bat")\
.set_index("striker").sort_values("runs_off_bat")\
.plot.barh(legend=False)
plt.ylabel(None);
players.query("wides != wides and noballs != noballs")\
.groupby("striker").ball.count()\
[qualifiers]\
.div(players.query("runs_off_bat == 6")\
.groupby("striker").runs_off_bat.count()
[qualifiers])\
.nsmallest(10).sort_values(ascending=False)\
.plot.barh(legend=False, figsize=(8,6))
plt.subplots_adjust(left=0.2)
plt.ylabel(None);
# plt.title("Least per six across all IPL Seasons\n"
# "Lower the better (Qualification: Balls faced ≥200)");
players.query("runs_off_bat == 4")\
.groupby("striker").runs_off_bat.count()\
.nlargest(10).sort_values()\
.plot.barh()
plt.ylabel(None);
players.query("runs_off_bat in [6,4]")\
.groupby(["striker", "runs_off_bat"]).ball.count().unstack()\
.assign(Total= lambda x: x[4] + x[6]).nlargest(10, "Total").sort_values("Total").drop(columns="Total")\
.plot(kind="barh", stacked=True)
plt.ylabel(None)
plt.legend();
(
players.query("runs_off_bat in [6,4]")
.groupby("striker").runs_off_bat.sum()
[qualifiers]
.div(players.groupby("striker").runs_off_bat.sum()
[qualifiers])
.nlargest(10).sort_values()
.plot.barh()
)
plt.ylabel(None);