Data Analysis Football Project Using Python

This project serves as a comprehensive guide to performing football analysis using Python. It covers data collection, cleaning, manipulation, and visualization to gain insights into football data.

Live Project Demo

You can view my live project demo: https://youtu.be/_7Q9iiH2A10

Requirements

To successfully run this project, you will need the following:

• Python 3.x
• Jupyter Notebook
• Pandas
• Numpy
• Matplotlib
• Seaborn

Installation

To get started, follow these installation steps:

• Install Python 3.x from the official website.
• Open the Command Prompt terminal.
• Type python --version to check if Python is already installed and if it is, verify that the version is up-to-date.
• Verify the preferred installer program, pip, by typing pip --version.
• Install Jupyterlab by running pip install jupyterlab in the Command Prompt terminal.
• Install Jupyter Notebook by running pip install jupyter notebook in the Command Prompt terminal.
• Once Jupyter Notebook has finished downloading, you will see the file path location in the terminal. Copy this path file location and paste it on a new line, and then hit enter. In my case, the file path is:
  

• Install the necessary packages by running pip install pandas, pip install numpy, pip install matplotlib, and pip install seaborn in the Command Prompt terminal.
• Create a folder on your desktop for this project. For example:
  
  
  
  
• Once you have created the desktop folder, open the folder and copy the folder path location. In my case, the folder path location is: C:\Users\ayans\Desktop\Test_Jupyter
• Paste the folder path location into the Command Prompt terminal by typing cd C:\Users\ayans\Desktop\Test_Jupyter
• After pasting the folder path location, the path will change to:
  
  
  
  
• Type jupyter notebook and hit enter. You will see a new tab open in your browser displaying the Jupyter Notebook dashboard.
• Click on the following options to get started, as shown in the image below:
  
  
  
  

Jupyter Notebook - Python Programming

To begin football data analysis with Python, follow these steps:

Step 1:

# Import Librarirs

import pandas as pd import numpy as np import matplotlib.pyplot as plt import seaborn as sns %matplotlib inline

Step 2:

# Load EPL data into a DataFrame

epldf = pd.readcsv('C:\\Users\\ayans\\Downloads\\EPL2021\\EPL2021.csv') epl_df.head()

Step 3:

# Print a summary of the DataFrame

epl_df.info()

Step 4:

# Print summary statistics of the DataFrame

epl_df.describe()

Step 5:

# Count the number of missing values (NaN) in each column of epl_df

epl_df.isna().sum()

Step 6:

# Create 2 new columns

epldf['MinsPerMatch'] = (epldf['Mins'] / epl_df['Matches']).astype(int) epldf['GoalsPerMatch'] = (epldf['Goals'] / epl_df['Matches']).astype(float) epl_df.head()

Step 7:

# Total Goals

TotalGoals = epldf['Goals'].sum() print(Total_Goals)

Step 8:

# Penalty Goals

TotalPenaltyGoals = epldf['Penalty_Goals'].sum() print(Total_PenaltyGoals)

Step 9:

# Penalty Attempts

TotalPenaltyAttempts = epldf['Penalty_Attempted'].sum() print(Total_PenaltyAttempts)

Step 10:

# Pie chart for penalties missed vs scored

plt.figure(figsize = (13, 6)) plnotscored = epldf['PenaltyAttempted'].sum() - Total_PenaltyGoals data = [plnotscored, Total_PenaltyGoals] labels = ['Penalties Missed', 'Penalty Scored'] colorpalette = sns.colorpalette("Paired") plt.pie(data, labels = labels, colors = color_palette, autopct = '%.0f%%') plt.show()

Step 11:

# Unique positions

epl_df['Position'].unique()

Step 12:

# Total FW Players

epldf[epldf['Position'] == 'FW']

Step 13:

# Players from different nations

np.size((epl_df['Nationality'].unique()))

Step 14:

# Most players from which countries

nationality = epldf.groupby('Nationality').size().sortvalues(ascending = False) nationality.head(10).plot(kind = 'bar', figsize = (12, 6), color = sns.color_palette('magma'))

Step 15:

# Clubs with maximum players in their squad

epldf['Club'].valuecounts().nlargest(5).plot(kind = 'bar', color = sns.color_palette("viridis"))

Step 16:

# Clubs with latest players in their squad

epldf['Club'].valuecounts().nsmallest(5).plot(kind = 'bar', color = sns.color_palette("viridis"))

Step 17:

# Players based on age group

Under20 = epldf[epldf['Age'] <= 20] age2025 = epldf[(epldf['Age'] > 20) & (epldf['Age'] <= 25)] age2530 = epldf[(epldf['Age'] > 25) & (epldf['Age'] <= 30)] Above30 = epldf[epldf['Age'] > 30]

Step 18:

# Assuming the following DataFrame exist: Under20, age2025, age2530 and Above30

x = np.array([Under20['Name'].count(),age2025['Name'].count(),age2530['Name'].count(),Above30['Name'].count()]) mylabels = ["<=20", ">20 & <=25", ">25 & <=30", ">30"] plt.title('Total Players with Age Groups', fontsize=20) plt.pie(x, labels=mylabels, autopct = "%.1f%%") plt.show()

Step 19:

# Total under 20 players in each club

playersunder20 = epldf[epldf['Age'] < 20] playersunder20['Club'].valuecounts().plot(kind = 'bar', color = sns.colorpalette("cubehelix"))

Step 20:

# Under 20 players in Manu

playersunder20[playersunder20["Club"] == 'Manchester United']

Step 21:

# Under 20 players in Chelsea

playersunder20[playersunder20["Club"] == 'Chelsea']

Step 22:

# Average age of players in each club

plt.figure(figsize = (12, 6)) sns.boxplot(x = 'Club', y = 'Age', data = epl_df) plt.xticks(rotation = 90)

Step 23:

# Group the English Premier League DataFrame (epl_df) by club and count the number of players in each club

numplayer = epldf.groupby('Club').size() data = (epldf.groupby('Club')['Age'].sum()) / numplayer data.sort_values(ascending = False)

Step 24:

# Total assists from each club

Assitsbyclub = pd.DataFrame(epldf.groupby('Club', asindex = False)['Assists'].sum()) sns.settheme(style = "whitegrid", colorcodes = True) ax = sns.barplot(x = 'Club', y = 'Assists', data = Assitsbyclub.sort_values(by = 'Assists'), palette = 'tab20') ax.set_xlabel("Club", fontsize = 30) ax.set_ylabel("Assists", fontsize = 20) plt.xticks(rotation = 75) plt.rcParams["figure.figsize"] = (20, 8) plt.title('Plot of Club vs Total Assists', fontsize = 20)

Step 25:

# Top 10 Assists

top10assists = epl_df[['Name', 'Club', 'Assists', 'Matches']].nlargest(n = 10, columns = 'Assists') top10assists

Step 26:

# Creating a DataFrame to group the total goals scored by each club

Goalsbyclubs = pd.DataFrame(epldf.groupby('Club', asindex = False)['Goals'].sum()) sns.settheme(style ="whitegrid", colorcodes = True) ax = sns.barplot(x = 'Club', y = 'Goals', data = Goalsbyclubs.sort_values(by ="Goals"), palette = 'rocket') ax.set_xlabel("Club", fontsize = 30) ax.set_ylabel("Goals", fontsize = 20) plt.xticks(rotation =75) plt.rcParams["figure.figsize"] = (20, 8) plt.title('Plot of Club vs Total Goals', fontsize = 20)

Step 27:

# Most goals by players

top10goals = epl_df[['Name', 'Club', 'Goals', 'Matches']].nlargest(n = 10, columns = 'Goals') top10goals

Step 28:

# Goals per match

top10goalspermatch = epl_df[['Name', 'GoalsPerMatch', 'Matches', 'Goals']].nlargest(n = 10, columns = 'GoalsPerMatch') top10goalspermatch

Step 29:

# Pie Chart - Goals with assist and without assist

plt.figure(figsize = (14, 7)) assists = epl_df['Assists'].sum() data = [Total_Goals - assists, assists] labels = ['Goals without assists', 'Goals with assists'] color = sns.color_palette('Set2') plt.pie(data, labels = labels, colors = color, autopct ='%.0f%%') plt.title('Percentage of Goals with Assists', fontsize = 20) plt.show()

Step 30:

# Top 10 players with most yellow cards

eplyellow = epldf.sortvalues(by = 'YellowCards', ascending = False)[:10] plt.figure(figsize = (20, 6)) plt.title("Players with the most yellow cards") c = sns.barplot(x = eplyellow['Name'], y = eplyellow['Yellow_Cards'], label = 'Players', color ='yellow') plt.ylabel('Number of Yellow Cards') c.setxticklabels(c.getxticklabels(), rotation = 45) plt.show()

Data Visualization

Finally, we will visualize the data to make it more accessible and understandable. This includes creating plots, charts, and graphs.

Conclusion

This project provides a comprehensive guide to performing data analysis football using Python. By following these steps, you will be able to collect, clean, manipulate, and visualize football data. Enjoy exploring the world of football data analytics!

Thank you for watching this project.