Football Data Analysis Project

Overview

This project focuses on the analysis of football player data, including both real-world statistics and EA SPORTS FC 24 game attributes. The analysis includes data scraping, cleaning, exploratory data analysis, clustering, and similarity measurement of players.

Project Structure

The project is structured into several Jupyter notebooks, datasets, and supporting folders and files as follows:

• analysis.ipynb
• cleaning.ipynb
• fbr_links.txt
• goalkeepers_df.csv
• goalkeepers_fifa_df.csv
• male_players.csv
• players_df.csv
• players_fifa_df.csv
• README.md
• scraping.ipynb
• tm_links.txt
• English Premier League
• French Ligue 1
• German Bundesliga
• Italian Serie A
• Spanish La Liga

Folders

• English Premier League: This folder contains all files and datasets related to English Premier League
• French Ligue 1: This folder contains all files and datasets related to French Ligue 1
• German Bundesliga: This folder contains all files and datasets related to German Bundesliga
• Italian Serie A: This folder contains all files and datasets related to Italian Serie A
• Spanish La Liga: This folder contains all files and datasets related to Spanish La Liga

Each folder contains the following datasets:

Datasets

• goalkeepers_df.csv: Dataset containing detailed information about goalkeepers.
• goalkeepers_fifa_df.csv: Dataset containing EA Sports FC 24 game attributes for goalkeepers.
• male_players.csv: Dataset with information about male football players.
• players_df.csv: Comprehensive dataset of football players with various attributes.
• players_fifa_df.csv: Dataset containing EA Sports FC 24 game attributes for the players.
• fbr_links.txt: A text file containing links to player data sources.
• tm_links.txt: A text file containing links to transfer market data sources.

Notebooks

• scraping.ipynb: This notebook contains the web scraping scripts used to collect football player data from various sources.
• cleaning.ipynb: This notebook is dedicated to cleaning and preprocessing the raw data obtained from scraping.
• analysis.ipynb: This notebook includes various analyses performed on the cleaned data, including comparative analysis, position-specific analysis, clustering, and similarity measurement.
• playstyles.ipynb: This notebook uses clustering technique to investigate player similarities and play styles
• find_similars.ipynb: This notebook creates a simple model to look for similar players to given ones

Prerequisites

To run the notebooks, you need the following Python packages:

• pandas
• numpy
• sklearn
• matplotlib
• scipy

You can install these packages using pip:

pip install pandas numpy scikit-learn matplotlib scipy

Contact

• Federico Paschetta - email