⚽ Belgian Jupiler Pro League Analysis 2021/22

A deep-dive analysis of the 2021/22 Belgian Jupiler Pro League season using StatsBomb event-level data. This project explores team and player performance metrics, with a focus on midfielder evaluation and Club Brugge's championship-winning campaign.

📊 Project Overview

An exploration of 1.1M+ match events from the Belgian top-flight, answering key performance questions and building custom midfielder evaluation metrics.

Key Insights

Midfielder Metrics Developed

Five custom metrics to evaluate midfielders:

1. Aerial Wins - Total headers won
2. Aerial Win % - Efficiency in aerial duels
3. Long Balls - Completed passes ≥35 yards
4. Assisted xG - Expected goals from shot-creating passes
5. Box Positioning - In-box touches during possession

🌟 Top Performers

🏆 Elite Midfielders by Category

Aerial Specialists:

• Vinicius de Souza Costa (98 wins)
• Julien De Sart (92 wins)

Long Ball Masters:

• Josh Cullen (233 completed)
• Julien De Sart (196 completed)

Complete Playmakers:

• Hans Vanaken (5.01 xG assisted, 327 box touches)
• Lior Refaelov (5.71 xG assisted, 223 box touches)

🗂️ Repository Structure

belgian-football-analysis/
│
├── 📓 notebooks/
│   └── jupiler_league_analysis.ipynb   # Main analysis
│
├── 📊 data/
│   ├── matches-events-bel1-2122.parquet
│   └── data-events-matches.zip
│
├── 📁 docs/
│   ├── documentation-events.pdf
│   └── documentation-matches.pdf
│
├── 📈 outputs/
│   ├── top_shooters.png
│   ├── team_xg.png
│   ├── yellow_cards.png
│   ├── cb_shot_assists.png
│   ├── cb_counterpressures.png
│   ├── midfielder_radar.png
│   ├── midfielder_rankings.csv
│   └── results_summary.json
│
├── README.md
├── requirements.txt
└── .gitignore

🚀 Quick Start

Prerequisites

• Python 3.9+
• pip

Installation

# Clone the repository
git clone https://github.com/yourusername/belgian-football-analysis.git
cd belgian-football-analysis

# Create virtual environment
python -m venv venv
source venv/bin/activate  # Windows: venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt

Run the Analysis

jupyter notebook notebooks/jupiler_league_analysis.ipynb

📈 Visualizations

The analysis includes several visualizations:

• 📊 Shot Leaders - Top 10 shooters across the league
• 📈 Team xG Rankings - All 18 teams by expected goals
• 🎯 Midfielder Radar - Multi-dimensional player profiles
• 🔥 Pressing Intensity - Counterpressure analysis
• 📍 Attacking Positions - Box involvement metrics

📦 Data Source

🛠️ Tech Stack

• pandas - Data manipulation
• NumPy - Numerical operations
• matplotlib - Visualizations
• seaborn - Statistical graphics
• pyarrow - Parquet handling

📝 License

MIT License - see LICENSE for details.

🙏 Acknowledgments

• StatsBomb for the open event data
• Belgian Jupiler Pro League for a great 2021/22 season
• Club Brugge - Champions! 🏆

Built with ❤️ for football analytics