2025 · Machine Learning
Player Market Value Analysis in Elite European Football
Machine Learning Sports Analytics Regression
Player Market Value Analysis in Elite European Football
University of Maryland | Sep 2025 – Dec 2025
Python Scikit-Learn Ridge/Lasso Regression Random Forest XGBoost
Project Overview
Conducted comprehensive econometric analysis of transfer market valuation for 600+ players across Europe’s top 5 leagues (Premier League, La Liga, Bundesliga, Serie A, Ligue 1).
Key Contributions
Feature Engineering: Engineered 15+ domain-specific features including goal contribution rate, minutes-per-appearance ratios, and league prestige weights to capture latent valuation drivers
Model Development: Benchmarked regularized regression (Ridge, Lasso, Elastic Net), Random Forest, and XGBoost models achieving optimal R² = 0.78 on holdout test set
Interpretability: Applied SHAP values and permutation importance to identify age, goals scored, and league tier as top-3 valuation predictors, informing transfer strategy recommendations
Technologies Used
- Languages: Python
- ML Libraries: Scikit-Learn, XGBoost
- Analysis: SHAP, Permutation Importance
- Data: TransferMarkt, FBref
Key Results
| Metric | Value |
|---|---|
| Players Analyzed | 600+ |
| Features Engineered | 15+ |
| Best Model R² | 0.78 |
| Top Leagues Covered | 5 |