Title
PlayerROI
Gallery
What it does
Decision Support System (DSS) for professional soccer clubs.
A full-stack transfer planning platform that helps clubs optimize their squad building. It reconciles performance data against market value and age-decline curves to find undervalued players and solve for optimal 25-man squad configurations.
Core Features
- ✦Multi-threaded pipeline for 40+ performance metrics
- ✦K-Means clustering for player archetype detection
- ✦Constrained linear programming for squad optimization
- ✦Automated ROI valuation & age-decline modeling
How I built it
The project began as a data scraping experiment to identify undervalued 'hidden gems' in European leagues. It evolved into a full DSS after implementing the linear programming solver. The most challenging part was reconciling disparate data sources into a unified metric schema that could feed the ROI algorithm.
System Flow & Architecture
Engineered a high-concurrency Python data pipeline that reduced scraping time by 90% via threading. The backend uses a two-stage optimization engine: scikit-learn for player clustering and PuLP for solving complex squad-building constraints (budget, HG status, tactical depth). Syncs directly to Supabase Postgres for real-time club advisory views.
Tech Stack