Predicting Young Football Talents’ Market Value and Optimizing Team Transfers

This thesis presents a data-driven framework for improving football club transfer strategies by predicting the market value of young players and optimizing player acquisitions. The study begins by constructing two original datasets: one focused on individual player performance, and the other on overall team performance. Using these datasets, the research explores the relationship between key performance indicators and market valuation, with an emphasis on feature engineering and exploratory data analysis. A predictive model is then developed to estimate the market value of young football talents based on historical and performance-related data. Following this, an optimization model is proposed using Pyomo and the Gurobi solver to assist clubs in selecting the most suitable players within given budgetary and squad constraints. The proposed methodology combines machine learning, statistical analysis, and operations research to support informed and strategic decision-making in professional football recruitment.