This project employs advanced machine learning techniques, including Linear Regression and Random Forest models, to predict and evaluate football player performance. By analyzing a dataset from the Transfermarkt database, which includes detailed player statistics and performance metrics, the project aims to assist football clubs in making informed decisions regarding player acquisitions. The analysis focuses on creating predictive models that can forecast player scores for upcoming seasons and assess their market value to identify undervalued players. This work emphasizes the development and comparison of different predictive models, and the integration of features like growth rates and performance trends to enhance prediction accuracy. Although the results show promise, the complexity of real-world sports analytics and the limitations of the available dataset suggest cautious application. This project not only demonstrates the potential of using data science in sports analytics but also outlines future pathways for refining these models to better meet the dynamic needs of football team management.