Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Performance Insights-based AI-driven Football Transfer Fee Prediction

Daniil Sulimov

TL;DR

The paper tackles the problem of predicting football transfer fees using performance-driven signals by integrating an expected-goals–based scoring system with a transfer-value forecasting component. It develops three predictive modules (an xG model, a goal-scoring predictor, and a transfer-fee forecaster) using event-stream data from StatsBomb and Kaggle, with CatBoost frequently delivering the strongest performance (e.g., high recall and AUC in the xG and scoring predictors). The forecasting module estimates transfer-value gains/losses with regression models, finding Random Forest to yield the best MAE around $4.26$ million, demonstrating robustness across large value ranges. Practically, the approach is validated on Euro 2020 data to produce a symbolic team and predicted value changes for selected players, illustrating tangible use for recruitment decisions and budgeting, while acknowledging avenues for improvement via optical tracking and richer feature sets.

Abstract

We developed an artificial intelligence approach to predict the transfer fee of a football player. This model can help clubs make better decisions about which players to buy and sell, which can lead to improved performance and increased club budgets. Having collected data on player performance, transfer fees, and other factors that might affect a player's value, we then used this data to train a machine learning model that can accurately predict a player's impact on the game. We further passed the obtained results as one of the features to the predictor of transfer fees. The model can help clubs identify players who are undervalued and who could be sold for a profit. It can also help clubs avoid overpaying for players. We believe that our model can be a valuable tool for football clubs. It can help them make better decisions about player recruitment and transfers.

Performance Insights-based AI-driven Football Transfer Fee Prediction

TL;DR

The paper tackles the problem of predicting football transfer fees using performance-driven signals by integrating an expected-goals–based scoring system with a transfer-value forecasting component. It develops three predictive modules (an xG model, a goal-scoring predictor, and a transfer-fee forecaster) using event-stream data from StatsBomb and Kaggle, with CatBoost frequently delivering the strongest performance (e.g., high recall and AUC in the xG and scoring predictors). The forecasting module estimates transfer-value gains/losses with regression models, finding Random Forest to yield the best MAE around million, demonstrating robustness across large value ranges. Practically, the approach is validated on Euro 2020 data to produce a symbolic team and predicted value changes for selected players, illustrating tangible use for recruitment decisions and budgeting, while acknowledging avenues for improvement via optical tracking and richer feature sets.

Abstract

We developed an artificial intelligence approach to predict the transfer fee of a football player. This model can help clubs make better decisions about which players to buy and sell, which can lead to improved performance and increased club budgets. Having collected data on player performance, transfer fees, and other factors that might affect a player's value, we then used this data to train a machine learning model that can accurately predict a player's impact on the game. We further passed the obtained results as one of the features to the predictor of transfer fees. The model can help clubs identify players who are undervalued and who could be sold for a profit. It can also help clubs avoid overpaying for players. We believe that our model can be a valuable tool for football clubs. It can help them make better decisions about player recruitment and transfers.
Paper Structure (8 sections, 7 equations, 5 figures, 7 tables)

This paper contains 8 sections, 7 equations, 5 figures, 7 tables.

Table of Contents

  1. Introduction
  2. Materials and methods
  3. xG model development
  4. Goal-scoring predictor development
  5. Development of forecasting model for transfer fees
  6. Results
  7. Discussion
  8. Conclusions

Figures (5)

  • Figure 1: Possession chain.Black points represent ball's position, arrows - direction of the actions performed over it, signed next to each arrow.
  • Figure 2: Examples of a shooting angle from different views.
  • Figure 3: Distance to the goals. Red dot represents the centre of the goals.
  • Figure 4: Precision-recall curves.
  • Figure 5: Team of the European Football Championship 2020, formed by the algorithm (letters stand for the pitch position)