Machine Learning for Soccer Match Result Prediction

TL;DR

This paper reviews machine learning methods for soccer match result prediction, highlighting current best models, datasets, and future research directions including model interpretability and richer data features.

Contribution

It provides a comprehensive overview of existing models, datasets, and evaluation methods, and identifies gaps such as the need for deeper comparisons and more interpretable models.

Findings

Gradient-boosted trees like CatBoost perform best on goal-only datasets.

Deep learning and Random Forest models require further comparison.

Incorporating spatiotemporal and event data could improve predictions.

Abstract

Machine learning has become a common approach to predicting the outcomes of soccer matches, and the body of literature in this domain has grown substantially in the past decade and a half. This chapter discusses available datasets, the types of models and features, and ways of evaluating model performance in this application domain. The aim of this chapter is to give a broad overview of the current state and potential future developments in machine learning for soccer match results prediction, as a resource for those interested in conducting future studies in the area. Our main findings are that while gradient-boosted tree models such as CatBoost, applied to soccer-specific ratings such as pi-ratings, are currently the best-performing models on datasets containing only goals as the match features, there needs to be a more thorough comparison of the performance of deep learning models and Random Forest on a range of datasets with different types of features. Furthermore, new rating systems using both player- and team-level information and incorporating additional information from, e.g., spatiotemporal tracking and event data, could be investigated further. Finally, the interpretability of match result prediction models needs to be enhanced for them to be more useful for team management.