Match predictions in soccer: Machine learning vs. Poisson approaches

TL;DR

This paper compares machine learning models and Poisson approaches for soccer match prediction across five European leagues, finding minimal impact of feature choice and model type on prediction accuracy.

Contribution

It provides a comparative analysis of ML and Poisson models for soccer predictions, highlighting the limited influence of feature and model selection on results.

Findings

ML and Poisson models perform similarly in predictions

Feature choice has minor impact on prediction quality

Team performance levels are stable throughout the season

Abstract

Predicting the results of soccer matches is of great interest. This is not only due to the popularity of the sport and the joy of private "betting rounds", but also due to the large sports betting market. Where previously expert knowledge and intuition were used, today there are models that analyze large amounts of data and make predictions based on them. In addition to Poisson models, approaches that belong to the machine learning (ML) category are increasingly being used. These include, for example, neural network or random forest models, which are compared in this article with each other as well as with Poisson models with regard to single-match prediction. In each case, the match results of a season are used as the data basis. The analysis is carried out for 5 European top leagues. A statistical analysis shows that the performance levels of the teams do not change systematically during a season. In order to characterize performance levels as accurately as possible, all match results, except from the match to be predicted, can be used as features with equal weighting. It can be seen that both, the exact choice of features and the choice of model, have only a minor influence on the prediction quality. Possible improvements in match prediction are discussed.