A Foundation Model for Soccer

TL;DR

This paper introduces a transformer-based foundation model for soccer that predicts match actions, demonstrating its effectiveness through comparison with baseline models and discussing potential real-world applications.

Contribution

It presents the first transformer-based foundation model for soccer action prediction trained on multiple seasons of professional data.

Findings

Transformer outperforms Markov and MLP models in prediction accuracy.

Model generalizes well across different match scenarios.

Open-source implementation available for further research.

Abstract

We propose a foundation model for soccer, which is able to predict subsequent actions in a soccer match from a given input sequence of actions. As a proof of concept, we train a transformer architecture on three seasons of data from a professional soccer league. We quantitatively and qualitatively compare the performance of this transformer architecture to two baseline models: a Markov model and a multi-layer perceptron. Additionally, we discuss potential applications of our model. We provide an open-source implementation of our methods at https://github.com/danielhocevar/Foundation-Model-for-Soccer.