A Foundation Model for Soccer
This work addresses the need for predictive modeling in soccer analytics, but it is incremental as it applies existing transformer methods to a new domain-specific dataset.
The authors tackled the problem of predicting subsequent actions in soccer matches by training a transformer model on three seasons of professional league data, achieving performance that was quantitatively and qualitatively compared to baseline models like a Markov model and a multi-layer perceptron.
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.