Automated Explanation of Machine Learning Models of Footballing Actions in Words

2025-04-01

AI Summary

In football analytics, a critical interpretability gap exists between machine learning models and domain experts (e.g., coaches, commentators): current models—such as xG estimators—lack mechanisms to translate statistical predictions into actionable, natural-language insights. To address this, we propose “wordalization,” a novel paradigm integrating semantic mapping of regression coefficients with large language model (LLM) prompting. Our method employs logistic regression for its intrinsic coefficient interpretability, extracting feature-level contributions to shot outcome prediction, then leverages an LLM to generate accurate, concise, and expressive Chinese textual explanations. We implement an open-source, interactive web system deployed for real-time analysis of recent international tournaments. This work is the first to systematically bridge the gap between statistical prediction and domain-expert communication, demonstrating significant improvements in coaching decision support and broadcast narrative quality.

Technology Category

• Natural Language Processing: Interpretability, Analysis, and Evaluation of NLP Models
• Machine Learning: Transparent, Interpretable, Explainable ML
• Humans and AI: Explainable AI (XAI) for Human Understanding

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

While football analytics has changed the way teams and analysts assess performance, there remains a communication gap between machine learning practice and how coaching staff talk about football. Coaches and practitioners require actionable insights, which are not always provided by models. To bridge this gap, we show how to build wordalizations (a novel approach that leverages large language models) for shots in football. Specifically, we first build an expected goals model using logistic regression. We then use the co-efficients of this regression model to write sentences describing how factors (such as distance, angle and defensive pressure) contribute to the model's prediction. Finally, we use large language models to give an entertaining description of the shot. We describe our approach in a model card and provide an interactive open-source application describing shots in recent tournaments. We discuss how shot wordalisations might aid communication in coaching and football commentary, and give a further example of how the same approach can be applied to other actions in football.

Problem

• Bridge communication gap between ML models and football coaching
• Generate actionable insights for coaches using LLMs
• Explain expected goals model factors in human-readable sentences

Innovation

• Leverage large language models for wordalizations
• Build expected goals model via logistic regression
• Generate entertaining shot descriptions automatically