Interpretable Low-Dimensional Modeling of Spatiotemporal Agent States for Decision Making in Football Tactics

2025-06-20

AI Summary

To address the high computational cost, poor interpretability, and strong data dependency of existing approaches in football tactical modeling, this paper proposes an interpretable, low-dimensional regularized modeling framework. Methodologically, it introduces the first integration of expert-knowledge-driven state variables—such as spatial scoring, relative distance, and motion dynamics—with a lightweight XGBoost classifier to model tactical interactions grounded in player–ball spatial relationships and dominance metrics. The key contribution lies in overcoming the black-box limitations of conventional reinforcement learning models and the prohibitive computational overhead of physics-based simulations. Evaluated on multi-source data from LaLiga 2023/24 (StatsBomb and SkillCorner), the framework identifies distance to the ball and spatial scoring as the two most critical predictors of pass success. It achieves high predictive accuracy while ensuring strong model interpretability and actionable insights for real-world tactical analysis.

Technology Category

• Reasoning under Uncertainty: Relational Probabilistic Models
• Machine Learning: Calibration & Uncertainty Quantification
• Intelligent Robots: State Estimation

Application Category

• User Modeling, Personalization and Recommendation: Metrics for user behavior and evaluating success
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Understanding football tactics is crucial for managers and analysts. Previous research has proposed models based on spatial and kinematic equations, but these are computationally expensive. Also, Reinforcement learning approaches use player positions and velocities but lack interpretability and require large datasets. Rule-based models align with expert knowledge but have not fully considered all players'states. This study explores whether low-dimensional, rule-based models using spatiotemporal data can effectively capture football tactics. Our approach defines interpretable state variables for both the ball-holder and potential pass receivers, based on criteria that explore options like passing. Through discussions with a manager, we identified key variables representing the game state. We then used StatsBomb event data and SkillCorner tracking data from the 2023$/$24 LaLiga season to train an XGBoost model to predict pass success. The analysis revealed that the distance between the player and the ball, as well as the player's space score, were key factors in determining successful passes. Our interpretable low-dimensional modeling facilitates tactical analysis through the use of intuitive variables and provides practical value as a tool to support decision-making in football.

Problem

• Develop interpretable low-dimensional football tactical models
• Overcome computational cost and lack of interpretability in existing methods
• Identify key spatiotemporal variables for effective pass prediction

Innovation

• Low-dimensional rule-based modeling for football tactics
• Interpretable state variables for ball-holder and receivers
• XGBoost model predicts pass success using key factors