Blame is easier than praise: Measuring off-ball defensive performance in football

2026-06-18

AI Summary

Traditional football defensive evaluation has predominantly relied on on-ball actions such as tackles, overlooking the sustained contribution of off-ball positioning. This work proposes a multi-agent spatiotemporal trajectory-based attribution framework that quantifies player involvement through Defensive Pressure Areas (DPA) and constructs role-conditioned baselines using automatically identified team structures to attribute changes in expected threat to individual players. The method enables, for the first time, reliable measurement of off-ball defensive positioning errors. Validated across 516 matches spanning genders and leagues, the new metric outperforms existing state-of-the-art action-based metrics by approximately one standard deviation and shows strong correlation with external expert ratings and player market values.

Abstract

The defensive performance of football players is commonly measured through a limited number of actions like tackles and interceptions while their continuous impact through positional behaviour has hardly been studied before. We formulate this problem as an attribution over multi-agent spatiotemporal trajectories without player-level ground truth labels, where event-level changes of expected threat are distributed among individuals. We propose a framework that performs this attribution using player involvement scores calculated from defensive pressure areas (DPAs). By computing role-conditioned baselines within automatically detected team structures, we can determine each defender's expected responsibility for threat created through arbitrary passes. The validity and robustness of this approach are evaluated on a uniquely extensive cross-gender and cross-competition data set, including positional and event data from 64 matches of the men's World Cup, 116 matches of the women's German Bundesliga and 336 matches of the men's German 3. Liga. In the absence of a ground truth, we propose an evaluation protocol that combines multiple relatively weak proxies into robust summary scores. We find a validity score that is improved by around 1 standard deviation compared to the best action-based metric and demonstrate that many popular measures show limited validity. The "blame" for conceding high-value actions shows especially strong correlations with external ratings and market values, making it the first published metric in football to reliably measure positioning errors. All code underlying this work is publicly available to support reproducibility and further research.

Problem

• defensive performance
• off-ball
• player positioning
• attribution
• expected threat

Innovation

• defensive pressure areas
• expected threat attribution
• multi-agent spatiotemporal trajectories
• role-conditioned baselines
• unsupervised defensive evaluation