Better Prevent than Tackle: Valuing Defense in Soccer Based on Graph Neural Networks

TL;DR

This paper introduces DEFCON, a graph neural network-based framework that quantifies defensive contributions in soccer by evaluating how defenders influence the expected possession value, addressing the challenge of measuring non-visible defensive actions.

Contribution

We propose DEFCON, a novel GNN-based method that accurately assesses player-level defensive impact in soccer, including preemptive actions not captured by traditional metrics.

Findings

DEFCON's defender credits correlate strongly with market valuations.

The framework enables detailed in-game defensive timelines and spatial analyses.

DEFCON effectively quantifies the impact of defenders on attacking opportunities.

Abstract

Evaluating defensive performance in soccer remains challenging, as effective defending is often expressed not through visible on-ball actions such as interceptions and tackles, but through preventing dangerous opportunities before they arise. Existing approaches have largely focused on valuing on-ball actions, leaving much of defenders' true impact unmeasured. To address this gap, we propose DEFCON (DEFensive CONtribution evaluator), a comprehensive framework that quantifies player-level defensive contributions for every attacking situation in soccer. Leveraging Graph Attention Networks, DEFCON estimates the success probability and expected value of each attacking option, along with each defender's responsibility for stopping it. These components yield an Expected Possession Value (EPV) for the attacking team before and after each action, and DEFCON assigns positive or negative credits to defenders according to whether they reduced or increased the opponent's EPV. Trained on 2023-24 and evaluated on 2024-25 Eredivisie event and tracking data, DEFCON's aggregated player credits exhibit strong positive correlations with market valuations. Finally, we showcase several practical applications, including in-game timelines of defensive contributions, spatial analyses across pitch zones, and pairwise summaries of attacker-defender interactions.