ScoutGPT: Capturing Player Impact from Team Action Sequences Using GPT-Based Framework

2025-12-19

AI Summary

Predicting football transfer success is challenging because player performance is highly contingent on dynamic tactical contexts and teammate coordination, whereas existing methods rely on static statistics or ex-post valuations, failing to model cross-context value transfer. This paper proposes the first counterfactual-simulation-capable, player-conditioned event prediction model: it formalizes matches as spatiotemporal action sequences annotated with residual On-Ball Value (rOBV), and employs a GPT-style autoregressive Transformer to jointly predict action type, location, timing, and rOBV. We introduce player-specific embeddings and a cross-cohort counterfactual injection mechanism to quantify individual adaptability and value portability across tactical systems. Evaluated on five seasons of Premier League event data, our model significantly outperforms state-of-the-art sequential baselines, achieving simultaneous gains in spatial precision and predictive accuracy. It enables practical applications including forward-role substitution identification and cross-system performance attribution analysis.

Technology Category

• Humans and AI: Human-Aware Planning and Behavior Prediction
• Game Theory and Economic Paradigms: Adversarial Learning
• Multiagent Systems: Adversarial Agents

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Transfers play a pivotal role in shaping a football club's success, yet forecasting whether a transfer will succeed remains difficult due to the strong context-dependence of on-field performance. Existing evaluation practices often rely on static summary statistics or post-hoc value models, which fail to capture how a player's contribution adapts to a new tactical environment or different teammates. To address this gap, we introduce EventGPT, a player-conditioned, value-aware next-event prediction model built on a GPT-style autoregressive transformer. Our model treats match play as a sequence of discrete tokens, jointly learning to predict the next on-ball action's type, location, timing, and its estimated residual On-Ball Value (rOBV) based on the preceding context and player identity. A key contribution of this framework is the ability to perform counterfactual simulations. By substituting learned player embeddings into new event sequences, we can simulate how a player's behavioral distribution and value profile would change when placed in a different team or tactical structure. Evaluated on five seasons of Premier League event data, EventGPT outperforms existing sequence-based baselines in next-event prediction accuracy and spatial precision. Furthermore, we demonstrate the model's practical utility for transfer analysis through case studies-such as comparing striker performance across different systems and identifying stylistic replacements for specific roles-showing that our approach provides a principled method for evaluating transfer fit.

Problem

• Predicting football transfer success using player performance context
• Capturing player adaptation to new tactical environments and teammates
• Simulating counterfactual player impact in different team structures

Innovation

• GPT-based model predicts next event and value
• Counterfactual simulations assess player-team fit
• Learns player embeddings for tactical adaptation analysis