Dynamic Graph-Based Forecasts of Bookmakers' Odds in Professional Tennis

2025-08-21

AI Summary

Professional tennis betting odds exhibit high predictive accuracy but suffer from significant publication latency, hindering pre-match analytical applications. To address this, we propose the first dynamic graph neural network (GNN)-based framework for pre-match odds prediction. Our approach models players, tournaments, and historical odds as a temporally evolving graph, integrating heterogeneous multi-source data—including head-to-head records, surface-specific performance, and real-time odds streams—to enable cross-surface and cross-tournament generalization. Unlike conventional methods relying on static feature engineering, our framework explicitly captures temporal relational dynamics and is the first to embed GNNs into the odds generation process. Evaluated on major 2024–2025 Grand Slam tournaments, it achieves strong alignment with ground-truth odds (MAE < 0.08), substantially outperforming baselines such as ATP rankings. The framework delivers an interpretable, production-ready tool for quantitative tennis analysis.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Knowledge Representation and Reasoning: Preferences
• Reasoning under Uncertainty: Graphical Models

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Bookmakers' odds consistently provide one of the most accurate methods for predicting the results of professional tennis matches. However, these odds usually only become available shortly before a match takes place, limiting their usefulness as an analysis tool. To ameliorate this issue, we introduce a novel dynamic graph-based model which aims to forecast bookmaker odds for any match on any surface, allowing effective and detailed pre-tournament predictions to be made. By leveraging the high-quality information contained in the odds, our model can keep pace with new innovations in tennis modelling. By analysing major tennis championships from 2024 and 2025, we show that our model achieves comparable accuracy both to the bookmakers and other models in the literature, while significantly outperforming rankings-based predictions.

Problem

• Forecasting bookmakers' odds before tennis matches availability
• Providing pre-tournament predictions across different court surfaces
• Overcoming limitations of rankings-based tennis match predictions

Innovation

• Dynamic graph model forecasts bookmaker odds
• Leverages odds information for pre-tournament predictions
• Achieves comparable accuracy to bookmakers and models