XGBoost Learning of Dynamic Wager Placement for In-Play Betting on an Agent-Based Model of a Sports Betting Exchange

TL;DR

This work addresses learning profitable dynamic in-play betting strategies from synthetic data generated by an agent-based sports betting exchange (BBE). It integrates XGBoost as a betting agent trained on the most profitable actions of a heterogeneous pool of baseline bettor strategies, then tests the learned strategy within the same ABM. Results demonstrate that the XGBoost-based bettor not only learns profitable patterns but also generalizes to outperform the training strategies across scenarios, validated by non-parametric hypothesis tests. The authors provide an open-source release of the extended BBE with XGBoost integration to promote further exploration of adaptive, data-driven wagering strategies in real-time markets.

Abstract

We present first results from the use of XGBoost, a highly effective machine learning (ML) method, within the Bristol Betting Exchange (BBE), an open-source agent-based model (ABM) designed to simulate a contemporary sports-betting exchange with in-play betting during track-racing events such as horse races. We use the BBE ABM and its array of minimally-simple bettor-agents as a synthetic data generator which feeds into our XGBoost ML system, with the intention that XGBoost discovers profitable dynamic betting strategies by learning from the more profitable bets made by the BBE bettor-agents. After this XGBoost training, which results in one or more decision trees, a bettor-agent with a betting strategy determined by the XGBoost-learned decision tree(s) is added to the BBE ABM and made to bet on a sequence of races under various conditions and betting-market scenarios, with profitability serving as the primary metric of comparison and evaluation. Our initial findings presented here show that XGBoost trained in this way can indeed learn profitable betting strategies, and can generalise to learn strategies that outperform each of the set of strategies used for creation of the training data. To foster further research and enhancements, the complete version of our extended BBE, including the XGBoost integration, has been made freely available as an open-source release on GitHub.

Figures (18)

• Figure 1: High-level overview of the experiment and the data flow of the system.
• Figure 2: Box-plot of the influence of XGBoost hyperparameter eta on the mean test score.
• Figure 3: Box-plot of the influence of XGBoost hyperparameter max_depth on the mean test score.
• Figure 4: Box-plot of the influence of XGBoost hyperparameter colsample_bytree on the mean test score.
• Figure 5: Box-plots illustrating the influence of subsample (left) and gamma (right) on the mean test score.