Hybrid models improve live tennis match winner predictions
Forecasting the Winner of a Live Tennis Match
Machine Learning
Summary
Predicting who will win a tennis match as it happens is tricky because the game changes with every point. This paper looks at how to best combine information from before the match and data collected live during the match to estimate each player's chances of winning. Using data from over 8,000 Grand Slam matches, the authors tested several prediction models. Their hybrid model called Trace was better as the match went on, reaching nearly 88% accuracy when three quarters of the match was complete. This suggests combining different types of information helps provide more accurate live predictions.
live sports bettingwin probabilitytennis forecastinghybrid modelGrand Slam matchesmatch progresschronological splitmodel accuracyperformance data
Authors
Charles Xie, Aneesh Muppidi
Abstract
With the rise of live sports betting in recent years, tennis forecasting has expanded from pre-match prediction to models that update win probabilities as a match unfolds. A central challenge in creating such a model is the constant need for models to adapt to score and performance changes. This study examines how pre-match and live information can be most effectively integrated into a model to produce accurate win-probability estimates. The analysis uses 8,222 Grand Slam matches containing a total of 1,505,355 points. Five models were evaluated using a chronological split, with matches from 2011-2021 used for training, 2022 for validation, and 2023-2024 for testing. Trace, a hybrid model, achieved accuracies of 76.06%, 82.15%, and 88.34% at 25%, 50%, and 75% match progress, suggesting that hybrid modeling is a practical approach to live tennis forecasting.