Machine Learning-Based Prediction and Quantification of OCD Surgery and Pedigree Effects on Racehorse Performance.

OCD surgery does not significantly impair long-term race performance; pedigree is the strongest predictor.

This retrospective study investigated the impact of osteochondritis dissecans (OCD) surgery on the race performance of Thoroughbred racehorses and evaluated the predictive power of pedigree and management variables using machine learning techniques. Data were collected from 75 Thoroughbreds that underwent OCD surgery between 2015 and 2017 and 257 maternal siblings without recorded OCD surgery (controls). The study included variables such as biometric, pedigree, surgical, and race performance measures. Additional predictors, including earnings per start and sale-price ratios, were derived, and racing performance was quantified using the field-adjusted percentile metric 'race_pts_avg'. Gradient-boosting models (XGBoost and CatBoost) were trained, and predictive performance was evaluated using R² scores and Shapley Additive Explanations (SHAP). Models incorporating derived features consistently outperformed those without, with the best model (CatBoost) achieving an R² of 0.7983. Variables related to surgical history, including age at surgery, showed limited predictive value, and lesion severity did not rank among the dominant predictors. In contrast, pedigree, particularly the group-mean encoded family identifier 'h_family', ranked highest. These findings suggest that OCD surgery does not significantly impair long-term race performance, while pedigree is the strongest predictor in this cohort. Although this observational study is limited by its regional scope and unmeasured factors that may influence both surgery status and performance, the results provide objective context for breeding, treatment, and sales decisions involving Thoroughbreds with a history of OCD surgery.

This summary was distilled by AI and may occasionally misinterpret data. Confirm critical details with the primary literature before clinical application.