Machine learning the harness track: Crowdsourcing and varying race history

Racing prediction schemes have been with mankind a long time. From following crowd wisdom and betting on favorites to mathematical methods like the Dr. Z System, we introduce a different class of prediction system, the S&C Racing system that derives from machine learning. We demonstrate the S&C Racing system using Support Vector Regression (SVR) to predict finishes and analyzed it on fifteen months of harness racing data from Northfield Park, Ohio. We found that within the domain of harness racing, our system outperforms crowds and Dr. Z Bettors in returns per dollar wagered on seven of the most frequently used wagers: Win $1.08 return, Place $2.30, Show $2.55, Exacta $19.24, Quiniela $18.93, Trifecta $3.56 and Trifecta Box $21.05. Furthermore, we also analyzed a range of race histories and found that a four race history maximized system accuracy and payout. The implications of this work suggest that an informational inequality exists within the harness racing market that was exploited by S&C Racing. While interesting, the implications of machine learning in this domain show promise.

► We tested out system against bettor crowdsourcing, Dr. Z bettors and random chance.
► Our system outperformed Dr. Z Bettors and random chance in accuracy and payout.
► A four race history maximized the system's predictive ability.
► Our system outperformed all tested predictive methods for seven wagers types.