Steven S. Vickner, PhD, Associate Professor and Steve Koch, Executive Director, Safety & Integrity Alliance, NTRA
Using data for 1,515 races over 165 race days from the 2011 Thoroughbred racing season at the Woodbine Racetrack, variability in all sources handle per race was explained as a function of field size, field quality and race conditions, race distance, surface, weather, and an array of seasonality control variables. Several robust and actionable insights emerged:
- Field size proved again to be the leading determinant of per race handle. However, each additional runner in the starting gate provided less wagering increase than the runner loaded just prior; this is diminishing marginal returns per entry (Figure 1).
- Restricted race conditions (i.e., race entries restricted to Ontario-Sired eligible runners) had a negative impact on handle. However, the larger average field sizes in those races did offset this effect consistent with prior studies.
- Diminishing marginal returns from field size and the negative effect of restricted races inspire the concept of return-on-entry (ROE). ROE can be evaluated where any runner in a given race represents an opportunity cost given that entrant may have otherwise filled another race at enhanced value to wagering (i.e., add to a smaller field and/or convert to an unrestricted race entry).
- Handle exhibited positive yet diminishing returns to race quality where outlier major stakes race events are controlled for separately in the econometric model.
- Races taken ‘off-the-turf’ suffer increased wagering penalties in addition to the impacts of reduced field size due to scratches.
- Finally, the data harbors a number of incidents where races aired on a ‘tape delay’ (i.e., not live) basis through the major distributing television/wagering network. These incidents proved harmful to total wagering.
A battery of empirical methods was tested with quite similar and robust results. A multilevel mixed effects modeling approach allowed for inferences beyond the 165 races herein and addresses higher-order autocorrelation across races within each race day. This latter point we dub ‘wagering inertia’ indicated handle in nearby races (i.e., race two and race one, or race three and race one) was not surprisingly correlated but handle in more distant races (i.e., race seven and race one) was not. On average 80 percent of the variability in handle per race day was explained by the model with highly statistically significant parameter estimates on not only key explanatory variables but also the usual seasonality and other control variables. Given only one race season is considered at one track, a hedonic specification is appropriate.
Research access to customer and track-level wagering data, horse racing performance data, and related sports betting data enables broader and deeper insights for the long-term benefit of the horse industry.
Koch, S. ‘Econometric models of all sources, HMA and export wagering on the Woodbine 2011 thoroughbred race product’, March 2, 2015.
Koch, S. ‘Declining field size: a global issue’, 2014 Global Symposium on Racing and Gaming, Tucson, Arizona, December, 2014.
Smith, M.D. (2001) ‘Breeding incentive programmes and demand for California thoroughbred racing: is there a quality/quantity tradeoff?’ Applied Economics, Vol. 33, pp. 1755-1762.
Dr. Steven S. Vickner is an economist and associate professor in the Equine Industry Program, College of Business, University of Louisville. He earned both his MS and PhD in agricultural and resource economics from Colorado State University, an MBA in management from the University of Denver and a bachelor’s degree in economics from Bowling Green State University. His research focuses on the role of information and technology on economic decision-making, and he has done years of extensive work studying yearling thoroughbred prices. For more information, contact Vickner at (502) 852-4888 or firstname.lastname@example.org.
Steve Koch is the Executive Director of the Safety & Integrity Alliance at the National Thoroughbred Racing Association. He previously was the vice president of racing at Woodbine Racetrack in Toronto, Canada. Prior to Woodbine, Koch worked at the Keeneland Association and he was raised as a horseman at Claiborne Farm. He earned both his BS and MS in agricultural economics from the University of Kentucky. For more information, contact Koch at email@example.com or www.linkedin.com/in/kochstephen.
The Equine Industry Program at UofL is the only accredited undergraduate business degree in the world with an equine focus. Graduates can be found in all aspects of the industry, from training to broadcasting to financial analysis.
The Analytics Brief series is published by the Equine Industry Program, College of Business, University of Louisville as a part of its ongoing research program and is intended to address contemporary economic, business, and legal/regulatory topics in the broadly defined equine industry, including but not limited to, pari-mutuel horse racing, wagering, gaming, auctions, breeding, other equestrian sports, equine-related tourism, and all relevant supply chains.
Analytics Brief 2018.2 (.pdf)