There’s nothing in racing quite like the size of the crowd on Kentucky Derby day. Last year, 158,070 people packed Churchill Downs to watch Always Dreaming win the Run for the Roses, and on Saturday, despite the rain, nearly as many - 157,813 - were there to see Justify’s magnificent performance. As an indication of wagering activity across the nation, however, these numbers are merely the tip of an enormous iceberg.
Of course, the Derby-day crowd is an aberration - far from the norm for a day at the races in the U.S. - yet it still represents a fraction of the total number of people who watch and wager on the race.
In fact, the on-track live attendance for any regular race day is only ten percent of the total that watches and wagers on horse racing. The other 90 percent are at remote locations, which could include OTB (Off-Track Betting) parlors or other racetracks, or via sports/racebook or wagers that are placed on a phone or a computer through an ADW (Advanced Deposit Wagering accounts).
Given that 90 percent of the wagering activity on horses is placed at off-track locations, it would be reasonable to assume that the post times of the individual races at the 34 Thoroughbred/Quarter Horse tracks that raced on Saturday were coordinated to maximize total wagering activity and customer satisfaction.
Frustrated customers
In 2017, 37,628 Thoroughbred races were run in the U.S. But the U.S. racing industry has no coordinated scheduling, and it is costing the tracks and purses significant revenue.
It frustrates our best customers when an important stakes race in New York goes off in close proximity to an important race in Florida, Kentucky or California.
The post times are set by the individual tracks, and not much discipline goes into maintaining the established post times throughout the day.
Tracks throughout Ireland, England and Europe generally identify the individual races by the off time of the race rather than the race number. So, for example, on Saturday at Goodwood in England, the races were the 1:35, the 2:05, the 2:40, the 3:15, the 3:50, the 4:25, the 5:00, and the 5:35. The races are more closely defined by the off time than they are by the race number.
The race stewards and racing officials pay very close attention to getting the races off on the proper designated post time. As a result, the simulcasting of the British, Irish and French races is much better organized and maximizes revenue better than simulcast races in the U.S.
A possible solution
Fortunately, a possible solution is being looked at by the racing industry in the United States.
In 2011, the Jockey Club commissioned McKinsey & Company to do a comprehensive economic study of Thoroughbred racing called Driving Sustainable Growth for Racing and Breeding. A number of important initiatives have been developed from this study.
At the Jockey Club Round Table meeting in Saratoga last August, Ben Vonwiller, partner and leader of McKinsey’s Global Media & Entertainment and Professional Sports Practices, presented an engaging report, Better Race Scheduling Using Big Data.
As Vonwiller explained, McKinsey had done a comprehensive review using big data and advanced analytic techniques for the National Football League to organize the most efficient and profitable schedule. The NFL has 32 teams that are organized in two conferences of four divisions each, and each team plays 16 regular season games in a 17-week period. The games are primarily played on Sunday afternoons, but they have productive revenue-generating additional slots on Sunday, Monday and Thursday nights. The new schedule was a resounding success.
Based on the McKinsey report, the executives of the Jockey Club concluded there was substantial handle/revenue opportunity for the American racing industry to consider adapting the McKinsey analytics and modeling to and industry simulcasting analysis.
The McKinsey team, working with the U.S. racetracks and Jockey Club executives, set out to do this analysis. Clearly, the most significant issue was the amount of overlap that existed given the number of tracks running at any one time and the number of races on the card. However, the data clearly indicated that there were efficiency opportunities in the scheduling.
In order to confirm that the assumptions were achievable in the real world, the McKinsey team spoke with racing secretaries and racing office personnel, track management, broadcasters and other industry participants to get their reactions to what their model was telling them.
Industry-wide cooperation the key
In sum, the business plan has to include an algorithm to achieve scheduling optimization, a commitment to maintain post-time integrity and someone to function in a master scheduling role.
Clearly the level of industry-wide cooperation will determine the potential handle gain, which McKinsey estimates to be about $400 million. This would be a huge gain for the U.S. Thoroughbred industry, as it represents approximately a four percent increase in annual handle.
The Spring/Summer racing season is well underway, with over 30 tracks running daily over the weekends. This would provide a robust pool of data for the racetracks, McKinsey & Co, and the Jockey Club to work together in evaluating and testing the modelling and assumptions that Vonwiller outlined in his Round Table presentation last summer.
Jockey Club President and COO Jim Gagliano is enthusiastic about this project. “Ben Vonwiller of McKinsey did an excellent job in his presentation on this subject last year,” Gagliano said. “Obviously, any advancement of the concepts that Ben discussed requires a degree of cooperation and communication among the principals for this issue - the racetracks. Our role is only to facilitate the process, which we are pleased to do.”
Management commitment
I would like to add a few further thoughts regarding this important initiative based on my experience as a racetrack operator at the New York Racing Association and also as a wagering customer.
There are a large number of moving parts and racing participants, human and equine, that are required to put on a successful race card. Establishing post times and getting races to go on time has to be a priority.
The horses need to be saddled, do a full turn in the walking ring and hit the track with ten minutes to post so that the live track and simulcast customers can see the horses in the post parade.
Any number of uncontrollable events can impact the post time of a race: a horse flips and injures himself in the paddock, a horse gets loose on the track, equipment malfunctions, a horse flips in the gate, a steward’s scratch, unforeseen weather developments, etc. However, there has to be a strong management commitment to maintaining the integrity of the post time.
Track management, racing participants and racing officials have a fiduciary responsibility to the betting customers and we need to protect their wagers as if it was our own money. Inevitably, there will be some circumstances resulting in a post-time delay, but that has to stop with the race in question and not cascade through the race card.
Finally, I still very much enjoy a day at the races betting on the NYRA tracks and one or two simulcast tracks. To bet on three tracks in a day requires proper preparation and planning bet management. I strongly believe that NYRA’s Martin Panza runs the races consistently right on their published post times. I can set my watch by the off time. This is good for both NYRA and the betting customers.
Conversely, post times at Gulfstream Park seem not to have any relationship to reality. It appears that every race, the horses enter the track with six-to-nine minutes to post. Unfortunately, when a Gulfstream Park race gets to zero minutes to post, it is consistently five-to-seven minutes from that point until the race goes off. This has introduced a new term in to the racing vernacular - post-time dragging.
Rest assured, I do not believe that this is a one-off occurrence but appears to be a consistent management practice. I presume that there is an economic advantage for Gulfstream in this, but it certainly negatively affects the customer betting on Gulfstream and also wanting to bet on other tracks.
Disorganized structure
In sum, having no mechanism to coordinate the scheduling of 36,000-plus Thoroughbred races that will be run in 2018 is unconscionable.
The more important simulcasting becomes for racetracks the more important race coordination and scheduling will become. Tracks will continue to lose considerable revenues, purses will receive less from the wagering and, most importantly, the betting customer will be poorly served by the current disorganized structure.
I applaud the effort that McKinsey, the Jockey Club and most importantly the racetracks are making to address this serious problem.