The Decade Double

Here’s a Daily Racing Form headline from 2024 (yes, the future):

North American Racing Handle Doubles Over the Last Decade

Do you find this headline completely unbelievable? It shouldn’t be. Let me ask this: how much would betting handle have to grow year-over-year for 10 years for that headline to be true? It’s not large – it’s only a 7.2% growth rate, compounded annually. In terms of growth above normal economic growth, it’s only a 3-4% adder to normal national growth trends.

Doubling handle would mean that contributions to track earnings and purses would also double during that period. (Neither earnings nor purses would double, since those are supported now by other sources like admission, concessions, and slots) I think most observers, seeing that handle was at an historic high, would no longer say that “horse racing is dead” but that racing was as good as it had been in 30-40 years.

Now, here’s two alternative beginnings to the article that accompany the headline. Which do you find more plausible?


1. Industry officials celebrated the 10th consecutive year of handle growth, noting that wagering on thoroughbred racing has doubled since the US marked the unofficial end of the Great Recession in 2014. Attendance and off-track wagering both doubled, track revenues increased 80%, and purse accounts increased by 60%. The purse account increase reflects that purse subsidies from other sources (racino/slots revenue, sales, supplemental fees) remained flat during this time. Tracks and horsemen used the windfall to increase races by 35% while the average purse went up 18%. Breeding finally reversed a two-decade long decline as the 2023 foal crop of 40,000 returned to levels not seen since 1991.

Most track officials credited their marketing and promotional efforts to get fans back to the track as the main source of success, but acknowledged that Jess’s Dream – the first foal of popular 2009 Horse of the Year Rachel Alexandra – winning the Triple Crown in 2015 kick-started their efforts. When “Taco” came back to race in 2016 and dueled in a cross-country campaign with the late-developing Cozmic One (Zenyatta’s first foal), the Breeders Cup Classic at Belmont Park featuring their final duel (won in a late nose by Cozmic One) set betting and ratings records for a non-Triple Crown race and energized the sport…


2. Industry officials acknowledged that the 2014 “Decade Double” initiative pioneered by The Jockey Club, NTRA, and a consortium of racetracks and other racing industry groups has met their goal of doubling betting handle on North American races in 10 years. The Decade Double initiative began with the premise that the $23 billion target for wagering on throughbreds would represent an all-time, inflation-adjusted, high indicator of interest in the sport. The leaders of the “Decade Double” campaign credit its focus on customers and getting buy-in from tracks and horsemen on how to share gains.

” We knew that the sport couldn’t grow without customer support,” said Jeff Gural, Jockey Club board member and head of the Decade Double Initiative. “Significant gains had to be realized by the customer – the bettor – and ultimately that meant lowering the price of betting on racing.”

“Working with our horsemen and tracks, we concluded – and believe me, it was a tense fight at times –  that bettors needed to see the lion’s share of gains, with tracks and horsemen splitting the rest. We settled on a 40/30/30 split, and that’s when efforts to reduce takeout by 40% began.”

This year, the average on-track takeout for a Win bet was 10%, which horseplayer’s Decade Double representative Andy Asaro noted was “much nearer betting the LA Jaguars and the points in the Super Bowl.” Exotics averaged 12-14%; in 2014, however, the typical takeout on exacta or trifecta pools was 20-25%.

The Decade Double and industry groups like NYRA and the CHRB aggressively promoted the takeout decreases, at first in hopes to keep track revenues and purse accounts level. Most groups acknowledge that the success was unexpected: track revenues have increased by 48% and total purses by 36%, despite the lower takeout. Racing days and total races have remained flat in response to a lower profile DD initiative meant to prop up field size in response to low foal crops. Even those have since recovered to a “healthy level” of 35,000, what many breeders consider sustainable at this level of betting…


It truly is amazing how growth can positively impact everyone while stagnation leads to tribalism and in-fighting and decision-making based on the fear of loss as opposed to the hope of gain. That’s unfortunately where horse racing is today.

Article 2, even if the numbers aren’t exact, shows that broad-based gains are possible if they accompany a plan and a target for growth. If we collectively bet $20B on racing, no one could rightly claim that racing was dead. It is, however, hard to envision that future if customers do not share in those gains. And again, the numbers are not daunting:

  • To double in volume, handle needs to increase by 7.2% a year.
  • To decrease takeout by 40% over 10 years, takeout needs to decrease by 5% a year.

The key, of course, is to offset the short-term revenue decrease from pricing with 2 other Ps of marketing.

  • Promote the heck out of the sport emphasizing lower prices (and other promotions)
  • Product quality needs to stay high /  improve (larger fields, showcase racing days, etc.)

The time element is the hardest part, because it’s not an overnight fix. Nothing worth doing ever is.

Racing’s Target Segments, Part 3 – Handicapping as a Serious Pursuit

Parts 1 and 2 found here.

In part 3, I’m going to concentrate on what I believe to be racing’s most important segment – the serious handicappers and professionals that make up the vast majority of wagering handle in North America. They are the segment that supports the sport outside of its biggest days, betting millions on races Sunday-Friday somewhat in the name of fun but more as an opportunity to make money by applying their intellect to the game.

This segment has significant overlap with the sports betting market generally, though that market is technically illegal outside of Nevada. Horse racing remains unique in that it can be bet without being present at the track, typically online through an Advanced Deposit Wagering service or at a simulcast outlet. Most readers of this site  – mostly the horseplayer community on Twitter – are familiar with the basics of how this works. This segment truly values one thing – a positive expectation bet (or the perception therein). Handicappers call these opportunities overlays.

In order to increase interest from the serious/professional gambling segment, horse racing must increase the number of overlays, or positive expectation bets, in its races and racing cards that attract this class of bettors.

The mechanisms for creating more overlays in racing are well-known but largely boils down to increasing the size of betting pools, increasing the number of combinations that can be played, and increasing the payoffs of winning bets. Several factors can go into this, but there are three main ones with pari-mutuel betting:

  • Increase play from casual gamblers / non-handicappers
  • Lower the price of betting (the takeout rate)
  • Increase field size

Having “less informed” money bet into the pools – from casual gamblers – was largely the subject of the first two posts, but it helps explain why bigger race days have better betting opportunities. It also explains the growth of poker to a degree – the early days of the boom, pros had a field day with casual players coming in. When casual fans “play numbers”, play “names”, “bet favorites”, “bet longshots”, “box superfectas” – those benefit the dedicated players by increasing the odds on everyone else.

Which leaves takeout and field size. The digital literature on why field size and takeout rate and their impact on payouts/overlays is voluminous, so I’ll try not to belabor the point too much. My thoughts on takeout can be found here:

Takeout – The Price Isn’t Right: Part 1 | Part 2 | Part 3

To sum up, takeout decreases will naturally increase handle thru the churn mechanism but most likely will not increase track revenue because of how the takeout is shared amongst the track, bet-takers, and purse accounts. To truly drive gains, handle (price) decreases must be tied to a strong marketing plan that includes promotion on and off-track and alignment with horsemen. There is, however, a takeout rate (below 12%, I believe) where parimutuel bets on racing become competitive with sportsbooks. This price level will attract more large sports bettors (whales) to the pools and racing will see handle gains greater than that produced by churn.

Field Size

Field size is equally, if not more important, to serious horseplayers in the search for overlays. Having more horses in a field allows the handicapper more data points to seriously consider (or dismiss) in search of a winner. Consider, for example, what happens when a 20-1 shot scratches from a race at the gate. If the handicapper gave the 20-1 horse (public gives 5% chance of winning), only a 1% shot (true odds of 99-1) to win the race, then losing that horse is effectively a 4% point increase in takeout for the bettor.

It works in reverse as well – full fields are a tremendous place for a handicapper to find value as, if she can identify a number horses that are overbet, the bet she makes has a positive expectation. It’s why I especially love the Kentucky Derby; the combination of a 20-horse field and, since Mine that Bird’s win, a betting public that won’t let any horse go above 50-1 again. You only need eliminate 6 horses from contention to make the Derby win pool a positive expectation bet.

The Average Field Size Metric is Misleading, Especially for Horizontal Players

Many horseplayers find their value by betting into horizontal pools like the Pick 3, 4, 5 and 6. The allure to these bets, versus straight betting and parlaying the results, is that takeout only happens once. More often than not, the horizontal return exceeds that of a similar parlay despite being, theoretically, of similar difficulty to hit.

Tracks often tout average field size (total starters / total races) as an indicator of the quality of their fields for betting purposes. And it is, but doesn’t tell the whole story. Take, for example, two theoretical racetracks that boast an average field size of 8.0. One track got that AFS score by carding two competitive 8-horse claiming races while the other paired an 11-horse turf race and a 5-horse main track clunker. One has 64 possible double combos but, for the same AFS, the second track has only 55 – 14% less. Almost all horizontal players will tell you the presence of multiple small fields in a sequence will depress payouts – this is why.

Is there a metric that can quantify the combo-killing nature of regular small fields? Yes. If we take the square root of the above scenario, we see that the two 8 horse races contribute an average of 8 combos to the double whereas the 11-5 double only contributes 7.4 on average. It’s easy to calculate this across multiple sequences as well – take these identical 8.0 AFS cards of 8 races:

Average Field Contribution to Horizontals (AFCH)

For any X number of races, multiply the field sizes together and take the Xth root [Field Size Product ^(1/X)] to come up with AFCH. For example:

Race 1 2 3 4 5 6 7 8 AFS AFCH
Track 1 8 9 6 8 7 10 7 9 8.0 7.9
Track 2 6 10 4 8 12 8 5 11 8.0 7.5

So, here we have a metric that allows to not just look at the number of horses run but also look at how the cards are constructed to produce value for the horseplayer.

The important thing here is that AFCH is probably more important for serious horseplayers in the search of overlays at a track than just field size. Since spreading tickets is a popular horizontal strategy that introduces overlays in those pools, minimizing the number of “free squares” is distinctly to the track’s advantage.

Conclusion

Thus wraps up 4,000 words on the target segments of the horse racing market that used “Racing as Entertainment” as a starting point and veered into other discussions where that “meme” didn’t fit as well. Thanks for reading.