STUART S. JANNEY III: Thank you, Lindsay. That was a wonderful presentation.
Computer-assisted wagering is a complex issue in our sport. Pat Cummings of the Thoroughbred Idea Foundation is joining us as moderator of a panel with Professor Marshall Gramm of Rhodes College, and NYRA executive Joe Longo.
Gentlemen, thank you.
PAT CUMMINGS: Thank you, Stuart, and I greatly appreciate the invitation to be here with you today.
At the 1974 Round Table horse racing’s greatest advantage was addressed in the opening remarks of then Monmouth president, Phil Iselin. Horse racing, Iselin said, “is not just a spectator sport, but rather one in which the fan participates on a high level. The horse a fan bets on belongs in some small measure to the better by virtue of the bet.”
I think Lindsay captured that pretty well, as well. Today the competitive landscape for wagering is far different, but the connection between racing and its fans, the bettors, is fierce.
The history of wagering on U.S. racing is one filled with stops, starts, and seemingly endless change. It all got kicked off when Leonard Jerome imported four pari-mutuel machines from France at his Jerome Park. In 1871, America got its first taste of pari-mutuel wagering. The minimum bet, $5, the equivalent of about $125 today.
So, keep that in mind the next superfecta you put in.
Pari-mutuel wagering took hold in the 1930s as it became clear that guaranteed percentage of takeout from each bet was an economic driver for all involved. By the 1980s, off-track was launched in New York, and the simulcasting era was truly under way.
Congress passed the Interstate Horseracing Act in 1978, facilitating simulcasting’s growth. Truly landmark legislation at the time, though it did face some fierce resistance from many in the sport at that time.
Racing was an industry confronting transition. In 1983, just one month after this multi-day piece from Steve Crist in the New York Times, a New York legislative task force sought to address this period of change. Its chairman, senator John Dunne, asserted, “If we can’t bring more people to the track, then we must bring the tracks to the people.”
Distribution of track signals expanded and eventually would reach Americans directly in their homes, and not just the pavilion, as it was here in 1987.
On-track betting actually has plummeted across those intervening years. It was cut in half from one-quarter of total handle to one-eight of total handle across seven years from 1996 to 2003.
In 2001, the four largest online waging providers at the time handled just $50 million. In two years it grew 10 times that to more than half a billion dollars.
But concurrent with the rise of internet account wagering for ordinary horseplayers was the growth of a segment of bettors leveraging this new connectivity. These customers are often referred to as CAWs, computer-assisted wagerers.
The CAWs are individuals with well-resourced staffs. They develop couture models to assess vast amounts of data. They deploy finely honed algorithms to efficiently place bets, often at the last possible second, and all while receiving significant rebates for their play.
In pari-mutuel wagering, betting late has always been beneficial. But when all betting was done on track, betting late came with a significant risk, a chance you might not get your bet down at all.
That’s not a concern now, especially for the CAWs, who have been given the ability to place batch bets, dumping vast amounts in the final cycle of betting, and the impact of that is witnessed every day in nearly every race, and across almost every pool.
While total wagering from CAWs has grown in recent years, wagering from all other customers has not. Here are the numbers: In 2003 CAWs were roughly 8% of total handle, but last year, probably about one-third of all betting on U.S. Thoroughbred racing.
When you adjust these figures for inflation, the concern becomes obvious. Handle from all non-CAW customers is down by nearly two-thirds in just two decades. The greatest source of growth has been a handful of high-frequency bettors.
So how the evolution of CAWs is managed and how we support our existing customers might be key to the future of wagering on racing in today’s tech-forward world, one with unfathomable processing power, machine learning, AI, and more.
To discuss this evolution in much greater detail, please join me in welcoming our panelists, Joe Longo and Dr. Marshall Gramm.
(Applause.)
So, Marshall, one of the most interesting people I know in the sport, and I’ve been involved for probably 20 years or so. An owner, a breeder, an academic researcher, and a very serious horseplayer. If you could, just give us a little overview of your entire involvement in racing.
MARSHALL GRAMM: Well, I’ve kept track of every bet I’ve ever made. I remember 1991 I made my first bet on a Thoroughbred racehorse. I drove up to Pimlico, made a bet, won it 7 to 2.
2010 was a big year for me. I bet 49 grand that year; I got $52,000 back; I made $3,000. I also owned my first racehorse. I claimed a horse for $5,000 at Philadelphia Park, so suddenly I was an owner and winning horseplayer. I was top of the world.
In 2011 I got rebates for the first time. It allowed me to be a winning player. By 2015, I had bet $25 million in one year, so was a 500-fold increase with a betting partner, increase in handle from me as a result of having an opportunity to win and the pricing being different.
And I went from one racehorse to now I own 113 racehorses in partnership or outright. I own 60 in training with 12 different trainers. Many of those trainers have less than 30 horses. I have 21 broodmares, and I stand a stallion just down the road at Irish Hill Century Farm.
PAT CUMMINGS: So that’s a pretty significant change.
(Applause.)
Yeah, I think so. But it all started with betting.
MARSHALL GRAMM: All started with betting. And now, again, we are talking about computer-assisted wagering. I do some computer-assisted wagering. I have a model, my partner and I have a model. We use that model to calculate probabilities based on historical data, based on real-time odds. We take those probabilities, we turn them into bets, and we bet at the last minute, and we go right through the front end.
So, a lot of the teams that you read about, right, they have their own TRA codes, they have direct access to the tote system, but there are a lot of people out there that can do computer betting right through the front end.
In part, this is not going away. We need to figure out how to make this work for every player.
PAT CUMMINGS: So, from the operator side, Joe is sitting in today for David O’Rourke, and we appreciate you coming on at last-minute notice, Joe.
You are the general manager of NYRA’s content management services business. For those here that don’t necessarily understand or aren’t familiar with that title and how it fits under NYRA, could you just walk us through that for a minute?
JOE LONGO: Yeah, so we started NYRA Content Management Solutions about five, six years ago with the idea of effectively becoming a content agent. We’re not the first in that space. There are some others.
What we look to do is increase distribution and get fair market rates for the tracks that we represent. So, we started with eight; we are up to 15 now. I like to joke that in addition to Tony Allevato, my boss, I have 15 other bosses. It’s been quite a ride.
PAT CUMMINGS: So, we’ll get into more of some of the things NYRA has done, and truly a real leader in computer-assisted wagering management over time.
I think it’s also worth noting that a few years ago The New York Racing Association did become a minority owner in the betting services provider called Elite Turf Club, which works directly with a lot of the CAW teams, so we’ll weave that in in a minute here.
Marshall, in the last five to 10 years, you said 2015 you were handling $25 million a year. What has changed since then? What have you witnessed from a betting standpoint? How has it affected your business and operation of your model and wagering in this really modern way?
MARSHALL GRAMM: Well, the market has become much more efficient, and so, not only as a result of the computer teams and the sophisticated bettors, the professional bettors, but we’ve also lost many casual players to sports betting, to casinos.
The proliferation of low-priced alternatives has pushed people out of our wagering pools. So, the increased inefficiency, right, has hurt a lot of mid-level horseplayers. We’ll probably only end up betting only $7.5 million this year. It’s become a lot tougher. The space is a lot tougher.
We are always updating our models, but so are the teams. These are always works in progress. It’s not like they turn on their computers and models and it just goes.
It takes updating, it takes work, and the market has become more efficient, the price has become more accurate reflections of a horse’s true chance of winning.
For the everyday horseplayer we’ve seen a lot more odds movements. Now, these odds movements, these late price movements, are something not new. In fact, in 2002, the industry took a real good look at odds movements. In fact, Churchill Downs and some other tracks went to cutting off pools at zero minutes to post. We still see that now, and it’s very frustrating, especially to new horseplayers used to sports betting, used to fixed odds. To bet a horse at 4 to 1 and see it go off at 8 to 5. Or to bet a horse and think they’re getting $22 in the exacta, and they’re getting $12 in the exacta.
Our middle-market player, our player who’s a weekend warrior and bets a significant amount of money, I’ve talked to many of them and they’re betting a lot less. They’re getting squeezed out as a result of the shark-eat-shark environment.
So, it’s a lot tougher. It’s a lot tougher for all players.
PAT CUMMINGS: Joe, let’s talk specifically about what NYRA has done in this arena. Two years ago NYRA made the decision, and was public right around this time two years ago, and said they were going to cut off the CAW players from betting in the win pool inside of two minutes to post.
They then subsequently did this — cut out the CAWs from betting in the late pick 5 across NYRA’s tracks. You eliminated the jackpot pick 6 provision, changed some of that.
What was NYRA seeing in the leadup to that? What caused you to look at the investment from bettors across many different segments and say, we’re going to step in and make some changes here and really be the first in American racing to do that?
JOE LONGO: So, similar to what Marshall said, we had a lot of complaints from customers and even people in the industry specific to the win pool itself and the late odds drops. Knowing it is an efficient market, we simply turned around and said, look, we can’t have the late odds drops any longer at the last minute, money to the pool, so we let them wager into the pool up until three minutes to post. So come two minutes, they’re not allowed in the pools any longer.
What you’ve seen is that because there are no CAW effectively in the win pool at all, you are seeing payouts that when you’re comparing to the other pools they’re in, substantially higher. As it relates to the other pools they’re not in, there are several. Our pick 6 that we have, which is traditional, we eliminated the jackpot. Late pick 5 has always been retail only.
It is similar to, at least in our view — and we have the liquidity to do it so I understand that not all can — but running different spreads of poker. You can play a 2/4 limit, or a 5/10 no limit, so in our point of view there is nothing wrong with having some of these pools that everyday retail player can participate in and hopefully have a good result at the end.
I think also on the win pool side of things, fixed odds and that stuff, and I think we touch on that later, likely — it is legal in some states and could be here sooner than we know. In order to protect the pari-mutuel markets here, the reason you are going to move from pari-mutuel to fixed odds is the volatility.
We then remove the volatility from the win pool. Long-term this is something that’s going to benefit us. We’ve seen the win pool itself, the percentage of our handle prior to the restriction, is the same exact number it is after, which more retail people have moved in there.
So, it’s been very positive up to this point, and the end result is a net decrease of CAWs in the pools.
PAT CUMMINGS: Now, some other tracks in America have made some slight changes in the way in which they rebate their players. Rebates are not a four-letter word. It is a way to incentivize your biggest customers, just like you would get a discount if you were buying something wholesale.
No one in the landscape of wagering on American racing has made the move that NYRA has. Do you think it might be something we see considering the results have been pretty positive?
JOE LONGO: I do. And to be clear, some of the bigger tracks are exploring different ways to curtail this change. I think as an industry as a whole, if we can come up with one industrywide CAW policy, whether it’s win pool or any other pool, and allow customers to know they can bet with confidence, it eliminates the whole idea of the boogeyman in the room.
You can bet and expect what you’re going to bet is what you’ll get. I think that is something that I know that we’re working on that in the background. Hopefully we can see something about that in the next couple months.
PAT CUMMINGS: Marshall, we hear algorithms, machine learning. It feels sometimes it’s not just the regular horseplayer going to the track and cracking open the program and spending 10 minutes and having a bet.
This is really modern. This is data, science, evolution, and modernity. I think a lot of the times the CAWs get a bit of a bad rap as taking advantage of the system. Really, they’re just taking advantage of modern technology.
It’s not as if CAW play itself is bad, it’s that it has some effects that affect ordinary horse play.
MARSHALL GRAMM: Well, it’s the evolution of the betting product, right? So, I do think it’s important what you said about rebates aren’t necessarily bad.
My 500-fold increase in handle resulted in a 500-fold increase in contributions to tracks and purses. That margin, right, those rebates, they came from a reduction in what I was paying to the ADW. When I bet $50,000 a year at youbet.com, maybe they kept 12% of what I bet, so they made $6,000. When I bet $25 million a year, right, that resulted in the ADW keeping $500,000. So, this massive increase of what I contributed to purses, plus my ADW is better off, so it’s win-win.
What we see with these CAWs, right, and what we saw with me as a horseplayer, is that people are price sensitive. If we give people good pricing they can play more. If we allow people to be winning players, the amount they play might be exponential.
It’s also important to note that there is no turning back, right? The computers aren’t going away. We can’t step back in time. This winter I went to Sam Houston Park and that sort of was like it because everyone was cut off, but that’s not our solution. That wouldn’t work.
There is no stepping back. I bet right through the front of my ADW. We log in, we bet, we make sure we hit the site and stay logged in, but we go straight to the front end. We have no special back-end tote access.
So, people are able to do that. Our future horseplayer is very tech savvy. I teach a course at Rhodes College, the Economics of Racetrack Wagering Markets, which is effectively handicapping horses under the guise of there is decision making and economics and finance and psychology.
But the kids who I teach, many of them have a programming background. They’re interested in betting on the races. They have found some constraints in doing it, so they end up modeling golf for example and betting golf. This is our future.
Giving people the tools to do computer-assisted wagering, helping out current horseplayers bet more efficiently is, I think, the way to go. The competition is fierce and there is no walking back from it.
PAT CUMMINGS: Let’s talk about the infrastructure itself. As a bettor using an ADW, a mainstream ADW, you can upload files to place bets.
You’re limited as to how many bets you can get in per second. If you want to play yesterday’s pick 6, for example, had $180,000 carryover and generated almost $1 million in new money.
If you wanted to place, say, 2,000 bets in that using your program, how long would that take you to get those bets in it?
MARSHALL GRAMM: I’m not good in quick math, but it takes three bets per second. If you even made a simple Excel file and uploaded your pick 6, right, you’d have to wait for all those bets to process. The teams can do that with rapid speed.
So again, I would be for eliminating those differences. I think all the players should have the fast through-put. Now, this sort of brings us to the problem with the pari-mutuel system, right? When Pierre Oller at a perfumery in Paris invented the pari-mutuel system, it wasn’t intended that people would be using that information at the last minute to make bets, right?
So, the odds movements, which NYRA has worked hard and some other tracks are working to solve, is problematic, and that’s the thing I think we’re going to have to solve. The teams, what they need is the public’s information, too. They need those odds. That’s part of their models.
They are using those odds to bet late, and that is what’s changing the odds. The teams are good in the win pool. They are exceptional in more complicated wagers, pick 4s, pick 5s, trifectas, superfectas.
So, we need tools to help casual players bet those wagers more efficiently. We also need to steer newcomers into wagers that they can win and churn their money.
Oaklawn Park, my home track just three hours from Memphis, Tennessee, they have a 10 percent show wager. So that is appealing to a newcomer to the game. It allows them to churn some money, cash a few tickets, learn a little bit more before they jump into more complicated wagers.
In fact, I find myself, when I go to Oaklawn Park, I may bet $1,000 on the races like I normally would, and may churn $2,000 to the show pool playing a show parlay.
PAT CUMMINGS: All because of the pricing?
MARSHALL GRAMM: All because of the pricing, right. Again, it’s on track. My money is going to Oaklawn Park and the horsemen. I’m not playing through my ADW because I’m getting such a good price.
On-track incentives like that can help the track, can help the horsemen, and can help these new players, right, stay in the game longer, learn more about our sport, potentially more about handicapping and get more competitive.
I do think we need more winners. I know it’s a zero-sum game, but if we lower prices, people have more winning days. And as they have more winning days, they become more invested in our sport. They learn more about our sport. Great defenders of or sport, our horseplayers.
As a horseplayer it was always my dream to own a racehorse. As a result of being a horseplayer I now own a lot of them. Any horseplayer I know, they are great defenders of our sport. We need more fans, and those fans can be drawn to us through the betting window.
PAT CUMMINGS: Joe, we talked about the evolution of pari-mutuel wagering, but as you alluded to, there is a tremendous amount of growth in other forms of betting that are driving sports bettors all across America, and it’s only going to get bigger.
Fixed odds wagering is very popular in other parts of the world. There is a start to it here in America with what’s going on in New Jersey, but where do we stand or what do you see as the future of fixed odds becoming a little more widespread in an era where, as we’ve said here, late price changes can be tough to tolerate for ordinary horseplayers; in fixed odds, the price you bet is going to be the price you get.
Where do you think New York is on that path and really the greater industry?
JOE LONGO: We have been on record in New York saying it is something we do want to participate in, and hopefully we get there. We will see what happens in terms of that.
But it’s important, too, for us to collectively as an industry look at this bigger picture and have control around it in general. I think we’re kidding ourselves if we think that fixed odds coming into the States is not going to potentially cannibalize what we are doing pari-mutuelly.
How big of an effect it has is yet to be seen, but to the effect we can control it from the start, make sure that it’s successful, you know, I think it would be a good thing and supplement what we’re doing here.
I think we’re kidding ourselves with the expansion of sports betting around the world and even more so in the U.S. here that it will take off. There is no reason to think that it wouldn’t, particularly with profits and everything else that can be used as an angle to get more casual folks into the game.
You see what we are doing in New York with the television strategy, and that’s a perfect avenue for that casual sports customer and entry point into the pari-mutuel pools at some point.
Our biggest strength that we have are the exotic pools, so to that extent, do we have fixed odds in the win pool only and use it as an avenue pari-mutuelly for the exotics, which you see are one of our biggest things.
PAT CUMMINGS: From an economic perspective, to offer an alternative to complement wagering, Marshall, do you think that the introduction of, say, derivative products like a head-to-head matchup bet could not necessarily cannibalize the traditional win pool, but create new markets that could generate an increase.
MARSHALL GRAMM: I do. I love the pari-mutuel system. I think it’s the best way to conduct gambling. The late odds movement is the problem with it. The great thing is that the superfecta and trifecta and multi-leg wagers, they don’t matter as much.
You know, the problem with fixed odds are what it does to winning players, but I think it’s something we have to try. I would love to see an exchange market through our tote system that is well priced. I think an exchange system can allow for fixed odds betting. The computer teams would love it because they would get to help set the market. I think horseplayers would love it because they would get a fixed price.
To me that would be a win-win if we could make it work. Like in the nascent days of Betfair. It hasn’t taken off in New Jersey. The market I think is a little small. But that’s something — again, I like your idea of maybe the win pool is one that we can experiment with and then these other pools, which are very large, aren’t going to change.
Right now, I still love the pari-mutuel system.
PAT CUMMINGS: Joe, do you think we could get to a point where maybe more pools get limited to the CAW play? Do you think there is an evolution there to expand that experiment a little bit more?
JOE LONGO: I would say, at least from my perspective, and we’ll see what happens, but I think it’s really all on the table. It’s in our mission statement as stewards of the industry here in New York. We have to make sure the Thoroughbred industry thrives, and a critical part of this is the player, doing whatever we can to make sure that they’re on level playing fields in terms of developing tools that we can for them.
It’s something we have to do. For sure it’s all on the table for us.
PAT CUMMINGS: That wraps up our time. I really appreciate it, on behalf of Marshall and Joe.
Thanks to Stuart Janney.
STUART S. JANNEY III: Gentlemen, thank you very much. That was a very interesting presentation.
We’re going to have a 10-minute intermission now, so we’ll see you back here in 10 minutes.
|
