Equibase Rating Systems, Dumbed-Down Telecasts? This is a Complex Game, We Should Start Acting Like it

Equibase, to its credit, announced a new rating system for Thoroughbreds last week. This dovetails, somewhat, the rating system in Europe, as well as the Trackmaster rating system in harness racing. 

The reviews are plentiful and mostly not overly complimentary. 

The problem with this rating system, or ratings in general in horse racing, is that the game is really, really hard. 

Pinnacle Sports, which has been around forever with hundreds of billions of betting handle, dipped their toe into offering fixed odds awhile back. It might've been the quickest run to the exits in the history of bet offerings. They couldn't make it work. 

Odds lines, power ratings, how fast is horse A versus horse B with limited sample size is always a problem (especially with lower class stock). The game is difficult, even for the very best. 

Dumbing down such a complex game into a number is achievable, yes, but in most cases you aren't going to end up with an optimal number. 

We as humans like to dumb down just about everything because the human condition wants to live with heuristics or a few simple rules.  It makes us feel better. 

We see a lot of it in horse racing telecasts with rudimentary analysis ("this horse is dropping in class, this horse has a driver or rider change"), easy math ("tough race, hit the all button"), or dumb bets like Racing Roulette (I barely remember what that was, but I remember it wasn't good). 

It's been written here and elsewhere - where people are probably smarter and use better grammar - so I guess it's no use to go over how dumbing down the complex is like sticking a spoke in the wheel of a ten speed. Nor is it any use explaining why a hard game is an edge not a liability. If you read this blog you probably agree already. 

But I guess it could be worth touching on what a successful ecosystem in a complex game should be.

The answer to this question in a twitter thread is, I think fundamentally correct.

In summary, your skill game being hard is an asset and the difficulty of the game provides a breadth - the super sharps, the casual sharps, and the entertainment types. Where your game is sustained is how these three groups are handled. The super sharps can not have so much of an edge that they kill everyone else off. The sharp casuals need to be able to navigate the game in a sharp way, and grind their bankrolls, trying to get better. And finally, the fish can't be led to slaughter - they need to stick around and roll over their banks.

Horse racing clearly doesn't do this well. The CAW's have such a huge edge they're feeding on everyone. The sharps can't complete near as well. And the super casuals, for crying out loud - we're asking them to bet television pick 5 tickets and hitting the all button. 

I believe we need to embrace the toughness of the game, and build the game around it. Equibase ratings are hard. The game itself is hard. If we make this our default, the way we educate new players, adjudicate customer cohorts, run and present the game, sets a foundation that at the very least allows for the game to be responsive to its complex needs.

Have a great weekend everyone. 

Chasing the Steam

The Nobel Peace Prize winner was announced today, and economist Jason Furman had something to say about it. 

The other day a student asked me about the prevalence of insider trading in prediction markets. I now have an answer. pic.twitter.com/CpYLCstLw5

— Jason Furman (@jasonfurman) October 10, 2025

We see this more and more now, especially as prediction markets get more popular; namely, the steam is always right, because it has insider knowledge.

That's kind of the goal of prediction markets - to allow for insider steam to show - and we've been dealing with it for years in some form. 

In 2010, a rule of thumb for a serious player was that the steam was probably fake - a fake rumor, fish betting what they heard somewhere, or what have you. But now I suspect it's different. 

With technology, things move fast. Someone is going to know the length of a national anthem at the Super Bowl, who is going to be nominated for an Oscar, or who is winning a Nobel Peace Prize. And one text to five people can be 50k wagered on one side in a hurry. 

Because I don't bet these markets and ain't a professional I asked one for confirmation. Rob Pizzola, who many of you know, is a sharp and he texted that it is "very, very rare that the steam in these markets loses."

On the other hand, for those of us who play horses we know this to be true... sometimes. And sometimes ..... not true. We know it's much more complicated than steam on or off. And navigating it now in horse racing? I find it pretty difficult and ever-changing, but the lesson from sports bettors is not lost on me. 

If it was years ago, we'd almost always be able to fade steam, because the steam was public money. 

In the Rebel Stakes earlier this spring, after a very public horse like Sandman stumbled in the Southwest, we'd be all ready and firing to fade the early steam when the horse opened at 3-5. But that horse doesn't open at 3-5 anymore. In fact, he opened fairly meh and he closed at I think 3-1, which frankly was probably the horse's true odds. 

Now, with the way the markets are bet, it's hard to decipher what or where the steam even is. 

A lot of the time, I don't even think the sharps know for sure.  

Case in point (or many points), I will look at a win pool and figure out what the board will look like late (to the best of my ability). Sometimes I will see an outlier - on-track steam that says a horse is well-bet and might be going to run well, but the exchange markets (or other markets) say it's dead, square money. 

Honestly, with the sharp money playing an exchange, and with CAW money in the win and multi-leg pools, the breadth of what we see in the odds by market should be small. But many times it isn't. 

I've seen some on-track steam horses at 2-1 who are 10-1 on the exchange markets. 12-1 horses who are 120-1. And on and on. 

Just Wednesday at Keeneland, the Maker steam horse in the 4th - a horse my pal Les Stark likes to say is bet like "they already have tomorrow's newspaper" - was 8-5 on the board, 5-2 in multis, and 7-2 on the exchanges. 

In 2005, 2015 or 2022, or 2024, when I see this I am never betting the Maker horse. My mind can not compute betting a horse at 8-5 that I can get at 7-2 elsewhere. It just goes against everything I believe in betting markets. 

However, in 2025 I have pivoted and I am not so quick to discount it. If CAW's with millions invested in algos have a horse at 5-2; if the sharp exchange action has a horse at 7-2, and strong board action says the horse has a 40% or more chance to win (and the odds board shows signs it will not correct late), I respect it. 

If the horse looks sneaky - say off replay, or a trainer change - I respect it even more, because those types are likely to have the proper, sharp, on-board steam; sometimes much closer to fair odds than the algos. 

Maybe for many of you who just want to bet horses you like or bet against horses you do not, this post reads like nonsense gobbeldygook, and that's okay. The markets are weird, and I am probably weird trying to read them so much. But in 2025, with so many sharps playing this game, they do signal what's happening, and reading what they're saying in plain english can be extremely helpful..... when we get it right. 

When do we chase steam and when do we fade steam in horse racing? In 2025, in my opinion, while it may be clear in prediction markets what to do, horse racing is a kettle of fish. But I do believe quite strongly that insider action with easy to use technology spreading the information has more validity than ever. 

Have a nice Friday everyone.