Where do you get real time Totalizer data?

Horse racing- the sport of kings! The glory and grandeur, the tradition! And I want to monkey it up- reduce it to a formulaic, math-heavy, marginally but reliably profitable endeavor.

So where do I get the numbers for the global action on a race, to the second? In my experimental runs I've found that as much as 50% of the action is registered in the last minute before post, with the final data coming in after post and that is enough to confound what I am trying to do. Apparently there are people out there thinking like me. Is there anyplace where I can get the data as it comes in, fast enough to act on it? Thanks in advance.

Quote: AutomaticMonkey

Horse racing- the sport of kings! The glory and grandeur, the tradition! And I want to monkey it up- reduce it to a formulaic, math-heavy, marginally but reliably profitable endeavor.

So where do I get the numbers for the global action on a race, to the second? In my experimental runs I've found that as much as 50% of the action is registered in the last minute before post, with the final data coming in after post and that is enough to confound what I am trying to do. Apparently there are people out there thinking like me. Is there anyplace where I can get the data as it comes in, fast enough to act on it? Thanks in advance.
link to original post

I’m pretty sure the staggering of results and also the secret of which results it’s getting its results from is what makes them fair. And although your not the last person to do this I’m pretty sure what I just said is going to stop you from correlating which machines are using which races

Quote: AutomaticMonkey

So where do I get the numbers for the global action on a race, to the second? In my experimental runs I've found that as much as 50% of the action is registered in the last minute before post, with the final data coming in after post and that is enough to confound what I am trying to do.
link to original post

These aren't exactly about the lag in pari-mutuel pool numbers, but these articles do explain in part why you're seeing a lot of last-second action:

Computer Assisted Wagering (CAW): Anatomy of a Deal (Thoroughbred Daily News, 3/20/2024) - this one makes it sounds like the tracks can make deals with specific bettors
The Good, The Bad, The Future of Computer Betting on Races (Horse Racing Nation, 7/8/2024)

As for where you can get up-to-date betting pool numbers, does anybody have that information? Okay, it has been a while since I have been to a track, but even at the track, the numbers have a little refresh lag in them.

Quote: ThatDonGuy

Quote: AutomaticMonkey

So where do I get the numbers for the global action on a race, to the second? In my experimental runs I've found that as much as 50% of the action is registered in the last minute before post, with the final data coming in after post and that is enough to confound what I am trying to do.
link to original post

These aren't exactly about the lag in pari-mutuel pool numbers, but these articles do explain in part why you're seeing a lot of last-second action:

Computer Assisted Wagering (CAW): Anatomy of a Deal (Thoroughbred Daily News, 3/20/2024) - this one makes it sounds like the tracks can make deals with specific bettors
The Good, The Bad, The Future of Computer Betting on Races (Horse Racing Nation, 7/8/2024)

As for where you can get up-to-date betting pool numbers, does anybody have that information? Okay, it has been a while since I have been to a track, but even at the track, the numbers have a little refresh lag in them.
link to original post

Great information, thank you! I had a feeling something along these lines was happening.

This is what I was planning on doing: Here's what I know about horses- they have 4 legs, they eat hay, and they run fast. But I can start out with the assumption that the action in the win pool comes from people who actually know something about horses, and use that as a gauge to look for bargains in the place, show, and exacta pools. So I need real time information about all those pools. Now I would find something like that, usually the kind of thing where so much is on win that the bet will pay better on place or even show, and that's a bargain 15 seconds before post, but 15 seconds after post the numbers update one more time and someone stepped all over my bargain! So according to these links there are apparently cartels doing exactly this, and they have some kind of special access, and surely are in communication with one another.

Armed with that knowledge, my challenge could now be predicting what these CAW outfits are going to do before they do it and parasitizing them specifically.

Quote: AutomaticMonkey

Here's what I know about horses- they have 4 legs, they eat hay, and they run fast.
link to original post

this is exactly what i know about horses too

It is usually not possible to have a discussion related to racing (among other things) on this site without it quickly degenerating into a lot of extremely childish & poorly informed posturing by some individuals with no real interest in it, for performative 'social' reasons having nothing to do with the reality of the actual subject. But here's a small start at a hint or two if you want to consider doing some research elsewhere.

Quote: AutomaticMonkey

...to the second?

YOU don't. Though others do.

The useful bread crumbs sensibly provided by ThatDonGuy are on the right path to some of the reasons why that is so. You are correct to think that the independent pools for win, place, show, exacta, quinella, etc, are not equally efficient, potentially providing profitable opportunities for arbitrage. I was routinely doing so decades ago to pay my way through school, before advance deposit account individual electronic wagering existed. Since then much has been published on the matter. Including a rigorous and thorough book in 1984 examining the topic by a professor who was an Economist & Statistician in the School of Business at the University of British Columbia (I'm told that now he runs a hedge fund), after which there's been a lot of academic, journalistic, and published public follow-on discussion since, which not surprisingly has significantly affected the nature and extent of the opportunities available.

There are now well funded outfits with arrangements for direct access to the pools that doesn't have to wait for the track's system to calculate public "tote board" updates. But you'd be mistaken to think you can get that. They get those connection deals because of the very high dollar volume of their wagering. They also are not paying the same takeout, instead wagering through something commonly called "rebate shops" that result in them paying low single digit takeout rates, instead of the 15-16% which you will be paying to wager into the same pools. You need to carefully consider the effect of that differential on what you are thinking. You will not be able to access these or other wagering market advantages without frequently wagering VERY large amounts. I suggest you would be very unwise to do so without an understanding of the game; you will not know what it is that you don't know, and there is much more to know. Including, peculiarities of distribution among the muiltiple "winners" in the place & show pools; it isn't what you probably think it is.

There are ways and venues through which it is still possible to make use of mathematical calculation of dynamic market inefficiencies created by the existence of independent pools for wagers on the same event outcome. Oaklawn Park is a major US track that does not choose to make those deals with rebate shops or allow direct access to big players, and that has some effect on the wager volume-time curve, differently than others. But there will be experienced bettors making the same calculation you are with the same data access, and a degree of judgement of what the tote board is telling you (and predicting about itself) is required. Smaller tracks have greater anomolies, and usually a relative absence of calculators feeding off the guesstimators, but in the smaller pools your own significant sized wager may "move the needle" enough to step on your own payout.

On the other hand, there is no reason for any concern about someone having their wagering throttled. Tracks have every reason to love winners; that's fundamental to the parimutual system. You are not betting against the house, you are betting against me, and all their business needs to care about is seeking as much wagering volume as they can get, completely indifferent to who wins or loses. But do it for fun and the intellectual curiousity of it for a while, not with an expectation of paying the light bill.

By the way, the apparatus is called a "totalizator." (Or MUCH less commonly, a "totalisator.") . Not a "totalizer." They've been in use and called that worldwide since they were invented by an engineer in New Zealand in 1913. The spelling difference might matter if you plan to do some research where your intial search results depend on matching the terms. You'll also want to be very familiar with "tote" as in tote system, tote board, etc. Good luck. But not in my pools.

Quote: DrawingDead

On the other hand, there is no reason for any concern about someone having their wagering throttled. Tracks have every reason to love winners; that's fundamental to the parimutual system. You are not betting against the house, you are betting against me, and all their business needs to care about is seeking as much wagering volume as they can get, completely indifferent to who wins or loses.
link to original post

With one significant exception: minus pools.

Quote: DrawingDead

It is usually not possible to have a discussion related to racing (among other things) on this site without it quickly degenerating into a lot of extremely childish & poorly informed posturing by some individuals with no real interest in it, for performative 'social' reasons having nothing to do with the reality of the actual subject. But here's a small start at a hint or two if you want to consider doing some research elsewhere.

Quote: AutomaticMonkey

...to the second?

YOU don't. Though others do.
...

Thank you for your sternly worded, while very informative post. If this is so, I will absolutely stay away from this or any kind of bet where my actual competitors are given, through some arrangement, more information by the house than I am. To me that would be the same as playing poker, but where one of the players is given an extra card because he plays there a lot and he's a good customer. If it was the kind of thing where I could just pay a fee for access to an API with the information that would be different.

^ I can't say that "extra info at a poker table" analogy is all wrong. People in the industry have said some similar things about allowing those arrangements.

Quote: ThatDonGuy

Quote: DrawingDead

On the other hand, there is no reason for any concern about someone having their wagering throttled. Tracks have every reason to love winners; that's fundamental to the parimutual system. You are not betting against the house, you are betting against me, and all their business needs to care about is seeking as much wagering volume as they can get, completely indifferent to who wins or loses.
link to original post

With one significant exception: minus pools.
link to original post

True.

.
Bill Benter made millions betting Hong Kong racing (80s and 90s) using sophisticated mathematical analysis - see link
his net worth is close to $ 1 billion

a google search indicated HK racing has an average field size of 12.5 - way, way higher than U.S. racing

I've never heard him say this but I believe he went to HK because he believed there was a lot of soft money in the pools -

could it be done today - in HK or the U.S._____?______idk__________I doubt it in the U.S. - HK - not sure

the 2nd link is not current - it's from about 2009 - but it says Sha Tin averages $17 million (in U.S. currency) per race in the betting pools


https://www.horse4course-racetips.com/bill-benter.html

https://www.sha-tin.com/info.html#:~:text=Hong%20Kong%20Racing%20is%20thriving,form%20of%20luck%20and%20entertainment.

.