The Gambler Who Cracked the Horse-Racing Code

[cato]

Thansk to everyone for your responses on my questions.
All very interesting.

The thing that struck me was that i got the impression that the computer program(s) they started with and had success with (in HK) were not necessarily the complex unicorn program and machine i had always envisioned. As Dave said, it was simple. Yet they were wildly succesful.

And yes they were the first in HK.

It seems to me that there were a group of people in the US who were the first users of computer programs in the 80s and perhaps you could take that injto the early 90s and many of those programs and machines would have the same or greater computing power than what Bentner was using. Yet the success in the US was fleeting (at least that is my understanding and was my experience) - so i look to the differences between HK vs US racing (some of which i put in my original post) as one of the explanatuions.

BCCourtney had some interesting observations but surely some folks in the US had explored levels 2 and 3 as noted in bccourtneys post.

thanks again.

[dasch]

Originally Posted by dasch

Last night I was working and had TVG on for the 2nd race from Sha Tin as the announcer said that a horse got pounded(this was IN THE GATE not AFTER the start or DURING the race, IN THE GATE) so it caught my attention and I watched the race. The horse that took all the late money showed nothing really and finished poorly. Out of curiosity I looked at the Hong Kong Jockey Club website for the chart and saw that the jockey was questioned after the race about the "unexpected" poor performance. I then looked at other races and in EVERY race ANY horse who didn't run the way expected the jockeys were ALL questioned by the stewards with full explanations made available to the public. Then depending on the situation these horses were put on Steward or Vets list before they can race again! This kind of integrity is 2nd to none IMO and after spending a RIDICULOUS amount of time over the years on American racing I am 90% sure I am going to walk away in favor of Hong Kong. I don't care if every horse I bet drops from 10-1 to 5-2 as long as it happens BEFORE THE RACE STARTS. I guess I have been so wrapped up in my work on American racing I really had no idea the integrity over there went this deep. Anybody who is frustrated will all of the BS in the USA with "supertrainers",odds changing during the race, etc, should take a look.

I didn't realize when I posted this that there had been discussions about Hong Kong for about 3 years so this was nothing new to most of the people here. I really don't have time to read everything so I have to be selective and I really had no idea. I can say after watching and betting the last week and a half I continue to be impressed. Some new things to me and maybe to others:

1)A rider who caused trouble in a race was IMMEDIATELY suspended for a week(2 racing days) and fined HK$100k(about $13kUS). Has ANY rider EVER been fined $13k here in the USA?

2)The scale they use to weigh out the jockeys is digital and CLEARLY visible. Why does the USA only use scales where its hard to tell anything precisely?

3)No morning line(which i think is great), BUT you can see the money and odds of EVERY HORSE in EVERY race as soon as betting opens the day before! Why do the USA tracks hide this information until the previous race has been declared official?

I have only played 3 days and my ROI is over 500% which of course won't continue but I have doubled my bet amounts already and have been more and more impressed with what I see. I don't data mine or have an algorithm like the big teams do so I feel I am not competing for those dollars and I am(based on the odds movement) less than 20% of the time on one of their same horses. I love that they color code the odds so you can see when horses are bet down significantly(20% or 50%) even after the race or the next day.

[acorn54]

correct me if i am wrong but according to mr wong at one point he was the modeler for the program that the benter team used. however he state not all that they did was by statistics. they hired what are called "video formers" that video tape the races and observe and use "subjective" approaches to horses during the running of the races. also they hired "insiders" that got the poop on what was going on trackside. also mr wong states in the book he authored, that some things are better done without a computer in his experience. namely the form of the horse. he also states some popular, what he considers, misconceptions among many experienced horse bettors.
in any event for answers to alot of questions people may have regarding these "syndicates" as mr. wong called the benter team i highly recommend the book as a well worth read. mr. wongs credentials are that he was a former actuary. he also won honors in a math event of high status that they hold in hong kong.

[Unicorn]

There is definitely patterns and angles in racing you can figure out with time spent in front of a computer. Iv'e figured out some stuff with foreign racing, but most the time I think US wagering is a bit more difficult. That is probably the next step that I'll try to conquer. For now though I'm going to stick with what is working. I don't see how anyone can have the time to spend 24 hours on a computer all day race after race. As of this year I'm focusing on what Iv'e profited on. I do go to the track occasionally, but not as much as I use to. There are too many distractions at the track and not enough information available to me that is more beneficial then sitting at home on my computer.

[HalvOnHorseracing]

Originally Posted by PaceAdvantage

This should be required reading for anyone who ever comes on here again and tries to tell us these whales are myths or that nobody can beat the game with an algorithm (they must be past posting), or whatever other nonsense people like to spout who can't fathom that smart folks with computers can absolutely make a killing at this game.

One of the things about t-breds is that there are times it's one factor (SAR-Mdn-Turf-2YO-Chad Brown trains) and times when you have to go deeper. It used to be (Pletcher-dirt-2YO-Mdn-6F), although it was not his year in 2018. I've caught my share of races when the was a bias that the crowd hadn't jumped on, or when the horse met one of the angles I know about. Read this. Even though it's from 2014 it's worth the time.

http://halveyonhorseracing.com/?p=170

[DGroundhog]

Originally Posted by pandy

I'm not so sure he won all that money because of his handicapping skills. These syndicates have a lot of advantages over the average bettor, a big one being that they bet a ton of leveraged combinations in each race that they wager on. I know at the Meadowlands (harness) one syndicate bets about $400,000 a card, or an average of $40,000 a race. Once their program identifies one or more horses that are going off at favorable odds, it automatically scans all of the various exacta, trifecta, and superfecta pools and looks for overlays, then makes a whole bunch of "weighted" wagers including hedge bets that not only gives them a lot of chances to cash multiple tickets in the race, but also has big money on certain combinations that appear ripe. So if we wanted to try the same thing using our own handicapping skills and just using the odds board, we would have to have several people sitting at computers to try and get all the bets in at the last second and we would have to have at the very least several hundred thousands of dollars to start with. And if we started off with a few bad days, that would be the end of it.

So what I'm saying is, he has several big advantages, one, he has a huge bankroll, two, he's using a computer that places a ton of bets just in the last seconds.

Another advantage, he doesn't care if he has the winner. The problem most of us have is that we don't want to have a string of losers, we can't afford it, mentally, or financially. The syndicates are in it for the long run and so well bankrolled that they can get killed for a week or so and they'll just keep betting. Plus, since they play so many combinations, and bet a lot of higher odds horses, they have built in hedge bets. Most people can't afford to spread that much.

Specifically regarding scanning the superfecta pool for overlays - how does that happen, exactly?

A superfecta is a scanned pool. The combinations wagered by various outlets are unknown to the host track until after the results are entered. Only the amount wagered is known prior to that.

After the results are entered and the race is calculated a 'scan' is sent to each outlet betting into the host track. Every site wagering on them sends a response with how many winning combinations were wagered. Only then can the payoff for the superfecta calculated.

This sometimes causes a 'tote delay' if a scan is not received because a link has dropped or the scan process has crashed. WPS, Exacta and Trifecta prices are calculated immediately since all combos are transmitted in real time. It would seem impossible to find overlays in a superfecta pool.

Even analyzing overlays in trifecta pools would be unlikely unless a special connection to the host tote system was available. A few years back at Gulfstream it was revealed some 'high roller' off site outfit had a special connection into the tote system that would have allowed this - but after that incident I don't know of this ever occurring. Not impossible, but unlikely. I don't know of any track offering Trifecta probables to the general public and it should not be allowed to any group of high rollers. That would be a big problem if it were true. (I may be wrong on this one. It's just not convenient to show Trifecta probables but technically it is possible.)

In short, if you don't want some computer geek conglomerate analyzing overlays and auto betting them accordingly, play the Superfecta pool. You are probably safe even betting the Trifecta pool - although technically it would be possible - I just don't think it is occurring regularly (and it shouldn't happen at all).

[Half Smoke]

this is another fun story from 2001 about a crushing whale - Ernie Dahlman -

he has a different approach

the story says he put between $10 and 18 million into the pools and in a good year earns about $700K with a 3 or 4% edge

don't know if he's still at it



https://www.nytimes.com/2001/06/03/m...d-of-odds.html

[davew]

an interesting video on Bentor and history from early blackjack days

How This Man Profited $1 Billion Betting on Hong Kong Horse Races

YouTube

[picojim]

YouTube

[citygoat]

More money in the pool means a chance to win more money