He Could Have Gotten Richer Faster Working in Finance

June 26, 2023 12:12 PM

Veteran gamblers know you can’t beat the horses. There are too many variables and too many possible outcomes. Front-runners break a leg. Jockeys fall. Champion thoroughbreds decide, for no apparent reason, that they’re simply not in the mood ... Play for long enough, and failure isn’t just likely but inevitable—so the wisdom goes. “If you bet on horses, you will lose,” says Warwick Bartlett, who runs Global Betting & Gaming Consultants and has spent years studying the industry ... What if that wasn’t true? from The Gambler Who Cracked the Horse-Racing Code [Bloomberg; ungated]
posted by chavenet (38 comments total)

My first and only day at the track 30-odd years ago I won an exacta bet paying $176 on two middle-favorite horses. Quit while I was ahead.
posted by Heywood Mogroot III at 12:26 PM on June 26, 2023 [18 favorites]

The house horse always wins
posted by aubilenon at 12:30 PM on June 26, 2023 [8 favorites]

I think the most I ever gambled was in one sitting; $5 at slots. I won $20. woo.

~ The end ~
posted by bitterkitten at 12:57 PM on June 26, 2023 [2 favorites]

Way back in the day I won a million dollars betting on a horse. I still have it.

Unfortunately, the bet was in Turkish lira, and it came out to about $2.50 American dollars.
posted by BlueHorse at 1:19 PM on June 26, 2023 [15 favorites]

My first landlord in the big city was a guy who said he didn't have to work because he'd spent some time betting on horses mentored by an old guy who had a system.

Maybe he was just dealing drugs and that was his cover story, but I choose to believe.
posted by clawsoon at 1:24 PM on June 26, 2023 [3 favorites]

For work-related reasons, I ended up playing with some horse racing betting programs in the early 2000s. They weren't cheap. Data feeds (past performance tables) were expensive. The programs themselves also charged fairly hefty (for a single person) fees. I can't say I was able to get enough experience with them to say if they really worked. My limited access also meant I really couldn't delve too far into complicated wagers (trifectas or other complicated bets beyond simple win-place-show better).

My instinct is, however, if you had time and money to burn, you'd probably end up slightly better off than if you were juts trying to juggle all the data in your head (for example, favouring Bayer vs. Equibase speed figures). Or maybe not. I could be judging them based on my own ineptitude as a bettor. I love racing for the horses and the spectacle and the thrill. I can easily spend a day at a track without betting a dime.

I can see where days off between races would have an effect, but I'm a bit surprised that it could be quantified so easily as to become part of the betting program. I'm also a bit surprised the weather/temperature avenue was a dead end, as it seems there are horses that prefer certain conditions.

Thanks for posting, chavenet.
posted by sardonyx at 1:25 PM on June 26, 2023 [5 favorites]

Pari-mutuel betting is wacky, man.

I remember visiting Tampa as a kid and the folks took us to see a Jai-Alai match. The older folks were placing bets and we were helping them with making picks.

And then after watching the lead horse (cough cough sorry I mean pelotari) fumble a winning shot I realized holy shit, the animals can read their own odds...
posted by JoeZydeco at 1:46 PM on June 26, 2023 [5 favorites]

Microclimates can have a big impact localized weather. If the data used to for his study came from a weather station miles away (airport is common) it may not have been well correlated with specific conditions on the track.

Or, maybe the weather during the season is too consistent; the season might have even been choose for that reason.
posted by Mitheral at 1:50 PM on June 26, 2023 [1 favorite]

I got to the part where it said they weren't going to claim a multimillion dollar trifecta because it was unfair, and thought it sounded like a fairy story and bailed out.
posted by tavella at 1:59 PM on June 26, 2023 [8 favorites]

Many years ago my wife and I and my dad took the (at that time little) kids to the race park here in town, mostly to see the horses, and we bet a few bucks, and the three kids each picked a horse and won a...what's it called, where you pick the first three horses but not the order?

So, we won like $200 so we all went out to a nice dinner, including Grandpa, and the kids got to split the rest of the cash between them which was probably 20 bucks or so a piece. That's the most money any of us have won, probably ever.

I'm not much of a gambling guy, and I have friends who think going to the casino is the best time to be had. I don't go with them, it's so boring.
posted by AzraelBrown at 2:29 PM on June 26, 2023 [3 favorites]

Yeah casinos are real weird if you don't have the gambling bug, because they're super overstimulating and boring at the same time.

Like many of the folks in this thread, despite having worked as a parimutuel clerk, I'm not a big gambler but did one time win enough to cover my bus fare and lunch. My trick was to go to the setup stables and try to figure out which horse/jockey team were VERY BEST FRIENDS and it totally worked so that's my system.
posted by Phobos the Space Potato at 2:37 PM on June 26, 2023 [25 favorites]

I have no business being in this thread.
posted by clavdivs at 2:53 PM on June 26, 2023 [2 favorites]

I don't go with them, it's so boring.

I volunteer in a casino about once per year and it's hard on my nerves. About 4 hours into my first of two shifts, I'm just so done with it all. The incessant sounds from the VLT machines, the people who stay all the day long, some of them blowing their wages no doubt. You start to recognize casino workers after a few years of volunteering and it's maybe not surprising how many of them are just so into the scene, they travel to other casinos for vacation (many of them). Never been to a horse race but it looks like something to do once.
posted by elkevelvet at 2:55 PM on June 26, 2023 [3 favorites]

I went in trying to find out what system they were using but unfortunately the article is rather sparse on details. The article makes it seem like they started with a methodology that's based on using some kind of regression model to predict a horse's probability of winning based on its various properties. But then later, it makes it seem like they made a breakthrough in realizing that the publicly available odds data could also be used to form the basis of a prediction model.

However, the academic paper they mention in the article, "Searching for Positive Returns at the Track" cites the book "Beat the Racetrack" by Ziemba and Hausch. The main idea in that book is that you can exploit a certain structure in horse race betting to identify what amount to mispriced bets. Specifically, in horse racing, you can bet on a horse to win, place, or show (i.e.: finish in first, finish in the top two, or finish in the top three, resp.). Roughly speaking, the key structure being exploited is that these betting pools are separate and thus, by the parimutuel nature of horse race betting, the odds in these pools aren't directed linked which means things can sometimes go out of sync, e.g.: a horse could be heavily favored to win but only slightly favored to place. Given that the betting pools for win are usually much larger and heavily scrutinized than other pools, the basic idea is to take the win pool odds as the truest estimation of probabilities and then look for odds in the smaller pools that deviate and exploit them. This approach may not be super surprising when you learn that part of Ziemba's career as a business school professor was spent identifying cases of market failure and/or when the Efficient Market Hypothesis failed (technically, the joint hypothesis of EMH and the Capital Asset Pricing Model).

In any case, Ziemba is actually quoted in the article (quick side note: the article is from 2018, Ziemba died in 2022) but they don't really give any indication if the way that these guys were using the public odds data was in the same way as suggested in Beat the Racetrack.
posted by mhum at 3:00 PM on June 26, 2023 [21 favorites]

The horse always wins

Unfortunately I think the whole thing is often a losing proposition for the horses.
posted by atoxyl at 3:12 PM on June 26, 2023 [22 favorites]

There was a system from 2015 through 2019 which worked like a charm: Just bet on Winx to win.

If I've done the math right, and you bet $100 on Winx to win in the 2015 Sunshine Coast Guineas and let it ride through all 33 of her subsequent straight victories, which ended with her retirement after the 2019 Queen Elizabeth Stakes, you'd have $6,298,701.30 (ignoring that such large betting would have sunk the odds even lower than the paltry returns for her last couple of years once it was clear she was one of the all-time-great race horses).
posted by maxwelton at 3:13 PM on June 26, 2023 [3 favorites]

I used to go to the horse track quite often as a friend sang the National Anthem there and they had $1 beer and hotdogs. The races themselves were kinda boring. I always thought it was like Las Vegas - a key component for many there was that you could smoke indoors.
posted by The_Vegetables at 3:14 PM on June 26, 2023 [6 favorites]

I got to the part where it said they weren't going to claim a multimillion dollar trifecta because it was unfair, and thought it sounded like a fairy story and bailed out.
posted by tavella at 1:59 PM on June 26

I suspect that had less to do with being altruistic and more about not wanting the spotlight put on their activities in case it got them into trouble down the road.
posted by sardonyx at 3:24 PM on June 26, 2023 [6 favorites]

I used to go to the horse track quite often as a friend sang the National Anthem there and they had $1 beer and hotdogs

lol a friend kept trying to get everybody to go to the local track for the $1 beer and hotdogs but then COVID happened so we never ended up going
posted by atoxyl at 4:11 PM on June 26, 2023 [1 favorite]

The only interesting insight I've ever seen about betting on horse races is this: unlike at the casino where you're betting against the house, at the track you're betting against the other gamblers, because their bets determine the odds. It remains a mystery how to systematically leverage that into a concrete advantage.
posted by Flexagon at 4:11 PM on June 26, 2023 [4 favorites]

A quarter century ago I went to the horse track for the first time during the county fair. I went with a coworker who was familiar with The Ponies, and after we placed bets for the first race we got our drinks and cigars (as was the style of the time) and took our seats in the grandstand.

I had picked three horses, all three of them to show because I was a nervous and wary gambler; after all this wasn’t just six dollars, this was three two-dollar bills which were crisp and delightful to present to cashiers who would give you strange looks. A lot was riding on this, which is a pun I’ve been sitting on for years.

So I would technically “win” if any of my three horses took first, second, or third. I had picked them based upon their names or the number of vowels in their names or any personal connection I could think of to the jockeys name. Very scientific.

Before the race they paraded all of the horses and riders in front of us on their way to the starting gate. One was particularly upset, but not nearly as much as its rider after he was bucked off into the dirt, whereupon the horse was scratched.

“What does scratched mean?” I asked. “It means he’s not racing” my friend responded and that’s when I noticed this horse was one of my Golden Three. Oh.

Well, I still had two in this race, at least until another one of my horses threw his rider in the gate, and was scratched. Ah.

At this point I probably don’t need to say what happened with the final horse, but I did learn that part of the rules of horse racing is that it doesn’t count if your horse finishes but the rider does not. That rider was scraped off on the outside rail of the second turn, and probably fared the worst of all.

This set the tone for my entire career as a race handicapper, which ended only because of $2 bill insolvency.
posted by Skrubly at 6:06 PM on June 26, 2023 [16 favorites]

i have a 100 gecs song stuck in my head now thanks everyone

wait no i really love this song though thanks everyone unironically
posted by bombastic lowercase pronouncements at 6:09 PM on June 26, 2023 [2 favorites]

Phobos the Space Potato: My trick was to go to the setup stables and try to figure out which horse/jockey team were VERY BEST FRIENDS and it totally worked so that's my system.

maxwelton: There was a system from 2015 through 2019 which worked like a charm: Just bet on Winx to win.

Now I need to know: Was Winx VERY BEST FRIENDS with her jockey?
posted by clawsoon at 6:11 PM on June 26, 2023 [6 favorites]

Was Winx VERY BEST FRIENDS with her jockey?

she had a permanent jockey, which you'd do if the horse shows a clear preference for a jockey, but both her jockey and her trainer put Winx's success down to her just really wanting to win horse races, and their job was to just try and make sure she didn't hurt herself.
posted by Merus at 6:29 PM on June 26, 2023 [5 favorites]

I've twice left the track with a fair bit more cash than I arrived with. Both of those times I was tagging along with somebody who genuinely knew their horses. The first was a guy who'd more or less grown up around the track (his uncles and cousins were trainers). The second was a guy who worked for the gambling commission (or whatever they call it) in Britain.

Both made a study of what was going on, read the various form guides and in the end placed big bets on only one or two of the day's races. They won BIG. I won not so big because I didn't bet near as much, but I still bought a lot of drinks for people the next couple of days.

Anyway, based on this correlation, I find myself leaning toward this

Veteran gamblers know you can’t beat the horses

not being entirely accurate while this

He Could Have Gotten Richer Faster Working in Finance

may well be. Because in finance, it's kind of a given that the fix is in. Isn't it?
posted by philip-random at 6:44 PM on June 26, 2023 [3 favorites]

Twenty-fuckin-five to one
my gamblin' days is done
I bet on a horse called Bottle 'O Smoke
and my horse won
posted by Ickster at 6:54 PM on June 26, 2023 [5 favorites]

The Gambler Who Cracked the Horse-Racing Code
-at the track you're betting against the other gamblers, because their bets determine the odds
-the basic idea is to take the win pool odds as the truest estimation of probabilities and then look for odds in the smaller pools that deviate and exploit them
...so...
They Short Horses, Don't They?
posted by bartleby at 8:18 PM on June 26, 2023 [19 favorites]

MONA is the largest private art museum in Australia. It was built by a man named David Walsh, who makes his fortune at part of an international gambling syndicate.

According to this article, his syndicate bets about $10 billion a year across horse racing, sports betting and even lotteries.

Walsh joked that he should have ended up in the street or, even worse, in the audience of the Sohn conference.

“If you assess it with the information that was available to me at the time, it was literally the worst decision I’ve ever made. It turned out well.”

This was down to luck, Walsh said. When the syndicate started betting bigger amounts and playing a wider range of games, Walsh went back and checked the maths of the systems it was using. What he found was that the syndicate’s edge over the house was not nearly as strong as they thought it was, and there was an 80 per cent to 90 per cent chance that the whole enterprise would be bankrupted because it was betting too much. So while Walsh thought he had a system that could beat the market, it turned out he was mainly lucky.

“That means nine out of every 10 people with a similar set of skills than me go broke,” Walsh said.
posted by davidwitteveen at 8:30 PM on June 26, 2023 [4 favorites]

Way back in the 20th century I was tech staff at a university, and one of our summer work study hires was a student whose family raised race horses. They kid had spent his life at the track and could read horses like a book. He declared our local racetrack to be "bush league" and would regularly predict the winners. So we - me, my staff, and some of the Dominican guys who taught summer session Spanish - came up with a work study project for the kid. On his morning work board he would find an envelope containing our daily pool, take the subway to the track, and bet on horses. He never lost, and he kept 50% as long as he doubled our investment. Plus: he earned summer work study credit and wages! Everybody wins!
posted by zaelic at 9:54 PM on June 26, 2023 [9 favorites]

put Winx's success down to her just really wanting to win horse races

Plus she was durable and a great athlete, able to lengthen her stride when necessary--and perhaps most importantly, she had a freakish cadence where she could travel 3.5 strides every five seconds to your typical race horse's 3 strides every five seconds. So even though her stride was not particularly notable in length, being able to travel so many more of them was an advantage.

(Sorry, Winx is my favorite horsie.)
posted by maxwelton at 10:34 PM on June 26, 2023 [7 favorites]

This set the tone for my entire career as a race handicapper

When I was a kid we had a family friend who… bought a racehorse? Invested in a racehorse? I don’t recall the setup exactly but I remember my dad checking for the results of their big race in the news.

“There’s a little symbol annotation, let’s see, what does that mean? Okay, it says ‘no speed.’

Is that good?”
posted by atoxyl at 11:30 PM on June 26, 2023 [2 favorites]

An overview of Phil Bull (because of family connections I met him several times when I was young). His approach was a combination of mathematics and the speeds/times horses had previously clocked up while racing.

When I was into horse racing betting (long ago; stick to betting on just US/UK politics recently because reasons) I used the Timeform rating system that he originally developed, and had a few big wins. My partner at the time was dismissive, but less dismissive when I backed Miinnehoma in the 1994 Grand National well before the race at long odds, based on the weight-adjusted Timeform rating. The resultant big win paid for our luxury fortnight in Barbados.
posted by Wordshore at 11:35 PM on June 26, 2023 [3 favorites]

Going to the racetrack is all about wearing a racetrack suit.
posted by slkinsey at 4:23 AM on June 27, 2023 [2 favorites]

I don't gamble myself but love reading these stories.

Okay, at a very high level I understand that it's possible to make money in horse racing: it's parimutuel, so, as with poker, you just have to be smarter than the other bettors.

What I would love to know (so much so that I almost made an askme to see if anyone had links to stories like this one) is how many people (zero? more than zero? a lot?) make a living betting against the house, on soccer and basketball and stuff like that. As quants, with inside information, whatever. I imagine that anyone who is doing it well is doing it very quietly.
posted by sy at 5:05 AM on June 27, 2023 [1 favorite]

is how many people ... make a living betting against the house, on soccer and basketball and stuff like that. As quants, with inside information, whatever. I imagine that anyone who is doing it well is doing it very quietly.

I don't have a favourite summary of the past few years, but Tesla were heavily bet against by hedge funds shorting their price well before it became the highest-valued automotive company in the world. Similarly for the "Stonk Memes" using Robinhood to rally against short trades on GameStop and AMC to cause losses to those hedge funds.

Shorting is a big practice, but also human-managed funds always make poor choices eventually.
posted by k3ninho at 5:53 AM on June 27, 2023 [1 favorite]

I have spent the last 20 years developing software to predict the outcome of horse races, as a hobby. It's fun, and a nice intellectual puzzle. According to my records I have lost a total of $2000 over 20 years, which I count as success. Data (past performances) costs $1 per card (a day of racing on one track, maybe 10 or so races) I go through phases where I'll win with superfectas, ( 4 horses in order 1,2,3,4) then Daily Doubles (winner of each of 2 races in a row) or straight win bets. It's a difficult puzzle, I enjoy the challenge, and there are so many variables that it keeps me engaged, and occasionally I'll win a couple hundred bucks. The toll on the horses is hard to accept, though. These beautiful animals just LOVE to run, and when paired with a rapacious trainer, that drive can lead to the death of the horse, which dampens my enthusiasm. Now I think I'll read the article.
posted by Floydd at 3:57 PM on June 27, 2023 [3 favorites]

at the track you're betting against the other gamblers, because their bets determine the odds
-the basic idea is to take the win pool odds as the truest estimation of probabilities and then look for odds in the smaller pools that deviate and exploit them

Years ago, now, during the first dot com boom*, I worked for one of the first online betting companies, now long gone. If was founded by a family of on-course bookmakers who were traditionally sort of market makers for odds in the UK. It was a strange world, they would refuse to sign contracts because in their world business was done on handshakes so asking someone to put something in writing was a grave insult.

I didn't work on the betting side, I was the sysadmin. Being the overly logical type I went in thinking there must be some kind of complicated formula for working out how the odds were stacked in favour of the bookie. I was wrong. Bookies have a notion of a percentage, a rough formula for working the odds in their favour - i.e. money coming in will be more than going out but its not infallible. What's really going on is a market in odds. They start with a estimation of the likely outcome** then they adjust based on the bets coming in, lowering the odds the more people bet on something, and raise others to make them more attractive to try and keep the percentage on track.

It's not a casino, though, and they can, and do get it wrong and lose money. Bookies are gamblers too. The big difference is this: Punters are betting on the winners, bookies are betting on the losers.

So yeah, its possible to beat the system, but they have a simple solution to that: if you win too much they will stop taking your money, or only allow you penny bets, and they don't really worry about the why.

Inevitably I used to end up hanging round in pubs with bookies. People were always asking them "How do I win?" and they would answer "Don't bet, put your money in a saving account", without fail. One time, after a trade show, I was in the pub and impertinent questioning led to one of the peoplem, someone very senior in the industry there to say "OK, we will bet on a coin toss, the condition is you have to call heads or tails" and left the non-bookies scractching our heads and asking ourselves what the catch was? In reality there wasn't really one, as far as I can tell. A coin toss is truly random so it doesn't mattter who calls it, it doesn't effect the outcome. It does give the illusion of control though. The industry is based on exploiting a logical fallacy (the gamblers fallacy) , so it was a hook for suckers. If someone did take him up on the offer they would likely end up trying to chase streaks, which would make his job easy. He didn't care, though. All he had to do was to wait for randomness to put him one or two points ahead, then pickup the money off the table, and walk away with a profit, smiling, point made.

*Weirdly I just went looking for any traces to refresh my memory and the website of original incarnation, or rather the second after it went bust and tried to relaunch itself, is still hanging aroud like a ghost, 20 years later, in all its 2000 era glory.

**One of the bookies they employed was something of a savant for stats, but IIRc that was more about football than horses, and certainly not greyhounds of which there was a seemingly endless number playing on TV's that hung in the office, I took to using them as a random name generator as we had ended up with an accidental policy naming machines after ex-employees as people shuffled in and out .
posted by tallus at 1:38 AM on June 28, 2023 [6 favorites]

For a glimpse into the world of UK horse racing and betting, regarding a recent race with some ... unusual aspects ... there is an explainer in The Guardian.
posted by Wordshore at 1:27 PM on July 10, 2023 [1 favorite]