"a) it was basically possible to reasonably consistently show a profit, and b) to do so involved more and more tedious work than I was used to from having a full time job in the stock market"
IME this generalises - you can replace "in the stock market" with basically any job that uses the numerical skills and ability to deliver consistently required to operate any gambling scheme correctly, and add the qualifiers that (i) you need to have pockets much deeper than the amounts involved and/or access to cheap leverage, and (ii) you need to be able to stomach regular and painful losses, and find a huge range of betting and bet-like activity to which this applies. It is also salient that bookies are incredibly reluctant to let you cash out the sums you would want to in order to actually receive your profits while you are in the money in a way that could sustain a professional standard of living. One of the online betting outlets (Betfair?) even used to have a special charge they applied only to people who consistently made large profits. (And may still.) They were at pains to point out that almost nobody has to pay it in practice, which I was not surprised to learn, but was surprised to hear them say out loud.
One would think. But in Japan, for example, if you take a fistful of winning tickets down to the one window they keep open on non-racing days, they will cheerfully hand over (say) ¥40M in cash without either (a) asking for ID or (b) presenting a tax document along with it. [This was a few years ago; possibly the rules have changed.]
Caveat: if you are unwilling to bet electronically, your logistical overhead will increase noticeably.
Japan is very keen on cash, I guess. Racetrack bookies at the kind of British courses I go to usually won't take bets that could involve paying out more than five grand; I think the limits are higher in the prestige courses and the expensive enclosures but still well below the ML reporting threshold
But I am consitutionally not a gambler and never have been. As a teenager I devised what I thought was a pretty sure-fire plan for making a million pounds: earn an average of more than £25,000 a year over a 40 year working life. Which I did in fact achieve, well ahead of the deadline, but I have also learned over that time that even 2% inflation a year more than halves the real value of money over 40 years so a million quid is not what it was back then.
Nate Silver is not allowed to bet more than trifling amounts by any bookmaker with a pulse, as he admitted in an interview around his new book. There are plenty of gamblers in similar case, even in poker rooms.
The other problem is that, in most jurisdictions, horse race and lottery gambling is taxed more heavily than stock market or (more recently) crypto gambling. As access to the latter two has increased, more money is flowing that way. Among other things, that increases the relevance Keynes insight on having the allocation of capital decided as the by-product of the operations of a casino.
So you can win at the horses for the same reason you can win at poker ... because you're playing directly against other people, and some of those other people are stupid and/or not willing to put in the work to do it correctly. I had no idea of this at all (despite having spent untold hours in my teenage years devouring pretty much every book Dick Francis wrote). Thanks very much for this piece!
it is a bit different because a poker player can be pretty sure (win or lose) whether they played optimally or had bad luck, but a horse race punter never can tell whether they just had a bit of bad luck or whether they'd fundamentally missed something important. Another fun fact - in Dubai, they have cameras tracking every horse in the race, and they often find that a horse coming in fourth or fifth will (because of taking the outside of curves or having to move through the pack) have run a greater distance than its losing margin compared to the winner!
At least poker is zero-sum (more or less). In pari-mutuel horse betting you have to overcome a rake of 17-25% on every bet. The idea that one could consistently overcome such a large negative edge felt Promethean when I first learned about it.
Fortunately the "public model" is frequently quite terrible.
that sounds like quite a big over-round - for most races at the track it's more like 12-15% IME. and of course, some of that is the cost of being allowed to watch a dozen young men and women risk their lives over fences rather than passing round little bits of cardboard
horseworlddata.com reports 78% of total wagers returned to bettors via bookmakers, in the UK. (The Tote is a bit worse.) Of course there is variation among courses.
US win/place/show takeouts are occasionally (rarely) below 15% or above 18%; trifecta takeouts are rarely below 22%. Hong Kong takeouts range from 17.5 to 24, depending on bet type; Japan takes 25% across the board.
You can also add the fact that gambling (and the secondary stock market) serves little to no actual productive value, other than to redistribute money from unintelligent people to slightly less unintelligent (or just lucky) people.
I dunno, it gives a lot of people (including me when I can find a free afternoon during the season) entertainment. Equally as important as "horses for courses" is the concept of "getting a run for your money".
The solution there isn’t to continue allowing gambling, it’s to apply those same enjoyable principles to video games and other entertainment (while I’m not particularly familiar with the industry myself, almost certainly there are ways to provide all the same entertainment value of betting on horses, without the ability for some poor addict to bankrupt themselves). Heavily regulate and largely ban online gambling, and limit in person betting to specific venues like horse races and high-end casinos, with minimum income requirements and a cap on how much you can lose. Instead you scratch that itch from video games, which also have restrictions on predatory behaviour but which do at least promote creativity and useful skills (and absorb more labour). All the same enjoyment, with none of the addiction and bankruptcies arising from gambling.
I think you have to make a distinction between horse racing (which employs thousands and entertains millions, and only exists because of betting) and every other sport, where gambling is just a valueless parasite.
"a) it was basically possible to reasonably consistently show a profit, and b) to do so involved more and more tedious work than I was used to from having a full time job in the stock market"
IME this generalises - you can replace "in the stock market" with basically any job that uses the numerical skills and ability to deliver consistently required to operate any gambling scheme correctly, and add the qualifiers that (i) you need to have pockets much deeper than the amounts involved and/or access to cheap leverage, and (ii) you need to be able to stomach regular and painful losses, and find a huge range of betting and bet-like activity to which this applies. It is also salient that bookies are incredibly reluctant to let you cash out the sums you would want to in order to actually receive your profits while you are in the money in a way that could sustain a professional standard of living. One of the online betting outlets (Betfair?) even used to have a special charge they applied only to people who consistently made large profits. (And may still.) They were at pains to point out that almost nobody has to pay it in practice, which I was not surprised to learn, but was surprised to hear them say out loud.
I thought this blog from the Guardian was very insightful on the way that things have changed since bookmakers started getting all Moneyball and trying to manage their risk - I sympathise with the feeling of "keep your poker morality out of our sport" but suspect that the golden age never existed: https://www.theguardian.com/sport/article/2024/aug/26/talking-horses-proof-punter-bookie-relationship-has-changed-for-ever
Interesting. Relatedly, presumably in these days of KYC and money laundering regulations there are limits on anonymous cash payouts at racecourses?
One would think. But in Japan, for example, if you take a fistful of winning tickets down to the one window they keep open on non-racing days, they will cheerfully hand over (say) ¥40M in cash without either (a) asking for ID or (b) presenting a tax document along with it. [This was a few years ago; possibly the rules have changed.]
Caveat: if you are unwilling to bet electronically, your logistical overhead will increase noticeably.
Japan is very keen on cash, I guess. Racetrack bookies at the kind of British courses I go to usually won't take bets that could involve paying out more than five grand; I think the limits are higher in the prestige courses and the expensive enclosures but still well below the ML reporting threshold
Japan really is very keen on cash.
It's perhaps worth noting here that Japan (and, I believe, Hong Kong) have no UK-style local bookies. The game is quite different.
But I am consitutionally not a gambler and never have been. As a teenager I devised what I thought was a pretty sure-fire plan for making a million pounds: earn an average of more than £25,000 a year over a 40 year working life. Which I did in fact achieve, well ahead of the deadline, but I have also learned over that time that even 2% inflation a year more than halves the real value of money over 40 years so a million quid is not what it was back then.
Nate Silver is not allowed to bet more than trifling amounts by any bookmaker with a pulse, as he admitted in an interview around his new book. There are plenty of gamblers in similar case, even in poker rooms.
The other problem is that, in most jurisdictions, horse race and lottery gambling is taxed more heavily than stock market or (more recently) crypto gambling. As access to the latter two has increased, more money is flowing that way. Among other things, that increases the relevance Keynes insight on having the allocation of capital decided as the by-product of the operations of a casino.
Have you ever seen this paper? https://pubmed.ncbi.nlm.nih.gov/24292982/
So you can win at the horses for the same reason you can win at poker ... because you're playing directly against other people, and some of those other people are stupid and/or not willing to put in the work to do it correctly. I had no idea of this at all (despite having spent untold hours in my teenage years devouring pretty much every book Dick Francis wrote). Thanks very much for this piece!
it is a bit different because a poker player can be pretty sure (win or lose) whether they played optimally or had bad luck, but a horse race punter never can tell whether they just had a bit of bad luck or whether they'd fundamentally missed something important. Another fun fact - in Dubai, they have cameras tracking every horse in the race, and they often find that a horse coming in fourth or fifth will (because of taking the outside of curves or having to move through the pack) have run a greater distance than its losing margin compared to the winner!
At least poker is zero-sum (more or less). In pari-mutuel horse betting you have to overcome a rake of 17-25% on every bet. The idea that one could consistently overcome such a large negative edge felt Promethean when I first learned about it.
Fortunately the "public model" is frequently quite terrible.
17-25%??! WTF.
***crosses off "learn to bet on horse races" from bucket list***
that sounds like quite a big over-round - for most races at the track it's more like 12-15% IME. and of course, some of that is the cost of being allowed to watch a dozen young men and women risk their lives over fences rather than passing round little bits of cardboard
horseworlddata.com reports 78% of total wagers returned to bettors via bookmakers, in the UK. (The Tote is a bit worse.) Of course there is variation among courses.
US win/place/show takeouts are occasionally (rarely) below 15% or above 18%; trifecta takeouts are rarely below 22%. Hong Kong takeouts range from 17.5 to 24, depending on bet type; Japan takes 25% across the board.
You can also add the fact that gambling (and the secondary stock market) serves little to no actual productive value, other than to redistribute money from unintelligent people to slightly less unintelligent (or just lucky) people.
I dunno, it gives a lot of people (including me when I can find a free afternoon during the season) entertainment. Equally as important as "horses for courses" is the concept of "getting a run for your money".
The solution there isn’t to continue allowing gambling, it’s to apply those same enjoyable principles to video games and other entertainment (while I’m not particularly familiar with the industry myself, almost certainly there are ways to provide all the same entertainment value of betting on horses, without the ability for some poor addict to bankrupt themselves). Heavily regulate and largely ban online gambling, and limit in person betting to specific venues like horse races and high-end casinos, with minimum income requirements and a cap on how much you can lose. Instead you scratch that itch from video games, which also have restrictions on predatory behaviour but which do at least promote creativity and useful skills (and absorb more labour). All the same enjoyment, with none of the addiction and bankruptcies arising from gambling.
I think you have to make a distinction between horse racing (which employs thousands and entertains millions, and only exists because of betting) and every other sport, where gambling is just a valueless parasite.
If we were all to only write and think about things that are morally impeccable, substack and the world would be a much more boring place.