... and at investing.
Hands down one of the best articles I've read in a while
Hands down one of the best articles I've read in a while
I was well on my way to becoming a professor of Renaissance literature way back in the mid-1970s. One night a classmate invited me to the dog track. I said, "What's that?" About forty-five minutes later, my academic career was over! It was that easy. I had what we Joyce scholars like to call an epiphany. Something about the lights and the animals and especially the blinking numbers on the tote board struck a very deep chord with me. It probably had something to do with a childhood largely mis-spent playing Strat-O-Matic Baseball. I have discovered that many horseplayers used to play this game as children. You rolled dice and with statistically-based cards, you could play out an entire season. This is very similar to the horse racing and gambling that I have been doing ever since.
I am here to talk about why most of what you have heard about horse racing is wrong, and why horse racing is much more similar to what you do than other forms of gambling. The general public probably thinks that for the most part, horse racing is just like the state lottery or playing craps or roulette in a casino, except that you have horses running around in circles rather than ping pong balls or a spinning wheel.
In fact, horse racing is entirely different from those forms of gambling for one simple reason. In horse racing, you are not betting against the house. In other casino games, you cannot win, with the exception of blackjack and poker. A good litmus test for someone being a liar and an idiot is if someone ever tells you, "I am really good at roulette," or "I win at craps," or "I have a system for beating the slot machines." There is no such thing. These are games with fixed percentages. The casino might as well attach a leach to your forehead when you walk in the door because the longer you stay, the more you will lose, except for short-term, meaningless fluctuations.
The exceptions to that rule are blackjack and poker. If you count cards diligently in blackjack, you can get a 1.5 percent edge over the house. Casinos, of course, don't get built by players having edges, so the casinos will eject you if they figure out that you're counting cards. This happened to me - I was playing a friendly game of blackjack at the Barbary Coast in Las Vegas about ten years ago, and suddenly a floorman came up to me with an Instamatic camera. I thought, "Wow! This guy recognizes me from horse racing telecasts!" I thought he would take my picture and put it up on the gambling wall of fame or something. Instead, he took my picture and said, "Sir, you are no longer welcome to gamble in this casino." Even though I was only playing five and ten-dollar blackjack, I was still counting cards. That is a very small edge that they don't let you have.
The only other game in a casino that you can win at is poker, which is always situated right next to the horse racing area. The reason that you can win at poker and horse racing is the same - you are not betting against the house; you are betting against the other players. This is such a crucial and fundamental difference, and it is lost on the general public. The house is not setting the odds. In roulette, there are 38 spaces on the wheel, and if you pick the correct one, the house will pay you off at 35-to-one, and they will keep the difference. The longer you play, the more you lose and the more the house wins.
When the other players are setting the prices, it is an entirely different story because somewhere between frequently, occasionally and rarely, the public makes the wrong price. That is the beginning of the successful equation in horse racing. It took me about ten years as a racing reporter and columnist, trying to track down the elusive method of picking the winner of every race, to realize that that was a fool's errand. In ten minutes I can teach anyone in this room how to pick the most likely winner of a horse race. There are data about past performance that we publish in the Daily Racing Form that correlate very strongly with the most likely winner in the race. Most horse players spend their lives thinking that if they just studied a little bit harder or got a little bit smarter, they could pick the winner of the race enough to make some money. There is no such thing. Picking the most likely winner is no great feat.
What you really want to do is determine which most-likely winners are good prices and which most-likely winners are bad prices. It is a very simple equation:
Price X Probability = Value
The entire world of investing is that simple too. Here is what I mean. If a horse has a 33 percent change of winning a race, and if you can get odds of 2-to-1 on him (which means tripling your money), there is no value - the horse is priced correctly. If a horse is 6-to-5 (which means you will only get back 120 percent of your bet) and he is only 33 percent to win, then he is a terrible bet. If you're going to get 4-to-1 (quintupling your money) on a 33 percent chance winner, then it's a great bet.
The majority of people who play horses refuse to think that way. They sometimes say that no horse is worth taking a short price on. That's just not true. If a horse is 90 percent to win a race and you're going to get a 50 percent ROI, then he is one of the greatest bets in history. They sometimes say that all long shots are over-bet and that you should never bet on a long shot. That's not true either. If a horse has a 10 percent chance of winning a race and he's 20-to-1, then you're getting double the value than you should.
What you wait for as a horse player (and investors tell me that they wait for the same thing) is mispricing, for the public to make mistakes. I cannot say for sure why the public makes mistakes in your world, but I know why they make them in mine. The people who most influence the odds in racing are known as "public handicappers". These are the guys at the local paper who run a set of picks every day. "Clocker Joe" or "Cowboy Jim" give you their 1-2-3 picks in every race. Their 1-2-3 picks have an entirely different motivation than your presumptive motivation to make money. Clocker Joe and Cowboy Jim want to pick the greatest possible number of winners so that they can remind their boss at contract time that they picked 31 percent winners in the previous year. That's pretty good for a public handicapper. Yet, with a typical payoff structure, Clocker Joe will still have lost his customers money. We have already taken this analysis to a level of investment and mathematical sophistication far beyond the ken of any metropolitan sports editor. The sports editors continue to reward the Clocker Joes of the world who pick 31 percent winners at very low prices because they don't understand the equation of Price X Probability = Value.
There is also a highly-misunderstood thing at the track called "the morning line". For every race every day, a track employee writes down a set of odds on each horse. These are not the actual prices that you get when you place a bet. They are the prediction of one overworked track employee for how the public is going to bet the race. That employee is not saying, "I believe that horse #1 has a 20 percent chance of winning the race, so I will make him a 4-to-1." He is in effect saying, "For a horse whose past performances look like this, the public is going to make him a 4-to-1 favorite." Those are two extremely different things, and this is not at all understood by the wagering public at large. The public routinely thinks that it means something when a horse is lower than the morning line odds predicted. If a horse is going at 2-to-1 odds when the morning line was 4-to-1, the public thinks "There's action here! THEY know something."
There is a mythical creature at the racetrack that is called "They". The vast majority of people playing the horses believe that there is a "They" - some cartel of brilliant analysts with foreknowledge of the outcome of the races, and when a horse is going off at much lower than his morning line odds, "'They' know. 'They' are betting today." I used to believe in "They" when I first came to the racetrack, but the longer I hung around, the more I recognized that there is no "They." "They" don't exist. Once you believe this in your heart, you have made a huge stride. Some people might argue that it is a stride towards incredible arrogance, and that may well be true, but you cannot play intelligently or profitably laboring under the myth of "They." At least in my world, there really is no "They."
My best guess is that there are maybe 100 people making a good living betting on horses. It is a very tough game. It is much tougher than the investing game for one simple reason. There is a thing in horse racing called the "takeout". You have something in the investing world known as "inflation". The difference is that your $100 today will be worth $103 or $108 by next year. In horse racing, the $100 that the player begins the day with is decreased by 20 percent every time he makes a bet. This is the economic underpinning of horse racing.
In the first race of the day, say the public bets $100,000. The first thing that the track does is take out $20,000 for itself. That $20,000 is used to keep the track operating, to put up the prize money for the horses, to bribe politicians, etc. That is takeout. Takeout in horse racing is much much higher than in other forms of gambling. In blackjack, without card counting, the house edge is about 1.5 percent. In football betting, it's 4.5 percent. It's about the same in roulette. In slot machines the edge is still only set at 6 to 8 percent, and usually the maximum is 10 percent.
The takeout in horse racing is set at 20 percent, and it is kind of an historical accident. Fifty years ago in the U.S., horse racing had a legal monopoly on gambling. The government felt that it was its legal right to extract money from it. The takeout began at 10 percent, which most economists say is probably the optimal rate of takeout for horse racing. That is the number at which people will not notice it, and people will churn their money enough that it is ultimately to the track's benefit. Every five years or so, however, whenever there was a budget crisis, some politician would suggest raising the takeout by a point or two to get more money from the dumb horseplayers. 10 percent became 11 percent became 12 percent became 15 percent and eventually 20 percent. That is the reason that I would be surprised if more than 100 people are making money at horse racing. Fighting that 20 percent bite from the pool race after race is very difficult. If you do the math, if you locked the doors at the race track and people just kept betting, it would take 37 races for every nickel in every customer's pocket to be transferred over to the racetrack. That's one reason that they don't run 37 races per day. Instead, they run nine or ten races, let everyone go home, they go to the ATM to get some more ammo, and they come back the next day. It is extremely difficult to turn a profit over time. You have to really wait for those public mistakes.
Fortunately, there are enough public mistakes that an extremely disciplined and hard-working person can make a living at it. But please do not quit your day jobs. Please. It changes dramatically when you try to do this for a living. Playing horses is a much better hobby. How many hobbies can you show a profit on at the end of the year?
The "wisdom of crowds" phenomenon is common to both horse racing and investing. Over the course of a year, the best public handicappers (all of whom, of course, work for the Daily Racing Form) do not do as well as the assimilated public. We have a guy named David Litfin who is by far the best public handicapper in New York. He picks about 30 percent of the winners each year, which is a very high percentage. This does not get him even, however, because he is forced to make a pick in every race, even races that he doesn't like and wouldn't bet on with his own money. But the public always picks 33 percent winners. How can the public at large, which includes drunk people and crazy people and people betting on the cute one and the one with pink silks, be better than the Daily Racing Form's ace handicapper?
I found the answer to that in James Surowiecki's book, The Wisdom of Crowds. This book popularized a lot of research being done in this area. The most accessible example comes from the game show, "Who Wants to be a Millionaire?" Contestants could request help from three different "lifelines" when they found a question too difficult. They could ask a friend, ask an expert or ask the crowd at large. What they found over the several years of the show was that the crowd was much more likely to be correct than the expert. This may appear to be counterintuitive.
Another example was from the early 20th century in England. There was a contest at a local fair in which the public could guess the weight of an ox. Hundreds of people submitted guesses. Not one individual guess came within five pounds of the correct answer, but the average of all the guesses came within a tenth of a pound of the correct weight. There are certain situations where having hundreds or thousands of pairs of eyes on a problem helps create a higher winning percentage over time. That is why there is no great trick to picking the most likely winner of a horse race, and that is why picking the winner is not ultimately a profitable pursuit.
On the theme of disruptive innovation, I would say that half of the 100 people making money on horse racing are doing so because they came up with an extraordinarily clever idea a few years ago. As you might imagine given the increasing speed and capacity of personal computers, people thought that there had to be a way to feed all of the racing data that's out there into a computer, analyze it, and figure out a way to beat the game. A lot of very smart people backed by some very serious money have taken on this project. They have found that you can beat the crowd this way, but no one could get the return higher than 94 or 95 percent because of the takeout. With a 20 percent takeout, you are doing very well to earn a 94 percent return, but you're still not making any money. This was a huge source of frustration for these guys who couldn't push the program past doing 15 percent better than the general public. Unfortunately, you had to do 20 percent better than the general public just to break even at the game.
So they tried going through the back door. They acknowledged that they probably couldn't push their programs much beyond 15 percent, but they asked how they could lower the takeout. They couldn't ask the legislature to reduce the takeout because they wanted to make a lot of money. Instead, they decided to go into the bet-booking business themselves - "What if we became the house?" They only needed 4-5 points of the 20 percent takeout for their own operations, and they could pay themselves a rebate using the rest of the takeout. Over the last five years, rebating has become the biggest issue in horse racing.
I will probably lose a few of you here, but I will give you the very quick version of why rebating works. The 20 percent takeout model is based on live racing, where the track is responsible for paying for the upkeep of the physical race track, and paying out purse money. About 20 years ago, however, live racing began to be replaced by "simulcasting" - we can sit in a room here in Baltimore and bet on races being run in New York or California.
When that idea came along, the race tracks had no idea what to charge us in Baltimore for the privilege of betting on their races. They couldn't charge the whole 20 percent because someone else was putting on the show, feeding the gamblers, etc. In one of the great arbitrary decisions in history, the race tracks decided to charge 3 percent. It was a ridiculously low number, and it has come back to devastate the industry because 20 years ago, 90 percent of the betting was done on a live basis. Today, only 10 percent of betting is done live. If you are running a race track, you are only getting 20 percent on 10 percent of the bets made on your races. On 90 percent of your business, you are only getting 3 percent.
Here at the Baltimore trading pool, I only have to pay the New York racetrack 3 percent. The other 17 percent goes into my pocket as the Off Track Betting (OTB) operator. Now I can go to the "whales" (the big players) who are using computers to get their return up to 95 percent, and I can offer them a 10 percent rebate on their $20 million or $50 million of bets placed at my windows. The guys who had gotten their programs up to 95 percent are now getting a 10 percent rebate, and suddenly they're making a 5 percent profit on their handle. This is the kind of completely counterintuitive, through-the-back-door thinking that has made a very small number of people able to win at horse racing.
There is another parallel to the world of investing. Just as the location of betting has gone from 10 percent off-track to 90 percent off-track, the types of bets being placed have also undergone a seismic shift. When you first went to the race track, you probably heard that there are three kinds of bets that you can make - Win, Place and Show. A Win bet is for the first-place finisher, a Place bet is for one of the first two finishers, and a Show bet is for one of the first three finishers. Only 25 years ago, over 90 percent of the handle was for these kinds of bets - people would pick a single horse in a single race and decide whether to bet the horse to win, place or show. The only other kind of bet was the "Daily Double", where you would pick the winners of the first two races of the day. This was a promotion to get people to come out for the first race of the day.
The betting public eventually became bored with these three bets. Some of the more fringe types of racing like greyhound racing began to experiment with "exotic bets". Rather than just picking a single horse to win, place or show, they developed the "exacta", where you have to pick the first two finishers in the correct order. Then they developed the "trifecta," in which you pick the first three finishers in the correct order. The superfecta includes the first four finishers in the correct order. They also developed a set of multi-race bets in addition to the Daily Double. These include the Pick 3, the Pick 4, the Pick 5, the Pick 6, the Place 9, all of which are self-explanatory - you have to pick the winners of many races in advance. On one hand, this was just a sign of boredom and decadence on the side of the wagering public, much as it exists in the investing world, which leads you into things like derivatives, options-based assets and other fun games.
In horse racing, these exotic bets became not just a novelty, but also a real opportunity that had not existed previously. The very nature of the exotic bet offered value that was no longer available in the Win, Place and Show pools. There is not a great amount of value in the simple Win, Place and Show bets because the prices tend to correlate pretty well with the probabilities. So there may not be a lot of value to be found in picking horse number four to win a race, but the public might misprice the odds for a horse coming in second in an exacta bet. This new bet created a new opportunity for the public to make a mistake. If you are going to go three or four horses deep into a race, or if you are going five or six races deep into a racing day on a Pick 5 or 6, the chance that the public is going to make a serious mistake grows exponentially. It grows so fast and large that it is worth playing the bets almost blindly. The Pick 4 and Pick 6 routinely pay 50-100 percent higher than the parlay of those four, five or six individual winners does. It is because of inefficiencies in the betting pool and the increased likelihood of the public making a mistake.
The best friend that horseplayers have are the big race days. The saddest thing that has happened to the serious horseplayers has been the growth of casinos and racinos and slot machines over the last thirty years. The worst players (the best players from our point of view) - the ones betting on names and looks and colors - have all left the race track. They are now pulling handles on slot machines all day because of the higher leverage - they don't want to win 10-to-1 on a horse; they want the slot machine to pour out a life-changing amount of money. Horse racing cannot compete with that, so all of the crazy people are now gone from the grandstands. That is terrible for the serious horseplayers!
However, the crazies do come back a few times a year. Kentucky Derby day, Preakness day, Belmont Stakes day, Breeder's Cup day, almost every day at Saratoga, almost every day at Del Mar, the pools are filled with the money of tourists and amateurs - people paying attention to the selections of the public handicappers. It is really worthwhile to wait for those days and bet five times as much on those days as you would on a rainy Thursday at Aqueduct where you know that you are just playing against the other sharks. They'll make some mistakes too, but there will not be a lot of fat in those pools.
There is another phenomenon about the big days and the big races that is not only profitable, but also just as pleasing: once the general public and the media get involved in my little world of horse racing, the opportunities expand exponentially! They get it so wrong that there are wonderful opportunities. I will conclude with an example of that.
It was the spring of 2004, around the time when certain horses begin to emerge as contenders for the Kentucky Derby, which is run the first Saturday in May. It was an indifferent group of horses. There were no superstars or strong contenders from the previous year of two-year-old racing. A horse who began his career in relative obscurity - he was racing in Pennsylvania - slowly became one of the favorites for the Derby. He had a name that people loved - Smarty Jones.
By any historical measure (and there are some very good statistical metrics of pace and time), Smarty Jones was the best of a bad group. He didn't stack up with the great horses of history. As no other horses emerged, and as the general press began to jump on this story, Smarty Jones began to take on a buzz. He was the favorite in the Kentucky Derby, which most people forget because he was quickly recast as an outsider and a long shot. Come Kentucky Derby day, Smarty Jones was a kind of bad favorite. Most intelligent people bet against him. About an hour before the race, the skies opened up, and it began to pour at Churchill Downs. When it rains hard at the race track, the same, predictable thing happens - the horses with early speed do very well. There is a pretty basic reason - they get in front of the field and kick up mud, and the other horses get mad and decide that they don't want any part of this. One or two horses in the lead run their race and nothing much happens. Smarty Jones took the early lead and won the Kentucky Derby by four lengths. He was now certified as the only horse who could win the Triple Crown. People became very excited about him based on this victory in the slop.
Then another phenomenon started to happen. There is now only one horse that can win the Triple Crown since Affirmed did it in 1978. In the two weeks between the Derby and the Preakness, the general news media suddenly descend on the horse racing world. These are people who know absolutely nothing about horse racing. You could point at the stable pony and tell them that it was the Derby winner. You could tell them that the Preakness was 4.5 miles. They really don't know the difference. Since they don't have anything intelligent to say about horse racing, they do what most sports writers do today - rather than writing about sports, they write about human interest stories. This is my profession, but it is a profession that has sunk into the gutter. The journalists in every sport, including horse racing, basketball, baseball, football, hockey or whatever, live for somebody with a sick grandmother or someone with an exotic disease. More than anything, they want a rags-to-riches story or a plucky underdog story, whether or not there is any truth to it.
Don't like ads? Remove them while supporting the forum. Subscribe.