I will be heading out on a road trip – with a bit of luck, that will include observing a spectacular total eclipse in the beautiful state of Oregon, U.S.A.. If all goes well, I will blog about that later.

That will take some time away from blogging, so I am presenting a piece that I wrote quite some time ago (late 1990’s), concerning my attempt to “beat the horse races”, with the applications of computing power, data analysis, intelligence and dogged determination. I did ok for few years, and learned a lot of statistical techniques (via concurrent courses in multivariate analysis). I certainly learned plenty about the pluses and minuses of data mining, and the wisdom behind the notion that one should be careful about post-hoc analyses.

Plus, I had some fun – there is nothing quite like a big score on a long odds horse that your system predicted, that hardly anybody thought had a chance.

And some of the photos are old, so the quality isn't great, but they are authentic, so there's that. The better photos are from google images.

Horse racing Days (4)

II – The World of the Racetrack

Types of Bettors

People exhibited various levels of commitment to the game, and to gambling in general. There were those who were frankly gambling addicts, that needed the action, come what may. They could be reliably counted on to bet every race, even if they had not had a chance to examine a racing program. Those people tended to bet on anything; races, sports, card games, lotteries, even the proverbial raindrop running down the window pane. One fellow claimed, jokingly I hoped, that his ability to perform his conjugal functions depended on how well he was doing at the track. Needless to say, these types were always having money troubles, regardless of how much they earned. I knew one electrician who worked a regular job, and every under the table job he could wrangle, earning prodigious sums, which went straight away into gambling losses. He was likely to show up in the middle of the racing card, straight from a job, drop a quick hundred, and trudge straight back to work. Like anyone else, these types were not indifferent to money, but their main purpose in gambling was for the ‘juice’, the adrenaline high that betting gave.

Others were realists, who were just happy just to keep their losses to a manageable amount, sufficient to allow them to remain players. They husbanded their money carefully, so that it would last throughout the card. To them, racing was a hobby, a social club, a sport like any other, that one paid a certain amount for the privilege of following. They cared about winning money, but were pragmatic enough to know that it was a nearly impossible task.

Then there were the professionals, or at least would-be professionals. For them profit, or the reasonable expectation of profit, was the main thing. These were the people who always carried with them a copy of The Daily Racing Form, the horseplayer’s bible. They purchased books about handicapping, and studied them thoroughly. When computers came in, they purchased the requisite hardware and software, and tinkered endlessly. They were constantly quoting numbers and testing systems. They spoke the language of ‘value’, much like a stockbroker. A horse that they considered a good prospect, but that the crowd bet lightly was an ‘overlay’, a much sought after situation. The converse, when the crowd ‘jumped on’ a horse, and bet its odds down far lower than it truly rated was an ‘underlay’, a situation scrupulously to be avoided. If a race offered no value, it was best to give it a pass. The idea wasn’t to win races, but to win money. Spotting horses whose odds were way higher than they truly should have been, say a natural 2-1 horse winning at 5-1, was what made for a winning season. Naturally, this is easier said than done; I doubt that more than a few percent of ‘professionals’ made a profit in the long run.

As with any generalization, the real world of bettors was more complex than outlined above. Every individual had a little bit of every betting style within him. Gambling addicts could be as studious and methodical as the professional, at least for a while. Even the most machine-like professionals were not indifferent to the gambler’s rush. And the hope of winning shone forth in even the most pessimistic realist’s heart, for that brief moment when the starting gate opened.

In my opinion, the most difficult role to assume was that of the ‘professional’. Both one’s money and one’s ego were on the line. There was a very stringent and exact measurement for success; turning a profit. And not just any profit; it made no sense to spend hundreds of hours studying races, entering data, testing systems, and making bets, only to win a small return, or ‘chump change’, as they say. The gambling addict got his high from every race. The pessimistic realist had his sights set low, and was invariably confirmed in his wisdom. But the would-be professional laid his heart and soul on the line every day, and usually lost both.

The Mathematics of Betting on Horseracing

In general, the mathematics of gambling is brutal. In fact, in most forms of gambling, losing money is as inevitable as the rotation of the earth. Indeed, in what the statistician refers to as ‘negative expectation’ games, the result is fore-ordained; the so-called ‘gamblers ruin’. A little knowledge of probability theory and some minor algebra can prove this to anyone’s satisfaction, if they care to do the work.

All casino games, with the exception of blackjack are negative expectation games. This simply means that the house has an edge on the bettor, and over time this edge will move money from the bettor’s wallet to the house’s bank. In blackjack, if card counting is employed, the bettor can, in principle, have a small edge on the house. However, this takes practice, dedication, and patience and is also illegal, so for the vast majority of players blackjack is also a losing proposition. The same is true for lotteries, slot machines, and video lottery terminals.

Sports books and horse racing are different, at least in theory. In these games, the bettors’ money is pooled, a certain amount is siphoned off by the house or the track (the takeout is 15-20% at most tracks), and the remainder is then paid back to the winners. The amount of the payout per winner is related to the amount of money bet on each team or horse. Lightly bet winners lead to a high payout, while favorites lead to a low payout.

The idea then, is that you are not betting against the house, but rather matching wits against all of the other bettors in the pool. If the track’s takeout is not too excessive, and if there is a sufficient amount of undisciplined money around, the canny bettor has a chance to make a profit. Undisciplined money is money bet for unfounded reasons; on a whim, superstitiously, or for action. Canny bettors bet money for good reasons; speed, class, breeding, or any other of a number of factors that lead to superior performance. They also choose their spots, betting only when a horse’s real chance of success are much better than the crowd believes.

Winning, then, depends on three things; a reasonable takeout, a lot of uninformed bettors, and the necessary information and knowledge to identify good betting opportunities. Remove any of these, and disaster follows, like a man sitting on a three legged stool that has had a leg kicked out. I would estimate that, at best, only one to five percent of steady players at any racetrack can hope to sustain a profit. It is of such slender hopes obsessions are born.

------------------------------------------------------------------------------------------------------------

Real life is pretty interesting, at the track, but fiction can be almost as good. So, here’s a short story that I wrote in those early years, about a horse-player and the devil (probably).

A Dark Horse

Just what might a gambler give up, to go on the winning streak of his life? Even he can't know for sure. Christopher Marlowe's Doctor Faustus legend is given a Damon Runyon spin, in this short story. For those who aren’t familiar with it, the Faustus legend is about someone who sells his soul to you know who, for fame and fortune. Things are not nearly so simple for the character in the story, though.

This is a short story of about 6500 words, or about 35 to 45 minutes reading time, for typical readers.