This is an update of an article I did for Horseplayer, so if it sounds familiar, that is where you saw it.

Andrew Beyer once said, “literature on handicapping can be divided into two eras. Before Tom Ainslie and after Tom Ainslie.” Without Tom Ainslie proving that there was a market for intelligent, well-written books on horse racing, many of the sport’s best known authors – Andrew Byer, Dick Mitchell, Steve Davidowicz, Jim Quinn and Mark Cramer to name a few – may have never found publishers.

Tom Ainslie was the pen name of Richard Carter. While he had success as a writer beyond horse racing – he wrote biographies of Curt Flood and Dr. Jonas Salk – he will forever be remembered for his talents as a turf writer. I met Ainslie at Saratoga many years ago. I can attest that he was personable and likeable, and if I had sooner realized my own writing ambitions, I may have pestered him to the limits of his graciousness.

In 1968 he wrote the first version of what is certainly the groundbreaking book on horse racing: Ainslie’s Complete Guide to Thoroughbred Racing. Ainslie published two more editions of the book, the last in 1988. While racing has changed (In 1988 Ainslie referred to the Daily Double as a “lottery bet”) much of the basic information of racing remains consistent. For the emerging horseplayer, Ainslie’s Guide still might be the best place to start a handicapping education.

It’s also not a bad idea for the veteran punter to occasionally go back to the basics. With nothing better to do on a cold and snowy day a few weeks ago, I picked up my copy of the original edition of Ainslie’s Complete Guide to Thoroughbred Racing and started flipping though the pages. On page 38 was a section labeled, “The Magic Number.”

For anyone who has forgotten, the Magic Number is 17.

Anslie posited that even a person betting entirely at random should not have losses exceeding 17%. Why 17%? That was generally the track deduction (the “take”) on a win bet. The secret was to stretch your bankroll by making a large number of small dollar bets, in essence exploiting the arithmetic of the pari-mutuel system. So a hypothetical player with a $1, 000 bankroll betting $2 a race will make 500 total bets and in Ainslie’s words, “would almost certainly lose no more than $170,” regardless of his handicapping skill. To say it another way, since the crowd as a whole only loses 17% a day, a random better who bets enough races should do no worse.

You can see where this is going. All you have to do to become a consistent winner is to learn enough about handicapping to reduce the worst case 17% loss. If the complete know-nothing can limit his loses to $17 per $100 bet, the knowledgeable handicapper should be able to not only eliminate longer-term losses, but realize measurable profit. Theoretically this makes a lot of sense. Because the game is pari-mutuel, you don’t have to overcome the fixed odds of casino games. You just have to be a little more skilled than the next guy.

Which brings us to the less well-known Dr. Burton P. Fabricand. Dr. Fabricand was a Ph.D. from Columbia who published widely in the fields of atomic and nuclear physics, oceanography and finance. Being an aficionado of the Sport of Kings, he thought there had to be a way to apply a high level scientific approach to making money at the track, and in 1965 he published “Horse Sense: A New and Rigorous Application of Mathematical Methods to Successful Betting at the Track.” Fabricand studied 10,035 (I’m sure you’re wondering the same thing I am – why didn’t he stop at 10,000) races and found that a flat bet on every favorite would result in a loss of nine cents on the dollar. Based on the calculations from Fabricand, Ainslie pointed out that simply by betting on the race favorite, the 17% loss gets cut in half.

So there we are. Just a short 8 or 9% from the land of milk and honey. And all we have to do to get there is…learn how to handicap favorites. Now all of this is based on the concept of synergy, which in the case of the racetrack means that in the long run, the collective wisdom of the betting public exceeds the individual wisdom of any single player. It is a comforting thought. Most people relish the idea of being part of the majority.

Ainslie and Fabricand definitely had the right idea, although they were perhaps a little too mechanical and blind to the idea that grinding out a moderate profit lacks broad appeal. By finding what ultimately came to be known as the “false favorite” Ainslie and Fabricand showed it would be possible to turn your 8-9% loss into a positive if not massive profit. Unfortunately, as with everything that seems too easy, there is a catch or two. In this case, even if you are playing only one track, you are obligated to bet the same amount to win on an average 92 favorites out of every 100 races. Plus, given the relatively low rate of return, you’d have to bet substantially more than $2 a race to be the envy of your racetrack buddies.

I once read an article that noted a bet on the tenth choice in a race will win, on average, two times in one hundred. This makes the natural odds on any respective tenth choice 49-1. Instead, the average return on these horses is only 38-1. The moral of the story was simple; you can’t win if you bet longshots. But suppose you were a good enough handicapper to eliminate 62 out of 100 tenth choices. You might guess that it wouldn’t be that hard since most tenth choices really don’t have a prayer of winning. You might even postulate that it would be easier to eliminate 62 tenth choices than eight favorites. The point is that in theory, Ainslie’s idea works for favorites or longshots, as you’ll see below.

Perhaps Ainslie’s system of playing favorites would be a good way to be a professional horseplayer, but the average weekend player is more likely to subscribe to Halvey’s law of inverse synergy:

On any respective event, the individual wisdom of any single player can exceed the collective wisdom of the crowd.

In fact, to adapt the old racing saw, it is possible to beat a race AND beat the races. None of us may be the equal to the crowd in the long haul, but if we can get the best of them at some key moments, we can still be long-term winners. Most of us are not at the track every day. Most of us in the age of simulcasting would have a hard time limiting our action to one type of bet at one track. But most of all, not many of us have the ability to closet our passion for the game in favor of a rote investment strategy. Sure a solid favorite can be a profitable proposition, but the highlights of any handicapping career are when you hit a race the crowd totally missed.

What we need is to twist Ainslie’s idea slightly to allow for something more expansive than betting favorites. The method is devilishly simple: construct an accurate odds line and apply a conservative betting strategy. What Ainslie was suggesting was that if you could find the undeserving (overbet) favorite and eliminate him, profit was yours for the taking. But the exact same is true if you can eliminate any overbet commodity. And obviously, if there are overbet horses, it stands to reason that there must be underbet horses. If you can successfully assess a horse’s chances in a race, and limit your action to the “overlay” you have everything you need to be a long-run winner. Plus, you don’t have to bet almost every race every day. You can pick your spots and vary your bets based on what you calculate as your advantage. You can wind up on favorites or longshots.

When you set your odds line, in every race, the sum of probabilities of winning for all horses has to equal 100% PLUS the amount of the take. For example, at Belmont the take on a win bet is 16%, so the odds line should add up to 116%. Since the tote board odds are calculated after the track has removed its take, you must in essence replace it before setting your own odds line.

Here are my seven steps to creating an odds line:

1. List the horses in rank order from best chance of winning to worst chance of winning. Don’t spend an inordinate amount of time separating those horses at the bottom of the list.

2. Put a check mark next to any horse that you believe has less than a 5% chance of winning then draw a line that separates these horses from the others. These are your non-contenders.

3. Assign the horses below the 10% line an aggregate win percentage. For example, if there are three horses below the line, you may decide all three together only have a total 8% chance of winning. This simply provides you with the remaining percentage that you can divide among your contenders.

4. After you have taken care of your non-contenders, assign winning probability percentages, working your way up from the lowest contender to the highest contender. I like working from the bottom up because I think I am less likely to overrate the lesser animals. If you start at the top, you may “overbet” the horse, much like the crowd often does. The other issue is that errors at the low end are usually exaggerated relative to the high end. For example, the difference between a horse with a 39% chance (8-5) and a 42% chance (9-5) is relatively insignificant. On the other hand, the difference between a horse with a 5% chance (19-1) and a 2% chance (49-1) seems pretty substantial.

5. After you have assigned percentages, add them up to see if they equal 100 plus the take. If they do not, make reasonable adjustments until they do.

6. Look at the first two choices on last time and decide if the percentages you assigned are acceptable. Since these are the horses you most want to bet, their winning percentages are most critical.

7. Convert your percentages into betting odds.

Now the basic rules for using your odds line.

1. Limit investment to only those opportunities where the crowd’s assigned win odds are higher than the real probability of winning.

2. For horses on your odds line at even money or less, the tote board odds should be at least 1.5 times your odds before considering the horse for wagering. So on an even money shot, the minimum play odds are 3-2. The idea of asking for a premium simply gives you, the linemaker, a reasonable margin of error.

3. For horses between 6-5 and 7-2 on your odds line, tote board odds should be at least 1.75 times your odds.

4. For horses between 4-1 and 9-1 on your odds line, the tote board odds should be at least double your odds.

5. For horses between 9-1 and 19-1 on your odds line, the tote board odds should be at least triple your odds.

6. Horses at 20-1 or higher on your odds line should rarely be bet because the margin of error in assigning these horses a winning probability is greater than for the true contenders.

7. In races where there are two or more acceptable overlays, the highest overlay should always be played. The second overlay can be played as well, but in no circumstances is it sensible to bet more than two horses to win in one race.

While the crowd is more efficient than any individual, as Ainslie and Fabricand proved, they are still wrong two-thirds of the time. Exploiting the inefficiencies of the crowd provides the bettor with outstanding opportunities for a positive bankroll. The one great betting truth is: you will make money if you bet true overlays.