As most of us have learned, even after only a few years as a punter, a sensible approach to form will provide a goodly amount of winners. The catch is that everyone usually finds the same "form" horses.

This leads to the over-betting of well-fancied runners and so the story goes on ... overall, the odds are too short to maintain a level stakes profit on the ones that happen to win.

The answer is simple enough in outline but more difficult in reality. That is, crushing the crowd. In other words, taking on the public's picks and finding enough winners at the right prices so that a level stakes profit is achieved, even with a lowish win strike tally.

I was reminded of this "block"'in punting when scanning a copy of Barry Meadow's Racing Monthly newsletter (now, sadly, not published). The headline on the article read: "Using public perception to make a profit" and it's basically the story of a US semi-professional punter named Glenn Barton.

In the year previous to the article being published, Barton made 1,343 bets and won 177 of them. His return on investment was 13.7 per cent, and it was his fourth consecutive winning year. So what is the secret behind his success?

Well, Barton works at what is usually known as "crushing" the crowd. How he came to find his approach is an interesting story in that he read everything that was available about racing handicapping and then discovered that all the things he'd learned needed to be unlearned, though he credits all the authors for helping him lay the foundation for knowing what the public is going to do in a particular race.

By identifying what he calls the conventional handicapping factors, Barton has been able to make accurate predictions about how horses SHOULD be bet by the public in a race.

If the public is betting the favourite and second choice in a race as they should, he says, he feels much more confident in playing his overlooked 6/1 or 8/1 shot.

Barry Meadow, himself a highly successful punter, comments: "The public loves factors trumpeted by these authors (Ainslie, Quinn, Quirin, etc), such as good recent race form, consistency, class and no form gaps and that's not wrong. Horses showing these characteristics DO win more races than horses that don't.

"The essential problem, as many players have discovered to their own chagrin, is that while following these factors will land you on more winners, they won't earn you profits. At some point, every successful player has to learn that he needs to play AGAINST the public's selections."

The turning point for Glenn Barton was reading David Sklansky's book Getting The Best Of It. Sklansky is a respected poker expert with a particular skill at explaining the maths of various gambling games.

Barton explains: "The most important thing I learned from this book is that you have to look at any gambling game from the public's perspective. The parimutuel (tote) market is very efficient; favourites win more than second choices, and 8/5 shots win more than 9/5 shots, so you have to understand how the public arrives at its odds, because they are so good at it. Some authors seem to have some disdain for the public, but the public does a very, very good job of predicting winners."

Barton did not suddenly start winning but he began thinking more like the coach who not only tries to improve his own team, but looks for weaknesses in the opposition. Barton began spending a lot of time trying to analyse the public's betting behaviour.

"The public bets horses in the same way they bet the stock market," says Barton. "Instead of looking at a company's long-term prospects, they're seduced by the latest earnings report, or some story about a new product. Horse players get a little too excited about a good recent performance.

"Sure, some recent performances are more important than older performances, but not so much as the public thinks."

Barton's main handicapping component is final speed. His computer program looks at all of a horse's last 10 starts. Several logic passes are made on each horse's past performances in a bid to determine estimated public perception, as well as an additional effort to analyse in a more contrary fashion. Geometric weightings are utilised, along with corresponding exponential weights to assess public perception.

The software compares its own analysis and the public's analysis to come up with a betting line for each race.

Often, the predictions are off beam. Barton says he thinks the public will bet numbers three and eight but instead they go for two and nine. When that happens, he ignores the race.

"If I don't understand what the public is doing in the race, it usually means I don't have a very good handle on the race," says Barton. "If my analysis of the public perception is wrong, it is likely that my own perception of the race is wrong as well."

He prefers races where the public is betting its expected horses, and also NOT betting the horses he figured wouldn't get much action. He has discovered that the betting public is usually wary of horses returning after layoffs. He's found, though, that such layoff horses often offer significant value. So what Barton is really doing is betting against widespread public opinion. Provided the crowd bets as he expects them to, he is happy to bet against them with a value-priced chance.

In a race at Sam Houston Park, he nailed a 45/1 winner. He explains: "My software concluded that the public, for obvious conventional reasons, would heavily bet the eight and the six, and these two were, in fact, the top public choices. I played another horse, Super Dinner, who had received a respectable rating based on its true potential though he hadn't done much recently.

"The horse looked horrible from a conventional viewpoint but the public validated my analysis by selecting the obvious horses."

Another concept that Barton feels is important is whether the horse has been performing as the public has predicted.

A horse that wins while favoured and finishes 7th when 30/1, or who knocks out consistent ratings every time he races, is probably useless as a betting vehicle, says Barton. Much preferred is the horse whose performances are inconsistent. The more unreliability about the horse, the greater the opportunity for an overlay, he says.

What does Barton bet on? What is the key to not backing a lot of losers and backing enough winners to emerge with a profit while avoiding the favourites?

"I like to wager on a horse that has sound ability, but is being overlooked because of its conventional shortcomings, and/ or the overbetting of the obvious horses," he explains. "Horses that pay high dividends should look bad, but supposed overlays based on conventional analysis must be avoided. Estimating and exploiting public perception is the only way I've been able to show a profit.

Barry Meadow says Barton's approach is mirrored by the theories of author Michael Pizzolla, whose own software programs attempt to predict what the public will do.

One thing that Barton has learned is that there is nothing he will assume to be true. But his play now is based on two beliefs - that final speed is king, and that understanding traditional perception is huge.

"I don't focus so much on the handicapping aspect as compared with trying to figure out what the public is going to do," he says. "Most races don't present betting opportunities. I'd bet on only about 30 per cent of the races I analyse.

"Value is everything and even when you're getting good value, you are going to lose most races most of the time."

In 2000, his best year up to that point, he managed to go through one streak where he won exactly one bet out of 65 attempts! But by not overbetting (he bets exactly 4/10ths of 1 per cent of his bankroll on each race) he's able to stay in the game even during the darkest periods.

He follows two guidelines (outlined in Meadow's best-seller Money Secrets At The Racetrack): He requires that his fair odds be no higher than 6/1 (at least a 14 per cent probability of winning) and he demands at least a 50 per cent premium for his bets; in other words, if his software says the horse should be 6/1, he needs 9/1 before he'll bet.

He says: "Hard experience has led me to the 50 per cent premium figure. I've read books suggesting 20 per cent or 25 per cent but I could never profit with that narrow a margin. Either the public is even sharper than I think and/or my lines just aren't as accurate as I'd prefer."

Although successful with his betting, Barton is astute enough himself not to rely in it for his fulltime income. And he has no plans to turn what has been a recreational hobby into a full-time business.

"My sources of income are my business and my investments," he says. "And way down the line is tinkering with horse racing. I will continue to try to improve my approaches but I have no desire to be a full-time player. I see myself enjoying the intrigue of the game for many years to come.

"It's the challenge of beating the game, a difficult game, that got me interested and keeps me interested. If you spent the same amount of effort in other areas you'd get much more of a reward. But I've gotten a lot out of it. Designing my racing software made me a better software designer.

"The two guys who first introduced me to racing back in college remain my closest friends. My business partner happily and willingly attends racetracks with me across the nation on our business trips. My wife enjoys going with me to Saratoga every year.

"You get a lot more than just money out of this hobby, although the money certainly doesn't hurt."

So there is the Glenn Barton story. An interesting one for punters anywhere in the world. He's taken on the "crowd" and makes money doing so. It's not easy but it can be done.

** This article is based on a feature that appeared in Barry Meadow's Racing Monthly. You can visit Barry Meadow's website at Back copies of the newsletter are available as well as other publications written by Barry Meadow.

By Philip Roy