Why do they win one day and lose the next?

THE AMAZING ‘BOUNCE’ THEORY

What is the "bounce" theory? I guess it won't come as too much of a surprise to learn that it originated in the United States, where punters try to pare horse-racing form down to a mass of statistics.

Ten years ago, says the Today's Racing Digest, if you used the word "bounce" at the track it probably meant you wanted security to bounce the drunk who had just spilled his beer on your formguide.

Nowadays, it's become one of the most used, and least understood, words in American handicapping. Slowly, too, it's starting to come into vogue in Australia. Punters are starting to apply the term to horses that run a big race and then turn in a below-par run next time.

Today's Racing Digest comments: "The bounce is now a catch-all excuse for ratings players who want to rationalise why a horse didn't perform to their expectations. It's simply quicker and easier to say the horse bounced than to recognise that it was a faster pace or tougher competition or a different racing surface that caused the horse to regress."

Are you starting to get the picture about the bounce? That's right. It's a word that tries to "label" a horse's up-and-down performances. Not all players believe in it. Others say you have to take it into account.

Here in Australia, we tend to say that a horse has "gone off the boil" when it wins and then fails, or we say it "peaked" when it won and then began the tapering-off process.

Today's Racing Digest has this to say: "Form cycles are often difficult to figure since all horses are different and that's what makes all this bounce talk dangerous. Just because a horse happened to run a career-best race does not necessarily mean it won't repeat it, or improve further, next time out.

"This is especially true with younger horses as they mature physically, since they haven't really developed a ceiling on their talents. It may hold more validity for an older horse with established form cycle patterns but this requires in-depth analysis of the runner's entire career before an accurate and true picture can be formed."

It goes on to say that performance ratings are a "very subjective thing", adding: "Track variants can differ wildly depending on who's doing the analysis. Very fast and very slow pace situations will lead to inflated or deflated final numbers for those who take surface times as hard and fast.

"Bottom line is that there is a great margin for error in compiling these ratings and it's quite possible that a horse that earned ratings of, say, 7584-80 for its last three races actually ran about the same race each time."

Predicting the bounce is another thing altogether. How do you know when a horse will run a poor race after a big race performance? In 1998, in Meadow's Racing Monthly, Jim Bayle compiled several surveys which seemed to indicate that it is a difficult task to predict bounces.

In the same newsletter, run by Barry Meadow, the noted researcher Jim Cramer revealed details of the largest study ever done about bouncing.

It involved 223,591 horses who made a total of 3,939,265 starts. The survey took months to compile and yielded hundreds of pages of data.

Cramer's opinion after Surveying his results? "Horses do bounce. Big time. If a horse has just run a new top (rating), he's likely to be bet strongly next time out. And he's unlikely to repeat that effort.

"Some horses do run an improved race after a new top. But most don't. Approximately two-thirds of all horses run a WORSE race after a new top."

So there you have it. When a horse "peaks" and hits its best rating, it is far more likely to decline next start than improve. Cramer's extensive statistics show that after a new top rating, 67.1 per cent of horses showed a decline in form next start, 27 per cent improved and 5.9 per cent repeated.

Looking further into the bounce theory, Cramer wrote: "The survey shows that the idea of bouncing is no myth, though the reasons for the bounce vary. Some possible explanations:

• The horse ran his big race due to a confluence of his form at the time, and a favourable pace and trip. Front-runners got easy leads, closers benefited from a pace meltdown. Mud freaks ran big numbers in the slop and not-so-big numbers on fast surfaces.
• The race was simply a peak effort, and peaks don't get repeated in any sport. Athletes tend to regress to their mean performance.
• The race took a physical toll on the horse, wiping him out and leaving him unready to return to the races without a lengthy time to recover.
• The horse was moved up in class and found the tougher competition daunting.
• After the new top, the horse was asked to do something different and he was unable to repeat the big performance under the new circumstances.

Let's hear from another US expert, Neil Campbell, and his views of the bounce theory. Writing in a website newsletter, Campbell says:

"For four glorious days in the middle of July (1997), American golfer John Daly conquered his mind, mastered his concentration and played well enough to win the British Open. Less than a week after earning the most storied trophy in golf, Daly was in the Netherlands for the significantly less important Dutch Open. He played poorly and finished nowhere near the leaders.

"If this seems an odd way to begin a horse-racing discussion on the bounce theory, then it shouldn't. Many have been sceptical about the now-vogue line of reasoning that claims a racehorse is likely to fare relatively poorly, to bounce, after a race that is among the best of its career.

"Really, there's nothing revolutionary about this school of thought. Whether it's the golfer who has worked for months to a mental peak or the Olympic sprinter who has trained for years to attain the perfect physical edge, athletes have long slumped after reaching the summit.

"The word bounce entered the lexicon of racetrack handicappers thanks to a New Yorker named Len Ragozin, a Harvard-educated journalist who left the craft in the 1950s to concentrate on owning and analysing horses.

"His speed figures (numbers assigned to reflect the quality of every race run by every horse) were plotted on sheets of graph paper. (The Sheets, now a trademark, have become another popular term in the racing dictionary.)

"Ragozin realised that there were certain universal patterns in how horses ran, and the graph paper made those patterns easier to recognise. One was that a horse often bounced after an exceptional race.

"For years, the results spawned by the immense research were known only to Ragozin and a few faithful clients. But in the mid-1980s, as the handicapping-information revolution hit full stride, The Sheets and their theories were gaining widespread acceptance.

"Sheets players in New York were cutting horses' odds in half and the fever spread to other major racing centres. Several successful horseplayers spend more than $35 (US) per day to buy the Woodbine Sheets.

"Andrew Beyer, the Washington Post columnist who popularised the use of basic speed figures, admitted in his book Beyer On Speed, that he used to snicker at the bounce theory. But Beyer changed his mind somewhat.

"One of the experiments he conducted was to ask the computer how a horse fared after running three consecutive improving speed figures. Seventy-one per cent, a stunning number, ran a poorer race.

"Of course, the setback usually doesn't last forever. Often, the bounce is just a brief respite before the horse goes on slowly up the ladder again to bigger and better things.

"So, the next time your horse runs poorly just when you thought it was a cinch, don't fret. Think about those four days when you yourself worked harder than you ever had before and how on the fifth day you could barely think straight. Bouncing can happen to the best of us.

There are many racing experts who just won't have a bar of the bounce theory. One of them is David Scott, who says: "Where did this nonsensical notion of 'bounce' originate, anyway?

"My guess is that some journalist, at pains to explain why the horse he touted off a career-best Speed Figure never lifted his legs in his next outing at odds-on, developed the theory.

After all, there must be some logical reason why a horse that ran with such speed and enthusiasm in his previous start was so slow and lacklustre in his subsequent outing, right?

"The concept that an animal bred to run for many miles would, three or four weeks after running 1200m to 2000m, still be tired, is ridiculous.

"Lyle Lovett recovered from the loss of Julia Roberts in less time than that. If a human being ran his fastest 100, 200 or 400 metres in one week, do you think that would have any effect on him the next week? Look at the number of heats Michael Johnson had to run in the Olympics. To believe that a horse would still feel some debilitating physical toll and 'bounce' from a race he ran weeks earlier seems bizarre to me.

"What's more, there's just no evidence of the bounce theory in any other sport. Quarter horses don't bounce. Harness horses don't bounce. Greyhounds don't bounce.

"Football players who have a spectacular rushing, passing or receiving game never are accused of bouncing if their statistics are less compelling the next weekend. Besides, if a race really took that much out of a horse, why would a trainer even consider giving his animal a workout between starts? Generally, horses get a soft day or two after a race but are right back under tack after that. The whole idea of a bounce flies in the face of the evidence.

"Besides, there's no real pattern for a horse after he runs a big speed figure. Sometimes they run well and sometimes they don't. So while many professional players prefer not to bet a horse off a top, that has much more to do with his diminished price than his potential performance.

"Horses off a top often are overbet and have scant value. On the other hand, betting against a horse off a top is a favourite tactic of shrewd horseplayers.

"There are reasons why a horse might not run well but they have nothing to do with how he ran in his last outing. Sometimes the track condition has changed or the pace is different. Maybe the horse doesn't feel well. Who knows? There might be a hundred reasons why a horse runs well one week but doesn't return to that form in his next start.

"The point is, the categorical assertion that a horse ran poorly because he expended too much effort in his last start is unsupportable. There's just no evidence that happens.

"The truth is, despite all the speculation regarding 'bounce', no one can say with any certainty if it really exists. Only the racehorses know for sure. And they're not talking."

I must say I prefer the studied claims of Jim Cramer. He adds some more pertinent points to the debate with the following conclusions: "One of the articles of faith of the bounce theorists is that if a horse runs a big race off a layoff (a spell), he's going to bounce like a basketball next time out.

"This turned out to be true, unless he's given sufficient time to recover from the strenuous effort. Overall, 27 per cent of all horses run a new top in their next race after a new top, but this number increases the longer a horse is laid off after the strenuous new top.

"Another way of putting this is that if a horse returns too quickly after a hard race after a layoff, he is less likely to run well. Horses who run a big number (rating) off a layoff and are rushed back to competition on less than seven days' rest, improve their rating just 22.1 per cent of the time, compared with 31.4 per cent who run better figures with 9 to 10 weeks' rest."

FOOTNOTE: Jim Cramer heads Handicappers Data Warehouse www.horsedata.com, with all data supplied by Equibase. Barry Meadow's Meadow's Racing Monthly newsletter and his many excellent books on betting can be found at his website: www.trpublishing.com. Today's Racing Digest is at www.todaysracingdigest.com.

By Denton Jardine

PRACTICAL PUNTING - APRIL 2002