SYDNEY -- Shortly after my arrival here, I found the perfect horse with which to begin my destruction of Australia's bookmakers.

Federal Prince was a standout on the basis of the speed figures I had been calculating diligently for months. His performances had looked visually impressive on the films I had studied in detail. I thought Federal Prince was virtually unbeatable, and yet most of the bookies at Canterbury Race Course were offering odds of 7 to 1 on him.

I snickered when I saw the animal whom the bookmakers had posted as the 5-2 favorite. Top Deed's speed figures suggested he was a dozen lengths inferior. He had enjoyed the best of racing luck in his previous starts. And after racing only twice in his career at six furlongs, he was now being asked to go 1 1/4 miles -- a difficult assignment on any continent.

I wagered with confidence, and when Top Deed outdueled Federal Prince in the stretch, I did not understand how or why it had happened. What I did learn was that the bookmakers and professional bettors who determine the odds are very smart, and that the game is very tough as a result.

For the past three months, I have worked harder at handicapping than ever before in my life, yet I've felt lucky to eke out a betting profit of $1,500. (That figure does not include the $3,000 I had to spend on handicapping materials.) On the basis of these results, I will not be retiring to a penthouse apartment overlooking Sydney Harbor. And I will never again take for granted some of the advantages American horseplayers enjoy.

In most respects, Australia is a wonderful place to go to the races. The track facilities are excellent, and so is the quality of the competition. The game is honest because stewards oversee it much more vigilantly than their U.S. counterparts. But the most attractive part of Australian racing lies in the mathematics of betting.

With bookmakers working on a 5 percent margin of profit, and with the tote taking only 16 percent, gamblers here don't have to overcome the impossible odds Americans face when they bet a trifecta with a 25 percent "take." Moreover, the combination of the bookmakers and the large tote pools makes it possible for a gambler here to wager and win sums that would be unimaginable in the United States. Even on an event that might be the equivalent of a midweek maiden race at Delaware Park, a bettor conceivably could earn hundreds of thousands of dollars.

But exploiting these opportunities is very difficult. In the U.S., a professional bettor can obtain just about all the information he needs by spending $2.50 a day for the Daily Racing Form and taping his local track's races from a televised replay show. But acquiring the same information in Australia is neither easy nor cheap.

The papers that cover the major racing on Wednesdays and Saturdays lack much crucial data. They give either sketchy or nonexistent information about a horse's position during the running of his previous races, so it may be impossible to tell whether a field is filled with front-runners or devoid of them.

To get past performances for all tracks, I subscribed to the computer racing service of the Australian Associated Press (about $50 a week). To see how horses actually had run in their previous starts, I spent another $100 a week or more for race films. And I was just skimming the surface. A professional told me: "To get all the basic information you need here will cost you $500 a week, and that doesn't include salaries, but you need a staff to help you sort though it all."

Even the full-time pros employing staffs and computers must struggle to find edges in betting, because they are battling bookmakers who cannot survive unless their odds accurately reflect a horse's true chances. And even in the most difficult races -- such as 2-year-old events filled with first-time starters from far-flung training centers -- the odds do just that.

The bookies confidently will list one first-timer as a 5-4 favorite and post another at 500 to 1, and they almost always will know who has talent and who doesn't.

I did find occasional edges (most often created by track biases) but when my assessment of a race clashed sharply with the bookmakers', I usually was wrong and they usually were right. One rare exception occurred last week.

A colt named Moville Peter had finished third in a seemingly nondescript race at a provincial track, but the first- and fourth-place finishers both had won their subsequent starts against good company. I had watched the films of the race and Moville Peter had been the most impressive horse in the field, coming seven-wide around the turn and flying through the stretch. When he was entered in a $100,000 stakes race, most handicappers would have thought he was outclassed, but I was confident he was the best horse. And when he went off at odds of 55 to 1, I made a wager that would have turned my Australian venture into an unequivocal success.

Jostled as he left the gate, Moville Peter came flying at the finish -- and just missed by a half-length. Regardless of what hemisphere in which you are playing the horses, it helps to have a little luck too. I would have to settle for the small satisfaction of being able to say I made a profit in Australia. But given the difficulty of playing the game here, I am relieved to emerge with even a moral victory.