SYDNEY -- Before I came to Australia, I envisioned a nation of backward horse bettors, a place where a gambler armed with modern American handicapping techniques would have a huge edge over the unsophisticated locals.

How wrong I was.

Race after race, I have observed that bettors and bookmakers have an uncanny knack for knowing which horse should be the favorite, even in seemingly wide-open, 18-horse fields. But I didn't fully appreciate how tough the betting competition is until I spent a day with Mr. D.

A professional gambler who cherishes his anonymity, Mr. D.'s name would be known by few people in the racing business. He doesn't sell tips or write books. But it is hard to imagine a more sophisticated betting operation exists anywhere.

Mr. D. employs a three-person clerical staff, responsible for entering data into his computer system, and several part-time workers, who are needed to telephone his bets on busy racing days. On a Saturday, Mr. D. will play thousands of trifecta combinations on races throughout the country.

His computer analyzes every race and rates each horse's previous performances on the basis of four factors: the class of the race, the margin by which he won or lost, his odds, and the weight he carried. It translates each performance into a figure expressing the horse's percentage chance of winning today: A rating of 20 means that if he duplicated his previous effort he would have 1 chance in 5 of winning; i.e., he would properly be a 4-1 shot.

With these ratings as a starting point, Mr. D. studies each race from standpoints that the computer hasn't considered (such as the probable pace of the race and the fitness of individual horses), makes adjustments to the computer's ratings and decides what he thinks the "correct" odds for every horse should be.

He receives up-to-the-minute tote odds on races throughout Australia via computer, and 15 minutes before each race he gets a telephone report on bookmaker prices at the track. He compares these prices with his own, and looks for horses he considers slightly overvalued and undervalued. Then he types into his computer the numbers of horses he wishes to use in the trifecta, and the computer does the rest.

It prints out a list of the trifectas that Mr. D. will play -- often more than 100 combinations in a single race. Having estimated the payoff for each combination, the computer specifies the amount that Mr. D. should bet on each so that Mr. D. can win 5 percent of the total wagering pool on the event.

Now Mr. D's staff swings into action, calling the wagers into his telephone betting accounts. There are separate off-track betting systems in Australia's different states, and Mr. D. will be betting with all of them.

As I observed this high-tech operation, one thing perplexed me. Mr. D.'s method of playing so many trifecta combinations in every race clearly is geared toward grinding out a narrow margin of profit. But with his expertise, he should be ferreting out horses who can generate windfall profits. In the United States, after all, every handicapper strives to find horses he thinks are superior and are going off at long odds. Why doesn't Mr. D. simply wait for horses that, say, he thinks should be 2-1, but are going off at 10-1 or more?

"Because," he answered, "that never happens. Never, ever. If you had a horse at 2-1 and he went off at 5-2, that would be a luxury. If you got 3-1, you'd think that Christmas had come early."

Odds are an accurate reflection of a horse's true chances and the reason, Mr. D. said, is that, "Australia has bred a more sophisticated group of bookmakers than any country in the world." Bookmaking is a tough, competitive business, and members of the profession work on small profit margins.

Posting a legitimate 2-1 shot at 3-1 would obliterate that margin. Anybody who made such a serious mistake too often would be out of business. So the prices here are remarkably sharp.

As a result, playing the horses here is a bit like betting on professional football back home. Horses' odds, like point spreads, are generally so accurate that even the most astute professional isn't going to find any big mistakes on which he can capitalize. The sharp sports bettors in Las Vegas will get excited if the spread on a game is 3 1/2 and they think it ought to be 2 1/2. The name of the game is finding slim edges.

The experts can find those edges. Mr. D. specializes in playing trifectas because the betting in these pools (where nobody knows what a winning combination will pay) is less efficient than betting to win with bookmakers. "In any given trifecta pool," he said, "there will be a lot of combinations that will be double what they ought to and others that will pay half of what they ought to. You can capitalize on that."

But it's hardly the way I envisioned playing the game here. I thought I would be able to use American-style speed- handicapping methods and come up with 50-1 shots that I love. And, indeed, I've found plenty of them. Unfortunately, they all have run exactly like 50-1 shots.