A small group of horseplayers gathered for dinner the other night and talked about the races the way horseplayers always do -- with a few differences.

"It's not surprising that Santa Anita and Hollywood are so dissimilar," one of them said. "All you have to do is look at the fact that horses have to use 53 percent of their energy in the early part of a race at Santa Anita and 51.5 percent at Hollywood."

Another one nodded. "And just look at the way early pace has been winning at Santa Anita compared with sustained pace and average pace at Hollywood."

I sat mute amidst this discussion. Then somebody said something I understood. "Andy, we used to use a lot of the stuff from your books," one of these horseplayers said. "Of course," he added, "that was a long time ago."

California has been the birthplace of so many important innovations in our culture -- right turn on red, trout-and-avocado pizza, etc. -- that it is not surprising that a whole new approach to handicapping is being developed here. It was the brainchild of a psychotherapist, aided by a group of truck drivers.

Dr. Howard Sartin was conducting a group-therapy session for eight truck drivers who were compulsive gamblers. The traditional way to deal with this problem is to stop gambling entirely, but this is California and Sartin had a novel idea.

"The cure for losing is winning," Sartin said. "So I suggested that none of us place a bet until we established a method that picked 45 percent winners." The patients plunged into a study of handicapping, but none of them was more avid about the project than Sartin.

"It didn't take us long to isolate the importance of early speed," Sartin said. "In the writings of Pittsburgh Phil and Colonel Bradley (two famous handicappers of yore), both mentioned measuring horses' speed in feet per second. So we translated a horse's time into his velocity."

Instead of saying that a horse ran a quarter mile in 23 seconds and a half mile in 47, Sartin and his students would say that he went the first quarter in 57.39 feet per second and the next quarter in 55.0 feet per second.

They began to analyze every segment of a horse's performance in this fashion, and started developing a handicapping method based on these numbers. Sartin was sufficiently encouraged by his early results that he wrote an article on his work for "Gambling Times" magazine. Almost immediately, he became the guru for an ardent group of disciples.

As home computers have proliferated, growing numbers of horseplayers have looked for ways to apply this new technology to handicapping. One was Dick Mitchell. A college mathematics teacher here, he also had been doing research into pace and time, but when he saw Sartin's work, he admitted, "The Doc's stuff was better than mine, and truth is truth."

Now, Mitchell is feeding Sartin's numbers into some of the five computers in his apartment and developing sophisticated systems for betting strategy.

Tom Brohamer was one of the earliest of Sartin's disciples. He conceived the idea of using the numbers to develop a "track profile" -- a mathematical expression of the types of running styles that succeed at a particular track -- and the Brohamer Model is now part of the gospel for Sartin's disciples.

Sartin and his top lieutenants conduct seminars on their method, and any commercial handicapping device has to be viewed with a measure of skepticism.

One neutral observer is James Quinn, author of "The Literature of Thoroughbred Handicapping," who says, "I like the work they've been doing. The leading personalities are very impressive people, and they've developed and advanced the method over several years, based on a lot of data about what works best. I do know people who have become big winners using it."

The Sartin method is no magic formula, by any means, but it represents a serious effort by smart, well-intentioned people to unravel some of the mysteries of the game. "What we're dealing with," said Mitchell, "is the physics of the way horses run." Thursday: How the method works