07/03/2001 11:00PM

Playing percentages limits game's scope


LEXINGTON, Ky. - I'm infatuated with Astra. She is so immensely talented, so breathtaking - it's a pleasure to watch her win a race like the Grade 1 Beverly Hills despite being at a pace disadvantage.

Yet watching her compete is like being involved in a long-distance dating relationship. She's a blast when she's around. The problem is, she isn't around often.

We get to see her only every few months. So after making an appearance at Santa Anita for the Santa Barbara on April 14, she didn't reappear until late last month at Hollywood.

Now comes word from trainer Simon Bray that she may make just two other appearances this year: The Beverly D at Arlington on Aug. 18 and later the Breeders' Cup Filly and Mare Turf on Oct. 27.

Four races in a year - what is this, Europe? Racing in this country seems to be headed in that direction. Horses are running less and less. Trainers are picking their spots, especially with stakes horses. They are focusing on the races they would most like to win, and bringing in fresh runners.

I don't fault Bray for this style. I would manage Astra in exactly the same fashion.

She is a mare who thrives when given time between races, and as she is already a multiple Grade 1 winner, the aim is to make her a champion. The races that will ultimately lead to the championship are the Beverly D and Breeders' Cup Filly and Mare Turf. Other races in which she could race would play little part in determining the Eclipse.

Short schedules like these are frustrating for racegoers. Most fans long to see horses like Cigar and Skip Away, who ran consistently every three or four weeks.

It's difficult to wait two or three months for a horse to return. With activity that limited, handicappers must keep trip notes, if only to refresh their memory of a race months later.

It isn't just stakes horses who are making fewer starts. Last year, runners in North American averaged 7.1 starts per year, the lowest since The Jockey Club began keeping track of such statistics. By comparison, horses averaged 11.3 races per year in 1960, and roughly eight starts a year in the early 1990's.

Everyone has an explanation for why this has happened. Some believe horses are more fragile, perhaps because of too much inbreeding. Others feel that use of the diuretic Lasix causes weight loss, and requires horses to be given more time between races.

Those aren't the only explanations for the declining number of starts per year. Another reason is trainer statistics.

A couple of years ago a trainer brought this to my attention. He said he and his peers feel they are judged by their win percentage, and are less likely to give a horse a race for schooling or fitness.

He convinced me. A win percentage for a trainer is viewed similarly to a batting average for a baseball player. Trainers who win at a 20 percent clip are treated like baseball players hitting .300 - they are considered among the best at their profession. When a trainer wins at nearly a 40 percent rate, such as Scott Lake last year, he becomes the horse racing equivalent of home run king Mark McGwire.

I know if I were training and in the midst of a slump, I would be less inclined to take chances in rich races, knowing wins in cheaper races would drive up my percentage.

The solution isn't to do away with the statistics. They provide needed information.

But handicappers and owners would be well served to look beyond the win percentage when analyzing the skill of a trainer. It would be better to consider how he came to reach those numbers. Did he take gambles that paid off in big races? Did he win going up the ladder, or only when he dropped horses aggressively down in class?

Until the public looks beyond the win percentage, this will remain a contributing factor to the declining number of starts per runner. And more horses could join Astra with four-race seasons.