04/24/2002 12:00AM

An attempt to put class figs on maiden races


LEXINGTON, Ky. - Some speed- figure handicappers ignore class. I hope to quantify it.

It is vital to understand the quality of races as well as their times. Times and figures are important, but not so important that they entirely offset other key handicapping elements.

The problem with class is that it is hard to describe, much less quantify. In general, class is a reflection of heart, speed, consistency, and quality. But it is difficult to measure all that with numbers.

So a few weeks ago when Keeneland began its spring meet, I started a class study. Based on numerical factors I determined to be instrumental in handicapping inexperienced horses, I began a chart that will eventually list all the maiden races for 3-year-olds run at Keeneland this spring.

I chose 3-year-old maidens for a simple reason. There is less data in the past performances of this kind of horse than most others, which should make class analysis more rewarding.

Divided by sex, I listed the backgrounds of the top five finishers from each maiden race. I examined their records and figured their combined in-the-money percentage. I averaged their gross earnings per start, the group's top Beyer Speed Figures (two per horse when they had at least two starts), and their sales prices. When a sales price was not listed, I estimated a price by multiplying their sire's stud fee by three.

Finally, I ranked them in each category and added those numbers, which allowed me to generate a combined ranking for each race in the study. The final rankings of the meet's first nine races for 3-year-old maiden special weight males and the averages are displayed in the accompanying chart.

What struck me about this data was the large discrepancies from race to race. The ninth race on April 19 appeared far inferior to the first race on April 5, for example, even though they were both maiden routes.

On paper, the two races figured to attract the same kind of horses. But in terms of numbers, they didn't.

This data confirmed what I always have believed about Keeneland. Their best races are during the early part of their meet. Note how the in-the-money percentages, average earnings, and sales prices dropped for maiden races run between Apr. 17-21.

The reason for this may be timing. By running early, a horse has the opportunity to run twice during the lucrative Keeneland meet, or his trainer can have him ready for the start of Churchill Downs. Keep in mind that trainers based at Churchill Downs want to run early and often there to improve or maintain their stall allocation for the following year.

I admit that this data has a few faults. By averaging the top five horses, it applies the same weight to a horse that finishes first as one that runs fifth. Down the road, I might want to change how these numbers are tabulated and apply more weight to the runners with the better finishes.

Additionally, a chronic loser with a high in-the-money percentage can inflate his numbers, when in fact such a runner is a sucker horse. Fortunately, there haven't been as many of these in the sampled 3-year-old races as there have been in races for older maidens.

But overall, I view those issues as minor. What is significant is how horses fit when exiting these races. Who is the better horse, the one who ran fourth in the best maiden race of the meet? Or the one who was second in the sixth-best maiden race?

Honestly, I doubt there is a number that can simply tell that. It is up to the handicapper to interpret the data.

My intention is to upgrade the performance of horses exiting above-average maiden races, and to downgrade the efforts of those leaving the lesser maiden fields, even if they ran like the wind in the easier race.

Handicappers often witness horses winning easily against mediocre competition, then watch as that same horse fails in a subsequent race when it is looked in the eye. Just as often, we see horses dropping in class run much faster and more successfully when pitted against lesser foes.

I'm hopeful that keeping records such as these will help identify those runners making "hidden" drops in the maiden ranks. In the coming weeks a large number of the horses in my study will likely have returned, which should shed light on the study. I'll report back with the results.