10/04/2001 11:00PM

But serously folks, smaller meets can yield insights


LEXINGTON, Ky - Back in the 1980's, when I lived in L.A., I used to enjoy watching comedians perform live at the Comedy Store on the Sunset Strip in West Hollywood. Contrary to what you might expect, the best time to catch the top names wasn't during the weekend. It was much easier to get a good seat for an unannounced performance during midweek. Comedians like Robin Williams, Sam Kinison, and Louie Anderson were much too successful to need to work on a Wednesday or a Thursday night if they didn't feel like doing so. But they still wanted to.

The reason was that they wanted to try out new material with the goal of finding out whether or not it was good enough to be used when the stakes were much higher than they were while performing before just 50 or 100 people. The jokes that made a small audience howl with laughter at 10 p.m. on a weeknight would soon be heard nationally on "The Tonight Show," or might be used the next time they went on tour. The jokes that didn't work were either rewritten until they were more successful, or were quickly eliminated and forgotten.

Handicappers can learn a few things from the ways comedians manage their material. Although you might be very comfortable with your standard handicapping techniques, you should always be open to auditioning new ideas. For me, Keeneland and Churchill Downs are prime time. I get to write my analysis, and I have the opportunity to manage my hypothetical bankroll. I'm very busy during these meetings, and don't usually have the time available to spend my evenings testing new ideas by applying them to races randomly extracted from aging stacks of Daily Racing Forms.

The Turfway and Ellis meetings provide a better opportunity for me to try new handicapping concepts, and see if they might be productive enough to be useful at other venues. Anything that fares well on paper is tried in real time on races at tracks that have pace characteristics that are at least somewhat similar to Keeneland and Churchill. For example, any angle designed to show profits at a speed-favoring track like Keeneland should also be successful at a speed-favoring track like Turfway. A theory that is expected to be successful on turf races at Churchill should also be productive on the turf at Ellis, since the pace characteristics of winners in one-mile and 1 1/16-mile races on the grass are fairly similar in the most common field sizes at both tracks.

Here are some of the things I have learned over the last few months, the "new material" that seems ready for prime time: In the past, since Keeneland is such a short meeting, when I studied the pace characteristics of races at similar distances on the same surface, I lumped them together for the sake of developing a larger sample size. But I have learned that there can be significant differences in the way those races are usually won. For example, I have known for a long time that early speed wasn't the optimal pace style on the turf at Keeneland, but I didn't realize that front-runners have been less than half as successful in 1 1/16-mile races than they have been in mile races. I now fully appreciate the fact that significant edges can be discovered by questioning generalizations that most handicappers would view as being innocuous. I've also learned some of the finer points of integrating my pace figures more fully with my definition of what an improving horse should look like. Last, but not least, I expanded my research on form cycles and now have a better handle on the the ways I will judge the relationship between recent form and betting value.

Will any of this work yield noticeable results in my betting bankroll? Only time will tell. But regardless of what transpires at Keeneland and Churchill, I will continue to work on my new material until I get it right.

New policy: $40 max

Speaking of the bets I suggest in my analysis column, I'm trying something new for this Keeneland meeting. Regardless of whether I'm winning or losing at any time during the meet, each one of my bets will be for a total amount of $40 per horse. It might be divided into win and place bets, or used in exactas, but not more than $40 will be bet on a single horse (a $20 exacta box is within the parameters, but a $40 box, or multiple $40 exactas keying the same horse would not, since it would be the equivalent of $80, or more on the horse, or horses that are keyed. On occasion, when I dislike a favorite, or when I believe a race is wide open, I might bet $40 to win on different horses in the same race, with the expectation that their odds will be high enough to make the bets worthwhile.