COMMENTARY / HORSE RACING : A Sign of the Times: Inexactitude in High-Tech Era

Horse racing fans who watched the Olympics couldn’t help but be reminded that one aspect of their sport remains in the dark ages: the timing of races. Every athlete in an Olympic race had his performance clocked with precision. We know, to the hundredth of a second, how fast the last-place Ivory Coast relay team ran 400 meters. But we’ll never know how fast the favorite, Arazi, ran the 1992 Kentucky Derby.

Let’s see: The winner ran in 2:03, a change from the 2:04 that was originally posted because of a malfunction in the electric timer. Arazi lost by approximately 8 1/4 lengths, and one length equals one-fifth of a second, according to the old (but highly inaccurate) rule of thumb. So. . . .

Such inexactitude is preposterous in this high-tech era, especially in a game in which countless millions of dollars are wagered on the basis of horses’ times. But a radical change in the way horses are timed is now being tested at Laurel Race Course.

The new methodology is the brainchild of Cary Charlson, whose company, Turf Sciences Inc., has developed computer and video equipment to time every horse--at various stages of a race as well as the finish--with exactitude.

The Daily Racing Form is sufficiently interested in this system that it is working with Turf Sciences on this trial run. As a speed handicapper, I have watched with interest, too, and I have learned that the data on which horseplayers now base their decisions and their wagers is imprecise, misleading or--in some cases--wildly inaccurate. Times quoted in the traditional fifths of a second are obviously less precise than those measured in hundredths, and the difference can be significant in the case of two speedsters who run their opening quarter miles in 22.01 and 22.19 seconds, respectively.

The times and distances of most races are inexact because electric timers don’t start until horses have run 30 feet or so to activate a beam at the official “start” of a race. But the grossest inaccuracies come from measuring horses’ losses in terms of lengths.

If a horse finishes five lengths behind a winner whose time is 1:11 for six furlongs, how fast has he run? The answer is not 1:12; we know that the equation of one length with one-fifth of a second is a fallacy. The correct answer is that there is no answer at all.

The size of a length obviously varies with individual horses, and there is little agreement on what constitutes an average length. (Eight feet? Nine feet?) When even the most astute chart-callers for the Racing Form determine the margins at various stages of a race, there is an element of subjectivity involved. Nor are the official final margins, as determined by a photo-finish camera, objective, because the operators of those cameras employ vastly different methods to determine the number of lengths by which horses are beaten.

Charlson has been interested in these questions ever since he went to work for a video company that filmed the races at tracks in the Midwest. His passions are electronics and computers, and in 1989 he started trying to use these tools to measure races. His ideas didn’t always get the warmest reception from racetracks--especially at one Midwestern track that he measured and found to be 180 feet short of its official 1 1/8-mile circumference.

Turf Sciences films every race in a fairly conventional fashion; each frame of video covers .01725 of a second. The operator of the camera can move the film forward, frame by frame, until a horse’s nose appears on the finish line, or on a line drawn across the screen representing the quarter-mile mark, half-mile mark, etc. That horse’s precise time appears on the screen.

Charlson and his associates have measured the distances of races at Laurel, and they time each race from the instant the gate opens; races at six furlongs are in fact 4,010 feet. To make the horses’ speeds at such odd-looking distances easier to compare, Turf Sciences translates their performances into m.p.h.

When Superstrike won the Frank J. De Francis Memorial Dash, he officially ran six furlongs in 1:09 4-5. In fact, he covered 4,010 feet in 1:12.37, running the first quarter at 38.55 m.p.h., the second quarter at 38.98 m.p.h. and the stretch run at 35.94 m.p.h.

The newest trends in handicapping involve sophisticated analysis of horses’ times. Two recent books, Tom Brohamer’s “Modern Pace Handicapping” and James Quinn’s “Figure Handicapping,” advocate detailed analysis of horses’ fractional times. The disciples of West Coast guru Howard Sartin translate fractional times into feet per second and talk about horses in terms of their velocities.

Eventually, horseplayers will have a whole new array of data with which they can try to unravel the mysteries of handicapping.