"Learning, earning and loving the track."
Starting to Take a Position
on Starting Position
No matter how you cut it, the way the Daily Racing Form measures fractional times and horse�s lengths back is a little screwy and antiquated. They have five possible points of call, at which they note each horse�s position and (usually) lengths behind the winner. For example, depending on the distance of the race, the second call can come after 1/4 mile, 3/16 mile, 1/2 mile, or one full mile. Not only that, but there are distances where the times they record don�t match up to the points of call. For instance, they record the 1/4 mile, 1/2 mile and finish times for 5 furlong races. However, at the same distance, they mark the horses� position and lengths-behind at the start, after 3/16 mile, 3/8 mile, 1/2 mile and the finish. So, at the second and third points of call of a 5 furlong race, you know the horses� relative positions, but you have to guess how fast they were going. I guess we should consider ourselves lucky that some of the times and points of call actually match up.
It�s possible to imagine past performances where you can look down a column and everything is comparable: the first call is always at the start, the second call is always at 1/4 mile, the third call is always at 1/2 mile, the fourth call is always at 3/4 mile, the fifth call is always at one mile, etc. At the expense of two or three more columns of space, everything would make a lot more sense. Heck, if they�re willing to waste space on the DRF speed rating + track variant, they�ve definitely got room to spare.
Anyway, getting off my rant, one little-used piece of information is what�s sometimes called the "start call." This is the DRF�s first call in sprints. You can distinguish it from the other points of call because only the horse�s position is noted, not its beaten lengths. This call is "the horse's position immediately after leaving the starting gate," according to the DRF. Since the starting post can vary from almost nothing to over 100 feet past the gate depending on the track/distance/surface, the start call can mean different things in different situations. If you guess the average distance from the gate to the starting pole at around 50 feet, and you accept (as I do), Charles Carroll�s estimate in his book Handicapping Speed of a running horse�s length as 8 feet, we�re talking about six or so lengths out of the gate on the average. Given that this distance is only about 2 to 3% of your typical sprint�s length, the question is: Is this bit of information worth anything?
I recently did a "paper-and-pencil" research study (as opposed to a larger database query) of the start call in dirt sprints. I looked at 630 performances, and divided them between those who were in the first half at the start call and those who were in the second half at the start call. Those in the front half of the field won 68 of 320 performances, for a 21% win rate. Their Return On Investment (ROI) was .94 (-6 cents per dollar wagered). The horses in the rear half of the field at this point won 35 of 310 performances, for an 11% win rate and a dismal .56 ROI. (These stats might change in a larger sample, and I welcome anyone with a database they can query to follow up on this on The Grandstand message board.)
I think the general trend is clear. However, it�s important to distinguish between statistics that a handicapper has access to before the start of a race, and those that are only available after the gate has opened. These aforementioned stats won�t do you a bit of good by themselves, because racing tellers have a disagreeable tendency to refuse bets once the horses have actually started running. But we might be able to learn something here.
So, given that there is such a big difference in win% and ROI between these two groups, why, half-a-dozen lengths out of the gate, are the horses already so spread out? Some horses can be several lengths behind by the time the leader passes the starting post. This is especially relevant in a sprint, where a few lengths can be tough to make up in the stretch.
"Why, half-a-dozen lengths out of the gate, are the horses already so spread out?"
My cynical side pops up (over my left shoulder, wearing red and holding a pitchfork) and suggests that this is something trainers do to darken their horse�s form when they�re not really trying to win this particular race Maybe they want a good daytime workout so they�ll have good odds and a fit horse next time. If the horse is several lengths behind before the race even starts, the jockey doesn�t have to pull the horse up to get a reasonable work and a lousy finish out of his mount. Many times have I seen a sprinter start out several lengths behind the field, rush up at or near the lead, and then fade far back in the stretch. Is this on purpose or are these jockeys just asleep at the wheel and then overcompensating for their tardiness at the start? Only the Shadow knows. All I know is my larceny detector keeps buzzing. And if you scan a sprinter�s start positions, I�d have to say, without giving it a serious statistical study, that you�d be hard pressed to predict many horses� start position today from their previous start positions. (Doesn�t mean I won�t try.) Putting aside the incurable E horses and SS horses (to use RS/Pos terms), for many sprinters, their position at the start seems to vary widely and unpredictably. This, too, makes me suspicious of human manipulation. So, the first possible answer is, maybe the front half includes mostly "tryers" and the rear half includes mostly "non-tryers."
A second and less devious explanation is jockey running style. There seems to be some correlation between a particular jockey and a sprinter�s start position. This is a subject calling out for a good statistical study. ( I know Jim Cramer of HDW has investigated the "start call" in sprints, but I�m not sure what his conclusions are.) Jockeys are creatures of habit � they have certain positions they tend to be in at the start of a race. (You have to wonder why a jockey with the habit of losing three lengths in the first few seconds of a sprint would continue to get mounts, but that�s another question.) If a jockey has a style that means he is likely to be in the rear half of the field by the starting pole, that might be a jock to avoid in sprints.
Of course there are other influences: post position, gate behavior, troubled starts, the horse�s gate training, a horse�s inherent running style, etc. I guess this article is more an exploration than a declaration. We need to find a way to predict a sprinter�s approximate position at the starting pole. Maybe a good start would be the average start position of this horse when ridden by today�s jockey in the past, if applicable. If the jock has never ridden the horse in a sprint before, you might average the horse�s sprint start positions. This would give an average start position for each horse. Taking the half of the field with the lower averages might be a decent predictor. I�m curious if any of you have suggestions as to the best way to predict start position in sprints.
Bottom line: If you can predict start position for sprinters in even an approximate way, you have a head start over other handicappers in the sprint to profitability. NC
Copyright �2001 NetCapper Inc. All rights reserved.