Correct me if I'm wrong here but there are a number of ways to look at this.

1) Using 1 length / 5th is a convenient standard to even the playing field, as long as it's used with all horses in the race....or just maybe evening the playing field may not be a good thing.

2) If you're calculating an internal final fraction based on lengths gained or lost, it's difficult to be accurate without taking into account the early fraction as well. If I know a horse was used hard in the first half of a race (assuming he's running 1 length per 5th, I think it's safe to say that in the late part of the race he's probably running less than a full length per 5th. The same has to hold true for a slow first half.


If I'm off base, I apologize.

I did a study of 182 races at Aqueduct, 6 furlongs on dirt, no mud, cheaper horses. The results indicated that the 5 lengths per second is reasonable at the 2nd call and finish. At the 1st call 6 lengths per second seems to be the norm.

When I get around to it I plan to repeat the study for other tracks, distances and classier horses. First I need to rewrite the software. At this point a lot of the math is manual.

I've never found five lengths per second to be good enough at 6f unless it is a really slow track, even for the worst horses. Some horses that are not being persevered with much, sure, but not those in contention.

The truth is the rate varies from race to race, fraction to fraction, and depends on a lot of things. There really is no one size fits all.

Exactly, The value of how long it takes a horse to run a length, or how many lengths a horse covers in a second depends on how fast the horse is actually running. Carroll finds that using 5 lengths per second more accurately describes harness racing than T-Breds. I find that the formula of 6 lengths per second to be the most accurate in most cases but there will be exceptions.
A more accurate method is to use that particular races' actual velocity in that race segment to calculate how long it to took to run the distance in beaten lengths from the leader and just add that to the leader's time.

Take it one more step. By using three points of call (1st, 2nd and final for sprints) we can compute the rate of deceleration and theoretically figure the velocity at any point in the race. Use the stretch call for verification. The math is laborious so it takes a computer.

I got a computer, it's not worth much, but I do hold it in very high regard.

There have been quite a few suggestions about reading material to get better acquainted with a game that’s loaded with variety of opinions. Although at first glance this material might seem valid, I for one can’t say for sure how successful the authors of these books might have been using their own methodologies.

If you’re really interested in getting some worthwhile information I would suggest reading the material by the 2 most prolific and successful players of all time:
Pittsburg Phil & Bill Benter
You’ll certainly find that there’s a sharp contrast between them and their tactics.

Of course Pittsburg Phil’s information may seem dated, but if you read between the lines you’ll gain a real appreciation of his overall philosophy and a practical approach to the game. The “Axioms of Pittsburg Phil:
http://colinsghost.org/2010/05/racin...phil-1908.html

If you’re looking for a more high-tech approach then read anything that was either authored by or written about Bill Benter.
(Computer Based Horse Race Handicapping and Wagering Systems: A Report by William Benter) https://www.scribd.com/doc/166556276/Benter
A recent interview:
https://www.bloomberg.com/news/featu...se-racing-code

If you’re interested in knowing more about the horses themselves you might also want to read about one of the most significant aspects of the game called “physicality” authored by Joe Takach.
http://www.icapper.com/Takach_Nov1.html
http://www.icapper.com/Takach_Nov7.html
http://www.icapper.com/Takach_Nov14.html
Entire Index: http://www.icapper.com/

Thank you Nitro for the link to Pittsburgh Phil.