Early Speed Points

[Tom]

Got to thinking about this a lot, with the new book out, I did some research to see what was really going on. I always objected to giving one point for running first and the same weight to running third.

I looked at 17,819 6 furlong races on dirt, any condition, run at all tracks since June 1 2005 to November 30, 2005.

I looked at running position at the 1st call, and at lengths either behind or in front to see how important each was in predicting the 1st call postion in today's 6 furlong race (last race was used, no matter what distance, surface, how long ago, class, the whole enchilad - last race only period).

Here is what I found:

[Tom]

This is it----can't get that damn formatting to work....isn't it [code] [code\]?

[Tom]

Using this data, I am thinking I want to give 4 points to horse leading by over 2 lengths, and only 2 points to a horse 3rd by 2 lengths.

Any ideas on how to prceed, suggestions?

I am looking to do this for all distances, and maybe use field size instead of 1,2,3 positions - like divide position by field size? More work, but might be more predictive.

Also, I did not look at any prices - just predictability to lead at the first call today.

Maybe I would be better off using winning as my goal?

[BillW]

Tom,

Don't forget to take into acct. the dist. last! Your study should probably use only 6f last and 6f today then deal with projections from differing dist. last samples.

Bill

[Tom]

Thanks, Bill...I wanted to take a fast snapshot to see if it was worth the work ( you know how dediucated I can be to work!:rolleyes: )


I especially wnat to look at what route distances a horse can drop back from - lots of old, unproven yet ingrained ideas about distance floating around in my nut.

I think I need to lot at the expected percentage of winners and compare actual to that. Got some good ideas reading The Mathematicsw of Horseracing by Dade Fogel and from Klien's new book. I think if I have to calculate KSP, I will miss a lot of races! LOL!

[mcikey01]

Tom: Intriguing stuff...But, so many "what if"'s come to mind...Average vs. expected winners is important and so is field size.. How about post position in last race and today, how about prorportion of field that lead at 1st call last race or proprortion of field with BL levels the same or lower, how about track layout and post position bias or distance at 1st call run on the turn, how about average 1st call position within last 30 days etc... Good start by you... Keep in mind the importance of controlling race variables....Hope it doesn't drive you bonkers.

[andicap]

I think you can variable yourself out of a good method.
The original Quirin speed points probably didn't take all these variables into account -- it was simple (KISS) and it works fairly well.

I would love to see the results broken down by distance first to see if there is appreciable difference with the original results.

Once you start getting into variables you can go crazy --
there's no end. How about differences by sex? age? surface(d-t, s-r), etc. You can drive yourself batty.

To me it's like handicapping a race -- I can look at a handful of factors and do a race in 5 minutes or I can spend 30 minutes and look at 100 factors that logically play a role in the outcome of a race.

I didn't say don't look at ANY variables but after a while it can become an endless loop.

[Vegas711]

Good work TOM.

Maybe there are a few more early speed factors that we can check out. I always have been interested in knowing how much does a speed duel increase the likely hood of a closer winning.If you have 3 speeds is this more likely to set up a closer than when you have only 2. I am sure others can come up with more things to check out.

[Vegas711]

Here is another early speed situation to check out.

Compare the second paceline back and compare it to the most recent paceline, also can look at the 3rd paceline and compare it to the other 2.

My theory is that the most recent paceline will be more indictive in determining early speed position.

[Overlay]

Summarized findings from a study of 45,000 past performance lines in Quirin's Winning at the Races:

Among horses not changing distances radically (from a sprint to a route, or from a route to a sprint), only three out of every five that were running first, second, or third at the first call their last time out were able to duplicate that in their next start. This figure was only two out of five for horses dropping back from a route in their last start to a sprint today. For horses stretching out from a sprint in their last race where they were running first, second, or third early to a route today, it was four out of five. Quirin suggested that the lack of greater race-to-race predictability in early-speed performance for individual horses was understandable considering the hectic scramble for position at the start of any race, where the loss of a few steps at the break could mean the difference between being among the early leaders or being in the middle of the pack. This relative lack of race-to-race consistency was one of the reasons he used multiple past races for his determination of early-speed tendencies.

Of course, the landscape may have shifted since the time of Quirin's study, and also as a result of his findings. In light of the established and well-publicized importance of early speed, trainers may now be conditioning with an eye toward having their horses more consistently attempt to contest the early lead, with the result that past early speed (whether displayed in the last race only or in multiple prior races) would not be quite as predictive of a winning effort as it used to be. Horses with early speed may still be winning just as often, but there will be more horses trying to run on the lead or close to it in the early going than there used to be, so it would be harder to project which one will eventually win.

[mcikey01]

I have rudimentary knowledge of horse conformation. Is there anything in thoroughbred body structure characterisitic of advantageous early speed? (I keep thinking about the differences between quarterhorses and thoroughbreds)

[46zilzal]

I have rudimentary knowledge of horse conformation. Is there anything in thoroughbred body structure characterisitic of advantageous early speed? (I keep thinking about the differences between quarterhorses and thoroughbreds)
short coupled animals are quicker than their leggy counterparts, but that alone will not make a difference I'm afraid.