Optimal Pace

[Magister Ludi]

When does a horse run closest to his species’ optimal pace? When he’s setting a record.




f------name------------track---date-----final-----2f-----4f-----6f-----8f----10f----fin

5.5----Plenty Zloty-------TUP---4.18.95----61.2----21.6---21.8----------------------------17.8

6------G Malleah---------TUP----4.8.95-----66.6----21-----22-------------------------------23.6

6.5----Lucky Forever-----HOL---5.20.95----73.24--21.96--21.95--23.22-------------------6.11

7.5----Awesome Daze---HOL---11.23.97---86.26--22.19--22.6----23.93------------------17.54

8------Najran-------------BEL----5.7.03-----92.24--22.65--22.02--23.09------------------24.48

9.5----Farma Way--------PIM---5.11.91---112.4---23.55--23.36--23.16--24.25----------18.23

10.5--Gold Star Deputy--AQU---4.10.99--127.32--23.13--23.74--24.22--24.48--24.74---7.01

11----Demi’s Bret--------AQU--10.26.97--132.31--24.49--22.73--23.94--24.58----------36.57

North American Dirt Records 2002 [close enough]




f------2f----4f---6f--8f--10f--fin
5.5---103--102------------------94

6-----106--101------------------94

6.5---103--103---97------------92

7.5---104--102---96------------98

8-----102--105--100------------94

9.5---101--101--102--98-------97

10.5--105--102--100--99--98--86

11-----98--106--100--98-------99

Split Speed as Percentage of Average Speed



2f----4f---fin

103--102--94

Median Split Speed as Percentage of Average Speed



It appears that the optimal pace for a thoroughbred racehorse racing on a dirt track at a distance between 5.5f and 11f inclusive would be to run the first 2f at about 103% of its average speed with a fairly linear deceleration throughout the race to 94% of its average speed. Comments?

[guckers]

Originally Posted by Magister Ludi

It appears that the optimal pace for a thoroughbred racehorse racing on a dirt track at a distance between 5.5f and 11f inclusive would be to run the first 2f at about 103% of its average speed with a fairly linear deceleration throughout the race to 94% of its average speed. Comments?

I like the concept, could it be validated with more than the record setter?

[DeltaLover]

Originally Posted by Magister Ludi

the first 2f at about 103% of its average speed with a fairly linear deceleration throughout the race to 94% of its average speed. Comments?

I disagree about the linearity of deceleration.

The way I attacked this problem in the past is by the creation of a distance to average speed of winner function and then calculating the Least Square line which was a good fit for distances up to a certain limit. I do not exactly recall what this limit was, but it was certainly a division between sprint - route distances. Beyond this distance a linear approximation was not a good fit.

We can come up with several conjectures to explain why this is happening. For example the geometry of the track might be a good reason. Overall we should not care why this is happening rather that it is a fact.

[Robert Fischer]

Originally Posted by Magister Ludi

It appears that the optimal pace for a thoroughbred racehorse racing on a dirt track at a distance between 5.5f and 11f inclusive would be to run the first 2f at about 103% of its average speed with a fairly linear deceleration throughout the race to 94% of its average speed. Comments?

I think it is an interesting way to look at the pace.

Without considering run-up distances, and turn configurations these numbers will look a little different. You will want to consider them estimates in any case.

You may also want to consider percentage of peak spead, in addition to percentage of average speed.

[Magister Ludi]

Originally Posted by DeltaLover

I disagree about the linearity of deceleration.

The way I attacked this problem in the past is by the creation of a distance to average speed of winner function and then calculating the Least Square line which was a good fit for distances up to a certain limit. I do not exactly recall what this limit was, but it was certainly a division between sprint - route distances. Beyond this distance a linear approximation was not a good fit.

We can come up with several conjectures to explain why this is happening. For example the geometry of the track might be a good reason. Overall we should not care why this is happening rather that it is a fact.

It probably has something to do with the fact that at about 50s, aerobic power becomes larger than anaerobic power. Note the sharp drop in the final split for the shorter sprints. That may reflect the fact that record-setting sprinters are anaerobically gifted but aerobically challenged.

[DeltaLover]

Originally Posted by Magister Ludi

aerobic power becomes larger than anaerobic power.

This is an interesting statement.

Of course for our purposes as handicappers and bettors, we should not spend our mental energy try to explain why something like this is happening. It is sufficient the fact that we can detect it..

[HUSKER55]

I have a question. instead of using an average speed, wouldn't using final time or call1.

just curious.

thanks

[DeltaLover]

Originally Posted by HUSKER55

I have a question. instead of using an average speed, wouldn't using final time or call1.

just curious.

thanks

yes, we can use final time. The only difference is in this case the slope will be positive while if using speed it will be negative.

This mean that you should interpret your data differently but essentially the result will be exactly the same.

First call or any other fractional time presents handicapping challenges and should be avoided.

[HUSKER55]

thank you

[Robert Goren]

The problem with getting consistent run up distances even at the same track and distance let alone multiple tracks and multiple distances makes most of this conversation moot.