Stride Data

[martin_85]

Having not really had the time to follow racing the past couple of years, I'm looking to get back into it this year.

Whilst re-reading various books and publications, I always enjoy following Simon Rowlands views and data used by ATR etc. I'm sure it's an area that is already well trodden, but I wonder whether the use of his research on stride data can be used to identify horses that have started out at sprint distance but are more suited to a longer trip.

I'm primarily thinking of the AW, as that's what I tend to follow and the more consistent surface takes some uncertainty out of the equation.

A lot of horses start out over 5-7f, with Simon's data suggesting that horses displaying a stride frequency of <2.40 being better suited to 8f+. He also combines that with a look at stride length, discussing the fact that horses displaying a stride length of 25ft+ can show a horse of above average ability.

Now allowances for horses being 'green' need to be made, but could identifying horses that turn out in maidens with a set stride profile be used to follow them for when they take a step up in trip which may be better suited to their natural ability?

I am thinking:

1. Stride frequency <2.40 (possibly take some ambiguity out of this and make it <2.35)
2. Has initially run over 5-7f
3. Stride length >25ft
4. High FSP (>98%) to show they were still accelerating and had not completely collapsed in the race
5. Below average top speed for the field, to show the horse has not peaked within the race duration

The key will then be narrowing the resulting horses to identify how effective a step up in trip may be.

I'm aware that there is a genetic layer that could be added, in that I believe most sprinters tend to be 'C/C', middle distance 'C/T' and stayers 'T/T'. I, however, don't have access to any bloodstock databases and suspect they are fairly expensive to obtain. This would however add weight to the argument over whether they are ill-suited to a sprint distance.

Trainers are obviously aware of all this, so will factor that into their training and race placement for their horses. Attempting to create a profile of trainers who have a higher than average win % when horses step up in distance, seems the logical next step to this.

I use HRB, so need to give some more thought as to how I can create some systems to profile AW trainers. Criteria such as:

• last run being 5-7f
• in a maiden with less than 3 starts
• next run being 1m+
• finishing position 4th or worse in previous race

I need to give it some more thought, but if anyone has gone down this route already and has advice (or warnings!), please do chip in.

 

[Chesham]

Hi martin_85

This might help your research

 

[Larry]

There is a site that is ATR & Simon Rowland affiliated I don't know if I'm allowed to name it but it's easily googled, also the site says you can send in the name of any horse that has had a run & it will tell you that horses optimum distance

 

[martin_85]

Thanks for that Chesham

A stride frequency of <2.3/2.35 makes more sense off that data. Good to see the longer stride length generally providing a higher class of horse.

Presumably in the example of Poetic Flare, this would have resulted in a decent time figure but a poor efficiency rating.

 

[martin_85]

There is a site that is ATR & Simon Rowland affiliated I don't know if I'm allowed to name it but it's easily googled, also the site says you can send in the name of any horse that has had a run & it will tell you that horses optimum distance

Thanks for that Larry I'll have a look, as the answer doesn't obviously spring to mind!

 

[martin_85]

Thanks to Larry and his recommendation, I have advanced the workings to establish a somewhat practical first step in the process.

Using RaceIQ data, which rather than peak cadence, uses a minimum/average/max frequency we can combine that with the race pace to hopefully establish some optimal distances.

As an example, looking for a horse with 1-3 runs in a maiden over sprint distance, Cotai Eye Joe in the 18:00 at Wolverhampton on 16/02/26 is flagged as having an average stride length >25ft.

The FS% and race par show this race was run slowly with a sprint finish, which adds a level of caution to the result, but the optimal distance calculator shows a suggested trip of 1m2f.

I need to work this through to get a decent sample size. If horses have multiple runs then it should provide a clearer picture, as hopefully at least one race will be more evenly run.

The next step will be working out how to assess the horse once it is entered in a trip which we feel better suits it. One race gives very little to go off, particularly if over a completely unsuited distance, but 2-3 races will at least provide some evidence.

The graphic that Chesham posted above indicates that an average stride length of 7.68 puts the horse in the upper range of the 66-87 bracket, but again that is only based off one race and must be treated with caution.

Something to work on at least.