Draw Help Needed

[bluebook]

Would I be correct in saying that, since the stalls are placed on the inside at Kempton, in a ten runner race stall 10 should be entered as 1, stall 9 as 2, stall 8 as 3, etc when compiling draw stats ?

Thanks in advance.

blue

[Prufrock]

Yes.

[Prufrock]

You also ideally need to allow for the stalls being shuffled across when there are non-runners, unless those non-runners come late in proceedings.

Good luck with that project.

[bluebook]

Meaning, I suppose, that in a twelve runner race in which stalls 11 and 10 are non runners, 12 remains 1 but 9 becomes 2, 8 becomes 3, etc ?

[Prufrock]

Yes.

Difficult to do this retrospectively, but it is something that can be monitored on an ongoing basis.

[Artemis]

I,m a bit puzzled here, chaps.

I haven’t taken much interest in the Kempton racing as a punter, although i’ve watched it quite a lot.

You know, I haven’t taken much notice of how the stalls were numbered, maybe because you don’t see the horses head on as they leave the stalls(or do you?)

I just presumed that the stalls are ALWAYS numbered from RIGHT to LEFT as you face them from the front or LEFT to RIGHT as they are loaded behind. Is this not the case at Kempton?

[Prufrock]

That is correct, Artemis. Stalls are always numbered from left to right as the horses enter them.

The consequences for right-handed tracks, and for tracks where the stalls are against the stand rail on the straight course (such as usually at Ayr), are that the rail draw is the highest not the lowest number, and as such the stall number that occupies that position varies according to field size.

You need to recode any draw data into a back-to-front draw in such instances to take that into account.

E.g. at a mile at right-handed Kempton, in a 6-runner race stall 6 is next to the inside rail. In a 16-runner race stall 16 has that position. Stall 1 is 6 off the rail in the first instance and 16 off the rail in the second.

You need to recode draw as (n+1) – d, where n is number of runners and d is draw number.

[Artemis]

Thanks for that, Prufrock.

I know that high numbers are generally preferred on right handed round courses like Kempton and also at courses such as Goodwood and Beverley over 7/9 furlongs.

I presume you must be referring to some formula you use to work out the historical effect of the draw. Otherwise, you would know from the course, the size of field, and any race analysis, what effect the draw might have had in a particular race.

[bluebook]

Thanks for all your help Prufrock. Very much appreciated.

Spent ages compiling the data for all distances up to and including a mile at Kempton and have just finished compiling the respective IVs for each stall over 5f. The sample size I have for this trip is no where near big enough, taking in just 26 races, but the results, if correct, are interesting. Bearing in mind the fact that stall 1 here will always be the closest to the inside running rail, i.e. stall 12 in a 12 runner race, stall 9 in a 9 runner, etc, you find the following :

Stall 1 : 1.5
Stall 2 : 0.4
Stall 3 : 1.1
Stall 4 : 1.8
Stall 5 : 0.0
Stall 6 : 0.4
Stall 7 : 1.1
Stall 8 : 1.1
Stall 9 : 1.7
Stall 10 : 1.2

The average IV for stalls 1-5 is 0.8 and for 6-10 is 0.9, which in itself suggests there’s very little in the way of a bias, but it is interesting to note that the middlemost draws, 5 and 6, have such a lousy record.

Does this chime with your experience of Kempton at the 5 furlong trip ?

blue

[Prufrock]

I prefer average cumulative lengths beaten per runner or % of rivals beaten – both rounded to include the stall(s) either side and not just the stall in question itself – rather than concentrating just on winners.

With both measures, I had horses slightly off the rails doing better than those on them, and at a couple of distances I had widely drawn horses doing best of all.

That was about 6 months ago, however, and I haven’t revisited the data in the meantime.

[bluebook]

That’s the method Peter May recommends, but since I’m cranking these out by hand (…) I’ve settled on the Mordin/Gibby method.

Again, thanks for your input.

[bluebook]

Ended up with the following stats for 6 furlongs at Kempton from a sample size of 82 races :

Stall 1 : 1.0
Stall 2 : 1.7
Stall 3 : 1.5
Stall 4 : 1.2
Stall 5 : 0.9
Stall 6 : 0.6
Stall 7 : 0.9
Stall 8 : 0.4
Stall 9 : 0.7
Stall 10 : 0.3
Stall 11 : 1.5
Stall 12 : 1.9
Stall 13 : 0.0

The average for stalls 1-4 is 1.3, for stalls 5-8, 0.7 and for stalls 9-13, 0.9.

[Author]

Posts