CP your a dead set legend, thanks heaps!

Definitely fills the gaps, I guessed that the SD was about 2 for <1100 however I thought that it would increase from there up to 1800 and above but it seems that the standard deviation remains pretty consistent while only the starts this prep changes.

Again thanks for the data, and no need to apologise on the delay, im a serial offender of being busy with other tasks.

Here's some food for thought for either yourself or others.

From a small data set I took last year of non-maiden races, today I looked at horse winning %'s, their average and their standard deviation.

What is of interest is the following:

NOTE: The reasons for <1.15 Standard Deviations is that any horse below this win% makes up for just 12.50% of all winners for that code - something that may assist in making lay decisions

Code - Mean - SD - <1.15 SD's

SR - 33.84% - 19.47% - 11.45%
MR - 33.18% - 16.28% - 14.46%
AR - 27.16% - 18.23% - 6.20%
VR - 27.82% - 16.69% - 8.63%
QR - 21.29% - 12.42% - 7.01%
NR - 25.57% - 16.02% - 7.15%
PR - 26.15% - 17.13% - 6.45%
BR - 30.22% - 15.90% - 11.93%

Certainly shows that the tracks which generally have the higher quality races (MR, SR, BR) means you can accept a higher win % for your lay selections. Or conversely at the other tracks you can accept lower win % for your backing.

Anyway obviously this isn't as definitive as it could be because it should be based on tracks not codes however I don't have the data for that. Anyway just thought i'd share since its something i was working on today.

Cheers