Using TrueNicks to Predict Racetrack Success
Written by Byron Rogers | Mar 05, 2009 |
Comments
With the first Northern Hemisphere 2-year-old races
set to get underway shortly, we thought that this year we would do something
that allowed us to follow TrueNicks as a predictor of racetrack success. We debated a couple of ways we could do this, with the primary goals of offering a statistically sound study and one that is trackable in 2009.
An ideal solution is to take a "snapshot"
rating when a book of mares first visits a stallion, prior to the foals even
hitting the ground, and following these ratings through for the lifetime of the
horse. The lag time between the marriage of sire and dam and when their resulting progeny eventually hit the racetrack, however, means that such a study is years in the making. (An "historical" TrueNicks tool is on the "to do" list -- it will allow us to compute the nick rating that a cross would have received on a particular date in previous years. In the meantime, we are dealing with "present tense" nicks.)
With these constraints, we selected a handful of
stallions and ran their entire 2-year-old crop (foals of 2007) through the TrueNicks algorithm, returning all the ratings of each individual horse as of Jan. 1, 2009. We'll use this to
track this foal crop through the year -- it allows us to focus on whether the ratings change, and if so, why they change.
The seven stallions we selected are the first-crop sires Afleet Alex (TrueNicks,SRO), Eurosilver, and Roman Ruler (TrueNicks,SRO); the proven duo of Johannesburg
and Stormy Atlantic (TrueNicks) (both Storm Cat-line stallions); and the ill-fated Maria's Mon. Over the upcoming weeks we will post up some comments about each of the stallions and start to track
their runners.
Prior to this individual analyses, there
are some interesting facts on how the ratings look from a general population
viewpoint.
There are 879 individual horses in the
study. Of these:
- 248 (28%) rate A or better on the
TrueNicks scale. Those who have been reading this blog for some time, will have already picked up that this is an inordinately high number of A-rated horses. A couple of valid reasons exist for this number being generated.
- Firstly, in
general, the "commercial population" has a skew towards higher ratings
than the population at large. In almost all the other studies that we have
done, while the general population of "As" is usually about 13% of all foals,
we are finding that this is closer to 20% in the commercial market. This is primarily a byproduct of commercial breeders and buyers using nick ratings as a guide to their breeding and purchasing decisions.
- Secondly,
proven sires have a much higher percentage of A-rated foals than the unproven. To
wit, Johannesburg has 150 foals in his
current 2-year-old crop and 66 of these, or 44%, are rated A or
better. It certainly makes it harder for TrueNicks to be used as a discriminator
in his case but does highlight the fact that success breeds success as once a
stallion starts to establish a pattern, breeders and stallion managers focus on
getting more representatives from those particular broodmare sire lines to his
court.
- Moving down the scale, 2-year-olds rated B form 27% of the population;
- C rated 16%
- D rated 22%
- and F and/or "No Rating" yielded a smaller 7%.
Interestingly the three first-season sires have the greatest percentage of their crops rating
D or below, with Afleet Alex topping the list with 42% of his entire crop with these low ratings.This is a reverse of what we see in proven sires, and is partially accounted for by first-season sires recieving a far more diverse representation of broodmare sires than the proven stallions.
As a matter of interest -- overattention to variant scores is not a great practice in general but makes for some fun reading -- Maria's Mon has the highest-rated 2-year-old by
variant with the colt out of Cricket Wicket being an A++ (variant
of 270.061) to his name. This colt failed to find many admirers when he went
through the ring last September, falling to the bid of California trainer Mark
Glatt for just $6,000. It will be interesting to see how he fares later in
the year.
comments powered by