About Stallion Spectator Ratings (SSR)
Financial success in the thoroughbred business is a challenging affair. No matter which sector of the business you're in – racing or breeding – the odds are stacked against success in any individual case. The principle underlying the APEX ABC Index is that roughly 8% of horses which actually get to the races make money – just in comparison to their training costs, never mind their purchase price. The breeding side isn't that great a bet, either: over the last 30 years, we've calculated that the percentage of yearlings at the best auction sales which make money varies between 14-22%, depending on the strength of the market; at best, two out of nine make money. No matter how much we might all love the game, very few people have such deep pockets these days that they can afford to ignore the bottom line.
As you know, good statistical data and analysis which have practical application in the real world are vital components in the quest to achieve at least respectable financial results in the horse business. Nearly 30 years ago Bill Oppenheim developed the APEX (Annual Progeny Earnings indeX) method of rating sires. One of the key indices is the A Runner Index, which measures the frequency with which a sire gets runners whose earnings, calculated on an annual basis, figure among the top 2% of earners, from all runners, in each jurisdiction covered.
The A Runner Index has proved over the years to be a useful predictor of future continued sire success; going back 30 years I can remember APEX ratings saying very early on in their careers that horses like Pleasant Colony, Private Account, and Storm Cat were going to become very successful sires, and APEX ratings said Smart Strike was a top sire when he was still standing for $35,000. On the flip side, the one thing APEX ratings were never designed to do was to make finer distinctions. Roughly 4.00% of named foals of racing age by sires which get APEX ratings (777 of them at the end of 2018) become A Runners at any stage of their careers, which is about the same percentage of foals which become Listed or Graded/Group Black Type Winners. We needed to be able to discover further refinements.
In the anarchic universe of thoroughbred sire statistics, in which not much other than the total numbers of foals, runners, winners, black-type and Graded/Group winners/placed, and progeny earnings are really official, it is then left to individual practitioners to devise our own definitions of the highest-class horses. I had suspected for some time that Group/Grade 1 and 2 winners was the correct description I was looking for- though Group 1 winners really are the top of the pyramid, they alone are just too small a sample on which an operation can devise strategy: in other words, you can't just shoot for Group 1 winners. Moreover, though they are very much the exception rather than the rule among Group 2 winners, the fact that such successful sires as Distorted Humor, Pulpit, Dansili, and War Front were only Group 2 winners is another reason we should be measuring Group 2 as well as Group 1 winners. Sure enough – even though we are measuring not by earnings, as in APEX, but in black type – the percentage of G1/G2 winners to named foals is right at 0.80%, meaning the ratio of G1/G2 winners is 20% of the number of Unique A Runners. This looks to us like the correct ratio: one out of five Unique A Runners is a truly 'elite' runner.
Having established those categories, we now felt we had the right information to begin creating statistical models which would constitute a better sire rating. The first piece of constructing our SSR ratings was to create the SSR SIRE Points scale, which measures what the sire has achieved in terms of production of A Runners and G1G2 winners. If you only want to look at an improved version of the traditional sire rating, this is it.
But that's only the half of it, literally as well as figuratively; there's also the question of the mares the sires covered. How should they have done, given the caliber of mares they covered, and then, how did they do? This is another highly complex question, but we made one key discovery and one key decision which enabled us to analyze this all-important question. First, we established that there was an extremely high correlation between the two data elements which most analysts use to determine mare quality: the CPI (Class Performance Index) – which is based on the mare's race record – and the CI (Comparable Index), which is based on her production of racing stock. This meant we need not use both Indexes, and the decision we made was to use the CPI, since that does not change over a mare's production period, whereas a mare's CI can. For our purposes, the CPI is the more 'stable' statistic to use.
This analysis results in the SSR MARE Points scale, which rates sires on how they have 'utilized', if you will, the mares to which they were bred. Our thumbnail summary of the CPI numbers is this: 1.00 is average for the whole population, but that includes all sires, not just APEX or 'commercial' sires, both of which are more restricted populations. We estimate 2.00 is about the true “commercial average” CPI. When we combine the two indexes, SIRE plus MARE Points, the result is our signature SSR Sire Rating List. Given the mares out of which they produced foals, and given their records as sires using some key metrics, we conclude this is the best way to find the best sires in North America and Europe.
We could have stopped there and been quite satisfied with having produced the first revolutionary new way of measuring sire success in recent memory. But researchers are pretty famous for asking questions, and the one we asked was: now that we know we can identify the top up-and-coming younger sires as well as the more established ones, what happens when these young sires start getting the better mares? We didn't get the answer we expected: not only do they not improve when they start to get the better mares, there is compelling evidence that the top young sires, as a group, are never as good again as they were before we knew they were good. At best they can continue to be nearly as good as they were to start with, but the suggestion is that they actually might need to be attracting almost continually better mares to maintain their levels of success.
This is staggering information, although, in a sense, typical of the difficulties inherent in achieving success in the horse business. Here we are: we discover how to identify the top up-and-coming young sires, but then, in typical horse business fashion, it's not as easy to capitalize on these discoveries as we might have thought. Formulating strategies are still difficult: commercially, breeders still have to use the top (and most expensive) sires available, even though the evidence is clear that their actual success rate of producing A Runners and G1/G2 winners may suffer.