The Class Distribution
by Roger Lyons
Of the well over 15,000 winners of unrestricted North American stakes and blacktype-qualifying foreign stakes from 2001 through 2009, about 45% won a graded/group stakes during their careers. G1 stakes were won by about 13% of those stakes winners, and about 26% were G1 or G2 winners. Those three numbers–13%, 26%, and 45%–should be kept in mind when assessing breeding methods. That’s why the online eCompuSire facility, which is available by subscription at eNicks.com (after you sign in to eNicks), allows users to select lists of G1 winners, G1-2 winners, G1-3 winners, or all stakes winners when researching breeding methods. I don’t mind promoting mentioning that program because I designed it and benefit from subscriptions, just so you know.
Any stallion whose record of graded/group stakes production corresponds with those percentages is going to be very busy come breeding season. Stallions whose stakes production exceeds those numbers will bring in a lot more revenue for the effort while stallions that fall short will bring in less.
That’s obvious, but, just as these percentages indirectly determine the value of a stallion, they can and should be used directly in the assessment of breeding methods. I’ll go so far as to say that knowing how much opportunity a given breeding method has had pales by comparison with the importance of knowing the proportion of graded/group production among total stakes winners resulting from that method.
For example, the stallion Malibu Moon has had three-year-olds and older from 149 mares with Mr. Prospector in their ancestries, and only six of those mares produced SWs for him. It happens, though, that those SWs include three G1 winners–Declan’s Moon, Malibu Prayer, and Devil May Care–along with G3 winner Odysseus. That’s 50% G1 winners, 50% G1-2 winners, and 75% G1-3 winners. Class trumps opportunity every time.
Contrary to prevailing impressions, most methods of relatively close inbreeding yield graded/group distributions that fall well below the population norms of class. According to eCompuSire, 42 stakes winners to date have been inbred to Seattle Slew, including two G1 winners (Hollywood Starlet S. winner Turbulent Descent, by Congrats, the most recent), six G1-2 winners, and 12 G1-3 winners. That’s 4.8%, 14.0%, 28.6%–far below the 13%, 26%, 45% distribution in the stakes population as a whole. Does that mean inbreeding to Seattle Slew is a bad thing?
There are no absolutes. If you sort the list by “Pedigree” and look at individual sires, you’ll see that Tiznow accounts for four SWs inbred to Seattle Slew, two of them out of the same mare, including a G1 winner and two G2 winners. Hence, Tiznow’s class distribution is 25%, 75%, 75%. That’s very competitive with the 21%, 55%, 66% proportions for his overall stakes record, and Tiznow’s overall class distribution happens to be very close to that of Unbridled (29%, 46%, 61%), remembered as the consummate big-horse sire. That Tiznow can make a method of inbreeding look good doesn’t mean any stallion can.
Frankly, almost any breeding method you can think of will appear not very effective when assessed with indifference to the variety of pedigree contexts in which it has been applied; but, for almost all plausible breeding methods, conditions of possible effectiveness can be found. The norms of class I have described here provide a measure you can use, along with eCompuSire, to discover those conditions.