cj55,
I was looking at that question and well as a couple of others (i.e. raced and unraced broodmares as ascribed to by Joe Estes and his theory of A.E.I.) and decided to dust off my Statistics 2000 course book from 3rd year university and see what I could find out.
Using the Kentucky Derby as my "control race" I went from 1947 to 2007. I picked 1947 because that is a "recognized" expansion of the thoroughbred population in North America; It is also ideal as a restriced race (the horses are all "3" year olds) run at the same distance on the same course on the same day; records are also fairly well kept as far as racing performance of the dams (though there are a couple of ? on my chart). I also examined all the derby winners to see if they were closely inbred (4 Generations is what I decided to call close). I used the pedigree query form and this is what I came up with.
80% (48 of 60) of all derby winners had dams who had raced and won, 7 of the 12 non-winners had raced, leaving 5 of the 60 to have not raced.
Now, as far as inbreeding - I found that 40 of the 60 Derby winners were not inbred within the first 4 generations, 23 of the 40 not inbred within the first 4 generations are inbred on the fifth with at least one 4th generation ancestor.
As an aside, Real Quiet is "quite" inbred by comparison (Cequillo: X4S X5S, Dr. Fager: X3S X4S, Raise A Native: X3D X4S, Rough'n Tumble: X4S X5S X5D, Nearco: X5S X5D) to the rest of the population.
Now, the question to ask but I am not sure of if it is answerable is would be what percentage of the overall population matched those trends in the thoroghbred population.
This is also flawed in that it cannot factor injuries etc to the three year old population aimed at the race, or exclusion to the race based on earnings.
So... I think the answer is it depends.
