18 January 2009
By Tom Reed
Horse International Vol 1 2009
In my recent article, "Every Man A Rembrandt! Every Breeder a Melchior!" (Horse International No. 5 – 2008), I discussed the pros and cons of several popular breeding indices and rankings and offered some concrete advice to the WBFSH on how to improve their rankings of sires. With the silly season now upon us and with studbooks and the WBFSH publishing their rankings and indices, I thought it would be valuable to return to this topic using the WBFSH 2008 Jumping Sires as an example.
Here are some problems that must be addressed by the WBFSH if their rankings are to have increased value to breeders:
1. The Number of Foals Sired by each Stallion Must be Disclosed.
The ranking of sires would be much more valuable if we knew how many foals each stallion has sired. It would be even more helpful if we also knew how many progeny each stallion has sired that has reached the age of six years. (For our purposes, the youngest a horse can be to earn points for his or her sire in international showjumping that will be included in the ranking is six years old).
It is possible to get an indication of the wide variation in the number of progeny by looking at data published by the KWPN and Holsteiner studbooks. Consider the differences in the number of progeny sired by these stallions:
Rank / Stallion / Number of Foals / As of
4 / Concorde / 2,727 / 2006
6 / Caretino / 1,485 / 2007
7 / Burggraaf / 3,446 / 2006
8 / Libero H / 567 / 2006
9 / Indoctro / 2,815 / 2006
11 / Cassini / 1,544 / 2007
13 / Calido / 776 / 2007
25 / Hamlet / 799 / 2006
27 / Mr. Blue / 126 / 2006
Note: The Number of Foals is only the number registered by the stallion's "main" studbook, either the KWPN or Holsteiner studbook.
If we are to make a clear and accurate assessment of the genetic worth of stallions we must be able to take into account the number of chances each stallion has had to produce international showjumpers. Burggraaf, for example, produced approximately 27 times as many progeny for the KWPN as did Mr. Blue. Yes, both stallions also sired progeny for other studbooks, but the numbers are striking and clearly have implications for the rankings.
2. The Average Number of Points Earned by Each Progeny Must be Disclosed.
I performed some easy but time-consuming calculations on several of the top 30 stallions. Darco, ranked 1st as a sire, has 58 progeny in the list of international showjumpers upon which the sire ranking is based. Darco's 8,411 points divided by 58 progeny equals 145 points per progeny.
Now let's compare the average number of points earned by each of several of stallions' progeny to Darco's progeny:
Stallion / Points / International Jumpers / Average Number Points
Hamlet / 2,230 / 1 / 2,230
Mr. Blue / 2,138 / 11 / 194
Libero H / 3,782 / 20 / 189
Caretino / 4,178 / 27 / 155
Burggraaf / 3,721 / 25 / 149
Darco /8,411 / 58 / 145
Indoctro / 3,567 / 26 / 137
Cassini I / 3,282 / 27 / 122
Calido / 3,120 / 26 / 120
To the casual observer Hamlet, ranked 25th, looks like a super sire. But by digging into the data we see that Hamlet has sired only one progeny contained in the WBFSH rankings of international showjumpers: the world's number one showjumper Hickstead (out of a mare by Ekstein). We also see that progeny of Mr. Blue and Libero H earn, on average, considerably more points than the other stallions on the above list.
Along with the average (mean) number of points it would also be useful if the standard deviation were published. Stallions with high averages and low standard deviations means that they tend to produce a lot of high-performing progeny and relatively few low-performing progeny. The current ranking system rewards stallions that produce a lot of low-level international showjumpers.
3. The Strike Rate of Each Stallion Must be Disclosed.
By "strike rate" I mean the percentage of progeny that become international showjumpers. The statistics I present below should be viewed with caution as they do not include all the progeny produced by these stallions (many of these stallions produce progeny for multiple studbooks) and they also include only the progeny jumping internationally in the 2008 season. But by limiting the analysis to the strike rate within each stallion's home or main studbook we can get a glimpse of their worth as a sire.
Stallion / Int'l Jumpers (Main Book) / Progeny (Main Book) / Strike Rate
Mr. Blue / 7 / 126 / 5.56 %
Calido / 16 / 776 / 2.06 %
Caretino / 26 / 1,485 / 1.75 %
Cassini I / 25 / 1,544 / 1.62 %
Libero H / 7 / 567 / 1.23 %
Indoctro / 23 / 2,815 / 0.82 %
Burggraaf / 23 / 3,446 / 0.67 %
Hamlet / 1 / 799 / 0.13 %
As noted above, the strike rate only includes data for the 2008 season. In earlier years, when the WBFSH pubished the rankings of sport horses and sires in book form, they also included 10-year rankings. It is unfortunate that these statistics have been dropped by the WBFSH as they would more easily allow us to compute a valid strike rate for each stallion.
I understand that breeders await the WBFSH rankings each year and these rankings influence the stallion choices of many breeders. I caution all of us to think carefully about the data and statistics and not to chase after the leaders in the rankings. Breeders should also recognize that many of these stallions serve as excellent sires in some studbooks and have much less success on other studbooks because of the differences in the mares. Indoctro, for example, appears to be more successful with KWPN mares (average number of points per progeny equals 153) than with mares in other studbooks (average number of points per progeny equals 137). In the case of Mr. Blue the results are even more striking: 263 points per progeny for the KWPN studbook versus 194 points per progeny for other studbooks. But again, caution is advised because some of the differences in average points earned may be attributable to the ages of the progeny.