Over on TrueNicks, I created a bit of a stir last week with the suggestion that the term "Sire of Sires", doesn't truly exist. You can read the post here by Alan Porter, but my thoughts are below in the comment field. After making the initial comment, I was asked to clarify my position, and I thought it worthwhile to make some further comment here to better explain my thoughts.
Firstly, it needs to be made clear that there are differences in the term "sireline" and "sire of sires". Obvious to some, but not to all. In regards to "sirelines", Alan in his post was referring to the historical classification of sirelines. In that respect the genetic data is quite clear. Over 95% of sirelines share the same Y chromosome sequence.
The only study (which you can read here for free) on the classification of the equine breed based on the Y Chromosome, did point to a mutation in the Y Chromosome that occurred somewhere between the father-son trio of Eclipse-Waxy-Pot8os with Whalebone (1807) being the common ancestor source based on the genetic data that they had in the study. This Y Chromosome haplotype now dominates 95% of the breed. The authors didn’t find any newer Y Chromosome mutation which would point to a more recent change in the breed. Thus, as far as a Y Chromosome is concerned (which presumably is the basis of a sireline), there is no Y Chromosome difference between Northern Dancer or Mr Prospector. They are one in the same sireline. It may make you feel better to classify them in terms of sirelines, but that is no different to using the sub-populations of family numbers (like say 1-x) which misleadingly signify a difference in mitochondrial DNA, where no such mutation/difference exists.
Now switching to the term "sire of sires" and why I think that it is a misnomer that hasn't got any real validity in thoroughbred breeding.
I guess the first aspect is defining the term "sire of sires". If you define it as a concept of the number of sons that retire to stud, then it is a popularity contest, not a concept of genetic transmission. If that is how you want to define a “sire of sires”, just based on the number of sons that retire to stud, then you can have at it. The commercial market, or more specifically the success of the first couple of sons to retire of a good stallion, defines your success there. For mine, I would define the concept of a sire of sires as 1) A successful stallion who 2) Left a number of successful sons whose stud career at least approached the success of their own sire.
So, let's take a look at Mr Prospector. He's a “sire of sires” right?
Mr Prospector has had 312 sons retire to stud worldwide that have sired offspring to race. Only 15 of these 312 have records that would indicate to me that they are at a minimum decent sire sons having sired at least 8 Gr1 winners in their career and having a SW/Fls% above 6%. That is not nearly as good as Mr Prospector himself, 14.8% SW/Fls and 47 G1 winners, but it is better than the average sire. The sons of Mr Prospector that make that grade are Alpha Plus, Carson City, Fappiano, Fast Gold, Gone West, Gulch, Hussonet, Kingmambo, Lode, Machiavellian, Miswaki, Numerous, Seeking the Gold, Smart Strike and Woodman. You can argue over the merits of some of these stallions who made their name in less competitive racing jurisdictions to argue against their inclusion, and you could knock out one and put a stallion like Placerville in who made a big impact in India but failed to meet the cutoff, but lets for now leave them in. 15 of 312 leaves a strike rate of 4.8% successful sons to retired sons.
What about Danzig? He's had 184 sons retire to stud to sire runners. Only 5 (maybe six) could be considered superior sires - Danehill, National Assembly, Anabaa, Green Desert and Belong to Me (with War Front being the other possible which we will include). That is 6 from 184 or 3.2%. What about his son Danehill? 202 sons to stud that have sired runners. Only Commands, Danehill Dancer, Dansili, Fastnet Rock, Flying Spur, Rock of Gibraltar and Redoute’s Choice could be considered truly elite – That is 7 of 202 or 3.4%.
What about Storm Cat? He's had 189 sons retired to have sired runners. Only 6 of these - Giant's Causeway, Bernstein, Pure Prize, Hennessy, Tale of the Cat and Easing Along could be called significant (and you could argue against Pure Prize). 6 of 189 or 3.17%. What about Sadler's Wells? He’s had 121 sons retire to stud to sire runners. Only 8 of these – Galileo, Montjeu, Fort Wood, Scenic, In the Wings, High Chaparral, El Prado and Poliglote are above the 6%/8Gr1 mark. That is 6.5%. There is an interesting aspect to Sadler's Wells as a sire of sires. He's had the least number of sons go to stud, mainly because putting In the Wings, El Prado and Scenic aside, generally speaking some early sons to retire to stud were good racehorses but very ordinary stallions (King of Kings, Old Vic, Johann Quatz, Entrepreneur, Dream Well, etc) which put a dampener on standing subsequent sons at stud until Monteju and Galileo came along and righted the ship.
So....the best stallions, considered by many as a "sire of sires", are having a strike rate of superior sire sons in a range of 3-7% of all his sons that retire. What that says to me is that there is next to no heritability of sire success, and thus, the concept of a sire of sires is a misnomer. If there was a true genetic occurrence of a “sire of sires” effect, then you would be looking at these stallions above having a strike rate of successful sons of at least double if not more.
Stallions become successful because they can sire high class horses. We then see a lot of sons of successful stallions retire to stud on the belief by breeders that they can pass the ability to sire good racehorses themselves on. They cant afford to go to the sire itself so they go to the son. Unfortunately this is not how the genetics of performance in the thoroughbred works. Just because their sire was a good sire, doesn’t necessarily make a stallion a good one also. Think about all those sons of Alydar. Equally, a mediocre sire has the chance to sire a very good racehorse and sire (e.g Tiznow).
We are doing a lot of work on the genetics behind stallion success. There are certainly some variations in genes that you want to see for a stallion to be successful, but they are nuclear genes, not Y Chromosome ones, so their inheritance is subject to Mendel. Furthermore, it is a lot more complex than just having these individual variants as some stallions are more successful breeding mares that share certain variants in distance related genes, and others in different mitochondrial families. It is very complex. That said, the more we are looking into the genetics of sire success, it is clearly apparent that you should treat every stallion as an individual chance of making it and the concept of “sire of sires” is just a romantic fallacy.