Ratings Generated in 2014
Featured Stallion

Frankel's First Mating: Lots of Inbreeding

15.33%. That's the coefficient of relatedness between Frankel and Chrysanthemum, the first mare scanned in foal to the champion racehorse during his first season at stud. As a point of comparison, first cousins are 12.5% related and half siblings are 25% related.

Chrysanthemum, owned by the Coolmore team of Tabor, Magnier, and Smith, won a pair of group III races and ran third in the Pretty Polly (Ire-I). By Danehill Dancer out of a Sadler's Wells mare, she is bred on the reverse Sadler's Wells/Danehill cross of Frankel, creating 3x3 inbreeding to both of those influential sires.

Heavy inbreeding is rare in modern Thoroughbreds, but 2008 champion 3-year-old Big Brown (TrueNicks,SRO) comes to mind as a recent top racehorse with a fair amount of inbreeding—his sire and dam were 10.35% related.

With the amount of Sadler's Wells and Danehill blood present among top class mares in Europe, such inbreeding might become a theme for Frankel's stud career.

Click to view report for Frankel-Chrysanthemum

Filed under: , , , , ,


It certainly will be the theme for the early career at stud for Frankel. As we all know Coolmore is a big professional organization with no shortage of money so they can afford to take a chance with this kind of close inbreeding and no doubt other rich outfits will be doing the same. It's only when these youngsters make it to the racetrack will we find out if it was the right thing to do.

John T 04 Mar 2013 9:36 PM

Chrysanthemum is an extremely promising broodmare prospect-just the kind I'd love to own. One never knows how long a broodmare's career may be, so to risk an early (in career) mating with a mare such as this with the likes of a Frankel, and with this degree of inbreeding, seems an imprudent move. My prediction: Chrysanthemum will prove to be a more successful breeding animal than her 2013 mate.

sceptre 04 Mar 2013 10:45 PM

If you think the Frankel/Chrysanthemum A+ Variant 4.49 has lots of inbreeding. Look at how a little Australian breeder using True Nicks hypothetical foal produced an A++ Variant 8.98. Sire: Reset(AUS)00 Dam: Glowing Tribute(AUS)02.

Sire: Reset Chief Earner: Pinker Pinker(AUS)$2,687,624

Broodmare Sire: Zeditave Chief Earner: Jumbo Star(AUS)$6,621,426.    

Bill Matsinos 05 Mar 2013 6:54 AM

Hi Bill,

Interesting cross, and along the same lines as Frankel-Chrysanthemum, with sire and dam being reverse crosses. Many times when you cross a stallion over a mare from the same sire line as his dam, you'll get a high nick rating since the sire himself often helped establish the nick in the first place.

I see the resulting 2011 colt's name is Master Reset. What are your opinions of him?

Ian Tapp 05 Mar 2013 10:44 AM

It is a shame that such a talented individual as Frankel may have trouble finding mares not too closely related. Shuttling would be a good thing in this case, but with his value, i doubt they'll do that. Hopefully, there will be enough breeders motivated to import mares for him. And it might be a good thing to broaden the gene pool by bringing in southern hemisphere mares or German bloodstock.

Karen in Indiana 05 Mar 2013 10:48 AM

This is exactly the sort of thing we in the thoroughbred industry need to get away from. All of this inbreeding is just continually watering down the stock and making the horses more fragile.

Ange 05 Mar 2013 10:54 AM

Although I admire the boldness of such a breeding decision I would not have planned such a mating in a young mare. I want to breed a young mare to a proven stallion or cross to evaluate her quality in those first 3 - 4 years as a producer. Although the nick is highly rated the 3x3 crossed x2 is too heavy for me. I wish them all the luck!

Smugglers 05 Mar 2013 11:03 AM


It seems to me that you have been reading too many uninformed internet forums. There is no evidence to suggest that there is a correlation between inbreeding and unsoundness. It is popular to say, but it is scientifically wrong.

The thoroughbred is quite inbred when compared to some other performance breeds, but it is a long way off being a problem. If it were, the first indication would not be unsoundness either, it would be fertility.

Inbreeding does not equal unsoundness.

Byron Rogers 05 Mar 2013 12:00 PM

The wisdom in the old days was breed a young mare to an old stallion and a young stallion to an old mare. Since you had a handle on what the 'old' partner could produce, you could tell if the 'young' partner was adding or subtracting to the norm. Easier to find and discard the Mogambos of the breed before they did serious damage to the gene pool.

Pedigree Ann 05 Mar 2013 12:02 PM

Ian, a question. Where does the inbreeding percentage (15.33%) appear on the TrueNicks report? I am still learning how to read these things. Thanks. Tom

JerseyTom 05 Mar 2013 12:51 PM


At this point the coefficients of relatedness appear only in the Key Ancestors Report with Analysis. We hope to add this figure to other reports in the future because it's a good way to quantify the degree of inbreeding in a particular mating.

Ian Tapp 05 Mar 2013 12:58 PM

While I am not afraid to inbreed to this extent, I do question the decision.  There is a total of 6 crosses of Northern Dancer in the resulting foal....2 in the 4th, 2 in the 5th, 1 in the 6th and 1 in the 7th generation.  Five of the 6 crosses are thru sons.  Only Northern Sea represents a daughter, the dam of Southern Halo.  I would have sought out a more balanced inbreeding on the pedigree than what we see with this mating.  Hopefully Frankel is dominate enough to put a good foal on the ground.  

Robert 05 Mar 2013 1:20 PM

Scientifically, inbreeding (whether you are talking about sweet peas or racehorses) results in a a large number of homozygous genes. If there is any inherent genetic weakness (whether for fertility, soundness, temperment, etc) it is more likely to show up in the offspring. On the other hand the "good" genes are also more likely to be homozygous.

You lose hybrid vigor, which can often produce an unexpectedly good individual, but if you produce a good offspring with a high degree of homozygous "good" genes, they are likely to be prepotent in their offspring.

What this means is that you want to be sure not to duplicate faults if closely inbreeding. by inbreeding, you increase the chances that an unknown fault or mutation will show up.

On the other hand, a horse such as Frankel is inarguably such a superior individual, that inbreeding to TALENTED close relations probably has a higher chance of success than it would with other stallions. He may have known faults, every horse has something, but they are not obvious to the outside observer. If it works it could be sensational. When it does not work, the "bad" ones are likely to be poor both on the track and in the breeding shed.

There are period throughout the development of the Thoroughbred, when individual stallions become so dominant that close inbreeding to them enjoyed a period of great success.

Donut Jimmy 05 Mar 2013 2:22 PM

There are dozens of false statements in just about everybody's lecture, which includes each statement about a real or theoretical horse's inheritance.

While we can make true statements about the average genetic similarity in 10,000 horses bred at a certain remove, to say that any two of those horses have that similarity is a wild guess.

War Admiral was a Sweep. It is quite possible for one of the grandparents to be bred right out of a foal, and it is just as likely as any other specific degree of affinity.

Cassandra.Says 05 Mar 2013 3:05 PM


The wikipedia page explains it this way:

"The purpose of such a coefficient is to express the likelihood of effects due to inbreeding to be expected based on a known pedigree..."

In other words, it deals with the probability of inheritance. True, the coefficient is a guess as to actual inheritance, but it is a guess based on probability.

Ian Tapp 05 Mar 2013 3:19 PM


We'll be long gone by the time it's clear if Frankel is the St. Simon or Nearco of the 21st Century or less likely that Chrysanthemum will be the Weekend Surprise or Nogara of the same time.

jim of G 05 Mar 2013 5:09 PM


And who wrote that Wikipedia page?

We can do much better than chance with the data available nowadays, but it requires horsemanship, study of individuals and a long attention span.

It is probable that selecting our stallions by the same test, racing them, means selecting horses more genetically similar than chance to go to stud. Method fail right off the bat. We also have enough data to trace some traits with 100 per cent certainty if we put our minds to it. You can trace Prince Rose's stamina through a pedigree almost as surely as you can trace the gene for greying.

It is highly probable that any given horse has a few ancestors in its first five generations bred right out. The chances of a linebreeding effect are nil. When a horse is as prevalent as Northern Dancer or Mr. Prospector, we are reaching numbers where we can detect those inbreeding routes that do not show an inbreeding effect.

Without hard data, we can do better than chance by inbreeding through stallions with dissimilar phenotypes: different color, shape, best distance, temperament, way of going -- those things the Colonel knew about Fair Play and Sweep.

Cassandra.Says 05 Mar 2013 6:41 PM

Compare inbred and outcrossed Northern Dancers for soundness and stamina and you'll have your evidence right there. Weight your results for quality of the mare as a racemare or producer and the results are strengthened.

Inbreeding increases the chances of having two chromosomes of a pair exact duplicates, cloned from the same ancestral chromosome. Sexual selection is a back-up system. The cell needs one working gene and has two chances of getting at least one that is functional.

Damaged genes are frequent -- every chromosome has some. The saving grace is that with two chromosomes to work from, the damaged genes will not be the same ones on both chromosomes. With inbreeding the possibility exists that they will be the same ones and the job that gene does will not get done.

Race horses, ultimate athletes, can't spare many functions. We breed beef cattle for a manageable number of readily measured genes -- $$ of feed in and $$ of meat out. My example of something obscure but necessary to good racing performance used to be liver function. Mo certainly demonstrated that for us.

I think we can spare ourselves an onerous amount of research on this one. It is logically impossible that inbreeding does not impair function.

We can spare ourselves an onerous amount of math by using a rule of thumb that is true to the odds: 5x5 is 1, 4x5 is 2, 3x3 is 16, 4x4x5, bracket and add: 4x4 x 5 = 4, 4 x 4x5 = 6. The total gives you a refined guess, but that's all we have anyway.

Cassandra.Says 05 Mar 2013 7:06 PM


I thought you might knock wikipedia--don't worry, coefficient of relatedness is a widely accepted term in genetics.

We as an industry are learning more and more about the genetic and phenotypic profiles required for elite racehorse performance (just ask Byron to elaborate).

Add this to the increasing amount of data tools available to evaluate racing and breeding performance, TrueNicks included. None of this is meant to replace considerations of class, conformation, aptitude, temperament, etc., but rather to enhance the selection process.

Ian Tapp 05 Mar 2013 7:15 PM

Cassandra Says:

Well, at least you gave me an excuse to procrastinate work on my taxes...You speak with far too much conviction for one so obviously lacking in sufficient genetic education. For openers, here's an idea-read up on meiosis, with particular focus on "crossing-over". It may alter some of your dogmatic pronouncements, as re- your "...it is quite possible for one of the grandparents to be bred right out of a foal..." Really, "quite possible"?-I don't think so. Also, your notion that it is safer to inbreed through representatives of differing phenotypes is basic hogwash. Such a notion presumes that phenotype offers a fairly accurate reflection of genotype, not to mention the false presumption that the average breeder (or any breeder) can identify the MAJORITY of phenotypic traits. Fact is, it's quite otherwise. You, however, seem to embrace this misconception when you further state that "You can trace Prince Rose's stamina through a pedigree almost as surely as you can trace the gene for greying." I'm sure you prefer to believe this, but "genetics" says otherwise. And to further underline my point: While War Admiral may have -"SHELL-WISE"- more closely resembled Sweep than Man O'War or Fair Play doesn't at all suggest that he inherited a greater proportion of Sweep's other phenotypic traits or genes... The practice of close inbreeding in a selectively designed breed, such as the thoroughbred, should be avoided. In breeds such as this there may be, all else equal, an even greater liklihood that the pairing up of recessive genes (occurs with greater frequency when inbreeding) can result in negative consequences. Close inbreeding also lessens protein diversity which can make the animal more prone to illness, etc. (ex.-less antibody diversity. Please research several articles on the pros vs cons of inbreeding. All else equal, close inbreeding does the (indiviual) resultant offspring no favor.

jim of G:

More than likely, we'll know if my prediction was right or wrong during our lifetimes.

sceptre 05 Mar 2013 9:43 PM


The goals breeding racehorses are not comparable to the goals breeding sweet peas.

Race horses can be rendered useless by a single damaged gene. A single flaw may be a single fatal flaw, such as brittle hooves or shallow development of the socket in a ball and socket joint.

But an inbred mare can have an outcross foal. She may have chromosome clones but she only transmits one of the pair. If it's not top and bottom, it's not inbreeding. Hence, hybrid vigor. One outcross can cure any amount of inbreeding. (Provided we are lucky and avoid the dread occurrence you mention, genetic damage in the reproductive system.)

We are buggy whip makers watching horseless carriages come down the pike. In a very few years, instead of calculating averages (and usually falsely assuming a bell curve, where the average is more likely than the extremes) we'll just spend $1000 looking at the genes themselves. We'll be able to say "This horse carries Ribot's no. 4 chromosome pair."

Cassandra.Says 05 Mar 2013 11:05 PM


I often think you make sense and display knowledge, but not in this post.

I have worked in vivo with crossing over trying to pick up the Siamese Himalayan gene but otherwise breed back to Persian type. (It's easier with species that have litters.)

I used to teach a course entitled "English as a second language for speakers of Legalese" and by temperament and training I avoid jargon that obscures more than it reveals. Since evolution is about experimentation and change, any flat statement about genetics is not invariably true, including this one. I think technical discussions of genetics obfuscates because we try to cram all the possible exceptions and complications Mother Nature has come up with into every sentence we write. And God forbid we try to add epigenetics to it, when we know more about epigenetics.

The simple fact is that when two zygotes are produced, lets say sperm, in every ejaculation there will be sperm that contain everything from one to 31 (ignoring Xy) chromosomes that are derived from the stallion's sire plus the mirror image of that distribution. This is not on a bell curve, it's 31 coin flips. All distributions are possible and the average is not more likely than any other. (Little test.)

BTW, you may say that I am considering only the "shell" but you are seriously underestimating Colonel Bradley.

Cassandra.Says 05 Mar 2013 11:29 PM


My intent is not to insult you, or try to go "one-up". Rather, my retorts to your posts are driven by concern that some others may accept your misinformed pronouncements. To be frank, I was stunned to read that you taught a course in "English as a second language...", as your posts are among the most difficult to decipher. Let's take a look at your most recent butchering of the language-"The simple fact is that when two zygotes are produced, lets say sperm..." Well, for openers, sperm is not a zygote...Your comment about crossing over suggests you have no understanding about this simple concept-again, read up on meiosis. Also,  your paragraph which begins-"The simple fact is.." is, as written, simply wrong on all counts. Again, don't be stubborn, instead read about meiosis. And lastly, please don't give us that Col. Bradley nonsense. While some breeders are better than others at identifying what is OBSERVABLE, none were/are capable of accurately "reading" the genome, or the many phenotypic traits beyond non-"scientific" recognition. And while the Bradley breeding program does deserve much praise, keep in mind that my old friend, Olin Gentry, was largely responsible for much of its success.  

sceptre 06 Mar 2013 11:28 AM

Breed the best to the best and hope for the best.  It has been proven time and time again that close relations don't always equal the same results.  Full siblings seldom show the same on the track or in the breeding shed so all the research and data still comes down to luck when breeding.  But good breeding makes your odds a little better.  

I'm not a fan of close breeding (1st cousin is to close) but if the horses had traits that would compliment each other I guess one can try.

Hardlyhatful 06 Mar 2013 6:49 PM

I have a Storm Cat grandaughter and her 2 year old filly daughter.  Neither one has the right knee problem from Teralingua.  They both are stockily built like photos of Storm Cat, and have his reported temperment and speed.  If I breed either of them to a Storm Cat son/grandson what are my chanches of getting an offset right knee??  Thanks.  This may seem off the topic, but we are talking about the problems of inbreeding and I would love to hear your arguements.

sassy old man 07 Mar 2013 4:55 PM

We are all forgetting one thing: LUCK

Bad Girl 08 Mar 2013 3:51 PM

Riposte a half-sister to the dam of Frankel,Kind,broke her maiden to-day at Newmarket over 12 furlongs at the second time of asking.

John T 17 May 2013 10:04 PM

inbred horses bred to inbred horses produce great horses. Remember Domino.

janie boudreaux 20 Jul 2013 6:16 PM

Leave a Comment

All comments are moderated and must be approved before they are posted. The blog author reserves the right to edit or omit any comment.

  (Appears with your comment) (required)
  (Will not be published) (required)