Failure of selective breeding?

Mugabe asserts at pumpkinperson:

off topic but a huge “teachable moment” today for hereditists.

for the first time since affirmed in ’78 a horse has won the tiple crown.

what horse still holds the record times in all three triple crown races?

secretariat in ’73.

the lesson is that despite enormous incentive to breed a super horse none have succeeded in besting secretariat. and that’s been ca 14 generations of thoroughbreds since ’73.

While Mugabe's broader point is without merit (there's no shortage of examples of successful applications of selective breeding to animals), it does look like thoroughbred breeders may be doing it wrong. Steve Sailer noted in 2005:

Enormous amounts of money are spent to acquire the best breeding stock for thoroughbred horse racing. The average time of the winning horse in the Kentucky Derby dropped steadily up through middle of the 20th Century (as shown in the red line above), yet there has been virtually no overall improvement in average time since 1950 (as shown in the blue line). If the improvements seen from 1896-1949 had continued, the average winning time today would be about three seconds faster.

Expert ratings seem to point to a similar conclusion:

The trend line (red curve) now shows a clear dip between about 1940 and 1970. This means that a larger number of higher ranked horses are clustered in that time frame. The upward trend since then reinforces the position of the critics in that the horses on the list selected since mid-century are, as a group, not as highly regarded as those that came immediately before them. The overall pattern seems to have been continual improvement through the first half of the 20th century followed by a gradual decline since then. This study is far from definitive because the rankings are objective only to the extent that humans apply their best judgment. On the other hand, since there is no purely objective standard to use, the opinions of the seven experts are as a good as any and better than most. The result of their judgment is that changes have occurred and that the best of recent years are not quite up to the level of the best that preceded them.

In contrast, standardbred harness racing times have continued to decline:

From the above chart, starting in 1930 standardbreds have improved winning Hambo times by 7.9%, while thoroughbreds of the same period have improved Derby winning times by just 1.8%. If you start analyzing the data at the time of the breaking of the magic 2:00 threshold the differences are even more striking – 7.0% for trotters vs. an imperceptible 0.08% for thoroughbreds.

The horse trainer quoted above thinks this difference mostly comes down to thoroughbred trainers not having "a very good grasp of equine exercise physiology". But assuming it's a matter of breeding, here's one major difference:

in order for the resulting foal to be registered with The Jockey Club, Thoroughbred breeders must have their mares bred by live cover. As inconvenient as this might seem to those unfamiliar with the Thoroughbred industry, this policy ensures the ability of more colts to end up at stud. If Thoroughbred mare owners weren't restricted by geography and location, they would no doubt be interested only in the sperm of the best of the best, thereby devaluing the stud services of all other stallions. In fact this is the case in the Standardbred industry. Because artificial insemination is the norm, the breed has improved much faster in comparison to the Thoroughbred. However, the Standardbred colt owner must make a profit at the racetrack because unless the colt is literally the best of the best there is no chance of making a profit on the colt as a stud (1).

There's also the question of what's actually being selected for.

"They're the ones who created this tragedy," Parker said. "Robert Clay is smart enough to know better. He bred her. That's where it starts. You don't blame the trainer, who does not have the reputation of breaking horses down, and you don't blame the poor little jockey. ... She was inbred three times to Raise a Native! [She broke down] right where Raise a Native was the weakest, right in the ankles, and everybody acts like they don't know what caused this filly to break down. It's written right there for everyone to see! Except they refuse to see it. To admit it is to address the fact that all these stallions that are bred like that, that all the yearlings that are bred like that, are potential accidents waiting to happen. And they've got so much money wrapped up in this crap!" [. . .]

Through the first 60 years of the 20th century, most of the major stallions and many of the best mares were owned and controlled by some of the oldest families and richest sporting patrons in America, by the Whitneys and Woodwards, the Bradleys and Wideners, the Klebergs and Mellons. They bred horses to race them, not to sell them, and they did so in order to compete against one other -- to beat their fellow members of The Jockey Club, to see who had the fastest horse. A cardinal article of their faith was to "improve the breed," which meant to breed a horse with great speed, stamina and soundness. In fact, on the C.V. Whitney farm in Lexington, a foal born with a crooked leg was usually taken into the woods and shot, lest he or she pollute the Whitney bloodlines with this inherent deformity.

By the middle of the last century, this tight-knit racing world began to change. As these families died out and their blue-chip breeding stock was sold at dispersal auctions, the best stallions and mares fell into the hands of commercial breeders, whose central motivation was to breed, not so much a sound or durable horse, but rather an attractive horse, a "cosmetic horse," who showed well, who had a pedigree filled with fashionable names, preferably sire lines that glowed with speed, and who thus would draw the biggest price at the fanciest yearlings sales. Because they needed to look like show horses, these hothouse yearlings were often raised in small pens and not allowed to run free, or to kick, bite and roughhouse with their peers.

[. . .] This gradual softening and weakening of the breed has led to the use of more medications to keep these horses running sound, among them the corticosteroids injected into injured knees and ankles. [. . .] Drugs are only one way that the industry has been trying to make up for the weakening of the American thoroughbred. A number of racetracks have already replaced their dirt tracks with softer Polytrack surfaces, for the purpose of reducing breakdowns, but all we know about these tracks is that they often are the bane of true speed horses, favoring come-from-behind plodders. They have made the outcome of races so unpredictable that they have driven the high-rolling, sophisticated gamblers away from the betting parlors; and they may or may not save horses' lives. The jury remains sequestered.

All such expedients are aimed at forgiving commercial breeders for what they have done to the breed. At the core of the problem is the fact that the fastest and most popular sire lines in the world are the least durable and sound.

[Eight Belles' breakdown: a predictable tragedy]

This decrease in starts per horse over time also sheds light on another contemptible fact directly related to the absurdity of the economic covetousness of the industry. With exaggerated breeding fees and bloodstock sales that generate literally millions of dollars it has forced breeders to resort to breeding something fashionable that people will be interested in buying. Unfortunately the fastest sires are usually the most unsound. In the commercial sense of the word, horses are no longer bred to race but rather are bred to sell.

[HORSE RACING — BREEDING FOR TROUBLE]