With dogs, it’s the opposite. Larger breeds die younger, on average. Of course, they may have been bred up larger than their infrastructure can handle, long term.
This is always a contentious subject, but let’s take the OP’s question to the limit.
If a kennel operator were to state categorically, “Greyhounds are faster than Shar Peis,” nobody would bat an eyelash or scream that the operator is a “breedist.” But if a track coach were to suggest that, say, persons of West African descent are faster sprinters than Irishmen or Poles, he’d be ripped a new orifice, and forced to apologize for his offensive, racist statement.
In the same way, if a kennel operator said that, say, border collies are smarter than Basset hounds, no one would care. But if a researcher suggested that the Chinese or Koreans are smarter than Albanians and Basques, people would be quick to scream that he’s a racist or that “race” is a meaningless term.
Why the difference?
Because different breeds of dogs are deliberately bread for certain traits and people aren’t. The dog breeder would be talking about scientific data, and the track coach would be talking anecdotally.
Dangerous. An old anecdote has a famous actress telling George Bernard Shaw that they should have a child, because of how amazing a child would be with her looks and his brain. He refused, asking her to think of a child with his looks and her brains. 
Guide Dogs, at least the ones I know, have been selectively bred for about 60 years, with very little external genetic material being added. Breeders are carefully selected based on a very thorough examination. The dogs they breed with are carefully selected also. A remarkably large percentage of guide dog puppies have wonderful dispositions, and there are indeed health problem which get screened out.
GDB has stored breeding records for 50 years at least. They also have samples that can be used for DNA analysis. They have a project with UC Davis on this. As soon as they can sequence dog DNA, they are going to start looking for the sites of various characteristics. It should be interesting.
I’m not sure that the kennel operator’s observations would be any less anecdotal than the track coach’s. And what about the researcher who suggests “that the Chinese or Koreans are smarter than Albanians and Basques?” Seems like whenever there is an attempt to introduce scientific data (test scores and what have you), that’s when people really start screaming racist bloody murder.
ETA: “Because different breeds of dogs are deliberately bread for certain traits and people aren’t.”–I remember Jimmy the Greek got into some hot water on just this question as well.
Well, the kennel operator would presumably be speaking as an authority on dog breeding, which operates under pretty strict control. We don’t breed humans, and it’s easy to jump to erroneous conclusions that simply aren’t backed by enough facts when talking about race.
I’m sure there are many populations around the world with obvious physical differences that would give those populations an advantage/disadvantage in certain sports. For instance, you’re not going to see a lot of basketball players who trace their ancestry to the Mbuti pygmies of Africa. As to whether West Africans are genetically predisposed to be faster than other populations, that is an open question. There is some evidence for this belief, but it’s hardly an established scientific fact. And whether it’s a racist view or not, I think that depends on the intent of the person saying it. It’s certainly a legitimate question, and there are respected scientists who try underrstand such issues without being racists.
It could work, if you cloned them.
Just make sure you don’t clone the clones. Copies of copies, ya know. They never turn out quite right.
Because the situations are not even remotely comparable.
As has already been said, dog breeds are the result of extremely extensive selection for single traits over many generations. No human populations have been subject to anything like this. Human populations have been interbreeding with their neighbors for thousands of years; dog breeds are prevented from crossing with any other kinds of dogs. Dog breeds are much less numerous in number of individuals than human populations. All this results in a far, far lower range of physical variation in a dog breed than there is in any human population.
The real problem with selective breeding of humans is that any such breeding program would have to last many generations. And once the people who decided to enact the breeding program are dead, the desire to continue the breeding program will probably die with them.
And a major problem I had with the idea of “Methusalah’s Children” was the idea that the secret Howard Foundation, which was all controlled by Howard families, would continue to be used to reward the selective breeding of long-lived people. Pretty soon the secret money would be controlled by someone who was more interested in the secret money aspect and not so interested in the selective breeding aspect.
Real trustees in real trusts don’t often steal the money in the trust, but that’s because the trust isn’t secret, and they face a certain risk of going to jail if they steal the money. Why would a Howard trustee who happened to be not quite as long-lived as others altruistically give everyone else the money? Why wouldn’t they keep as much as they could for themselves? Who’s going to call the cops?
One word for you: Slavery.
Referring to a speculative piece of science fiction is one thing, but why not take a look at what’s actually happened in human history?
In fact, for most of human history, generations of people have been pressured to fill certain narrow niches of work against their will, by generations of people who had the power to influence or coerce their choice of mate and reproductive behavior.
I’d guess that if slaveowners believed (or observed?) that matching up certain types of slaves would yield economic benefit, such an incentive would guarantee that they’d stick with the program.
I’m not talking specifically about the antebellum south, nor even about “slavery” strictly defined, if we also look at some of the harsher social hierarchies and caste systems.
It’s one of those things it’s unpleasant to think on, and a question that might foster accusations of racism just for its asking. But have humans actually been subject to breeding programs in history? (Either deliberate or happenstance?) And if so, did they work?
Slavery with directed breeding to the extent it was ever practiced in the American south was not really replicated (so far as I know) in any other known historical “slave” populations.