Horse speed has not improve over the past 50 years but humans did drastically. Why?

Boring.

Human tracks, spikes, and shoes are significantly better. Horse tracks haven't changed much.

Can’t find it anymore but I think Epstein had a similar hypothesis and he had cyclists doing the same test as the ones in 1974. I believe it was the hour record, and if cyclists of today used bicycles of 1974 the times were actually similar.

The reason is the horses were doped to the gills in the past but now they are having a more difficult time as drug testing has evolved, and when Mr. Ed died, they lost their most valuable spokesperson or advocate in terms of getting them off legally.

What does it say for humans then if this hypothesis is correct?

Horses have been inbred for a long time and in such a way as to really limit what kind of training they can do from their incredibly fragile hooves and ankles (or whatever you call ankles in a horse). There is far more experimentation in human training than horse training, and you can get the buy in of the humans to the training, not really the horses to the horse training. The central governor can be unlocked to do what was not believed possible, as we see every time another barrier is broken and there are sudden time drops. Competition brings better times. By the way, some horse tracks are faster than in the past.

Carbon fiber horseshoes with rockered soles and converting Churchill Downs to Mondo rubber will revolutionize the sport!

They haven’t figured out a way to make super horse shoes.

Horse speed progression is similar to that of the human long jump.

If we shorten the timeframe to 24 years (1999)the human WR in the half marathon and marathon have improved significantly. They have improved marginally in the 800, not at all in the 1500, mile and 3000, and modestly in the 5000 and 10000.

Super shoes and elite 5k/10k runners dedicating themselves to the marathon explain most of drop in marathon times. Fueling, improved nutrition, and better recovery in training likely account for a smaller fraction of the improvement for those road distances.

The cynical take on the stagnation at the 1500, mile, and 3000 post 1999 is that those events are true Vo2 max events that unchecked EPO doping benefited the most. Once better EPO tests in 2005 and the ABP in 2009 restricted blood doping to microdosing; 3:28/7:24 became the performance limit.

I would bet even money that the outdoor 1500, mile, and 3000 records last to 2030 as long as shoes don't get any better than the Dragonfly.

Horse owners breed for speed with new generations every few years. If genetics were that important, horse times should be going way up.

But remember 50 years ago was Secretariat, an outlier who was way beyond the field like Bob Beamon's long jump. For this to be relevant, we would have to look at improvement rates of horses who are not Secretariat.

Go right to the source and ask the horse
He'll give you the answer that you'll endorse.

The430miler did a good job of documenting this. When Secretariat died, scientists discovered that his heart, lungs, brain, internal organs, and passion for running were MASSIVE compared to all other horses. Long story short, no one has been as good

It says absolutely nothing. I hope you can sleep tonight knowing this.

Secretariat was nowhere near Bob Beamon caliber. Those tracks were scraped all over the place in 1973. That's what is missing from the OP's summary. Track conditions play a massive role in horse racing final times. That's why handicappers use a track variant and adjust it to every day at every track. Sometimes there are adjustments within the same racing card if conditions change drastically.

It is unbelievable how fast the tracks were in 1973. When I began studying it nearly a decade later I was in disbelief. Seven of the 11 existing main track Belmont track records had been set within a few weeks either way of Secretariat's Belmont. Then that fall when he set the world record at 9 furlongs in the Marlboro Cup, a filly named Desert Vixen had equalled the track record at the same distance in the prior race. Now, Desert Vixen was good but there's no chance she was the equal of any horse who had ever run a mile and an eighth at storied Belmont Park. The track was a freeway. And likewise Churchill Downs and Pimlico were lightning fast. The sport was desperate for a Triple Crown legend since there had been a 25 year gap since Citation.

I realize this is an argument I can't win. Every simpleton fan doesn't know a damn thing about horse racing other than they somehow think Secretariat was open lengths better than any horse in history. That's how you have laughable "Dream Race" projections in which the field includes all the greats in history and yet they collapse one by one in midstretch as Secretariat rolls past. Yeah, they all turn into nothing more than Sham, he of the awesome total of one Grade One victory yet the fritoheads turn him into a world beater based on heart size and other laughable variables.

I wish those dream races could actually be run. Let's put it that way. Fans would be stunned. I appreciate that even a huge Secretariat fan like Andrew Beyer eventually conceded that the public had a severely inflated perspective toward how Secretariat would fare against other all time greats.

EPO doesn't work on horses.

This is interesting. I'd never thought about this. One thing that comes to mind is that as far as I know no one is interested in how fast horses cover their race distances as far as I can tell. Could that be a factor? And are horses faster now than they were more than fifty years ago? I do think this shows that a lot of improvement in distance records is due to faster tracks, faster shoes, professional rabbits, a focus on times and records often equal to the focus on place,and PEDs.

Secretariat's track record at the Kentucky Derby is from 1973. Only once, in 2001, has a horse run under 2 minutes since then. Pretty wild. Every human running record is significantly faster today than in 1973. Although it is certainly notable that the mile WR is now 24 years old. From Roger Bannister until 1999 it was broken, on average, every two years.

Cleary you know a lot more than I do about this. But what about Secretariat's margin of victory in the Belmont? Isn't that far and away the largest margin of victory of any triple crown race?

Also, looking up the Kentucky Derby and Preakness, Secretariat was 10.5 lengths ahead of 3rd place in both races (and 2.5 lengths ahead of Sham). This would be a record in the former and a near-record in the latter. So doesn't this support the conclusion that Sham really was way better than the rest of the field, and that Secretariat was far greater still?