I would agree that it doesn't show very much, but it is interesting. TBH a lot of research involving dogs uses tiny sample sizes, which is 100% a problem, especially when it is used to support advice to dog owners or changes in the law.
A well-known trainer who is all over socials was a while ago hymning a paper that he claimed showed resource guarding could be vastly improved entirely force-free. The paper had a sample of 4 dogs, 2 of which were lost to follow-up. Of the papers I've read about training and e-collars, a significant proportion cite a study on radio fences (where the electric wire is buried and the dog wears a special collar) which looked at a total, er, 5 cases drawn from a population of unknown size (but likely to be in the thousands or tens of thousands).
Still, you and @SpanielsGalore did inspire me to go and have another look and I found I was able to read the whole of a paper of which I'd only previously been able to access the abstract. It has a sample of 57 dogs and mixed results on stride length:
https://www.tandfonline.com/doi/full/10.1080/10888705.2023.2259796#d1e880
TLDR It seems to depend on breed. But again, by the time you get down to the breed some of the samples were tiny (4 French bulldogs), so it's indicative for further research rather than anything else.
I will take me half a day to parse it properly, but I'll do it at some point.
Incidentally, if you look at mushing harnesses, they seem to fit closely at the base of the neck and to have the strap that goes under the dog well behind the elbow. I assume that this is to keep the shoulder assembly completely unimpeded, as mushers look for speed, and you're not going to maximise that if your huskies can't fully stretch their legs.
Yeah, I know I'm a sad nerd... I don't need telling!