All in all, @
Salubrious, I think we agree. My wife, who is much smaller than I, and does not have disproportionately long legs, rides a 650c bike with 170 mm crank arms. Her bike is tiny, really looks comical next to one of my bikes., but it has longer crank arms. I am reluctant to shorten them, because (a) she likes them, and (b) if I shorten them* then I would have to raise her saddle by 17 mm or so, which she does not want.
(*Shortening them is not a problem; I have a 175 mm crank that is otherwise the same as what's on her bike, and I have taps etc for shortening it. I could safely cut the 175 mm crank down to 153 mm or so (or shorter). I've done this several times.)
Let me rephrase something. I subscribe to the principle that the ideal crank length should be proportional to some bodily measurement. I am not convinced it is inseam; maybe it's femur length, I don' t know, I haven't studied this. The reason I reject the usual recommendations is that they are based on an assumption, or the narrow range of crank arms that is readily available, rather than a full range of theoretically possible crank arms.
By "ideal crank arm length" I mean the crank arm length that gives the best compromise between power, efficiency, comfort, wear and tear on the body, or any other factor that might seem relevant. Given that the choice is between 165's and 170's, I think it important to emphasize that there is not much difference between them. I am convinced the shorter of the two is better; but I am not convinced it is shorter enough.
Ten years ago I read an interesting study on the internet, in which the tester had had several different riders try out cranks ranging from 140 to 180 mm, if I recall correctly. I cannot find this article any more. The result was that all riders strongly disliked the shortest crank arms
at first but soon became accustomed to them and stopped complaining; but when they went back to longer crank arms, they strongly disliked the longer ones, and it took longer for them to stop complaining. As for power, speed, comfort, etc, there was not much difference among the crank arms tested. Logic would dictate that they repeated the experiment with shorter and even shorter cranks (as well as longer and longer ones), but i don't know if they ever did that.
Mike Burrows and Mark Stonich --both of whom have an interest in recumbent bikes-- are both proponents of shorter cranks, both for recumbent riders and for the rest of us.