Originally Posted by Salubrious

Mark of Bikesmith lives in my town. I've met him and his wife, who is adamant that the shorter cranks made a huge difference for her.

I think that actual science is good too. I am an engineer for my day job. I just don't think this is good science in this study.

Here are the cranks used in that study

120 and 220 mm cranks
145 and 195mm cranks
120, 145, 170, 195, and 220mm cranks

I copied these right from the link itself (which is a series of Power Point panels). I did not even know that 220mm cranks existed, but we can see easily that there are big jumps in the crank lengths used. If we apply the French formula, (20-21% of inseam) its pretty evident that the gaps between sizes are so large that the peaks for most riders probably exist between the gaps. For example, the formula suggests that for me 167mm is ideal and I find I have good power with 165s and 170s. I notice that the 175s are not as easy to go long distances (IOW I can go faster and further with less fatigue if I have a 165 instead of a 175). I don't see a 160mm, 165mm (a common size) nor do I see 175mm. This is what I mean by the gaps are too large. For many riders its likely if they are less than 6' in height that the peak of performance for them exists between 155 and 170, so their peaks would simply not show up with this level of resolution.

In short, the gaps are too large in this study. I think it would have made more sense embarking on something like this if 5mm changes were made. Additionally, this study seems to have ignored the prior information (the French formula)- it could have sought to prove or disprove that at the same time.

Bottom line is it does not seem to present enough data to be useful.