Originally Posted by
RChung
Trakhak was saying that the only way for energy stored in the frame to come back to drive the bike forward was through the crank, but you're saying that there's an independent path through the chainstays. So if we took the chain (and/or cranks/BB) off a bike, mounted it in a trainer, then pressed sideways on the BB shell, first on the right side and then on the left, will the rear wheel rotate forward? Or do you need the chain and chainring?
You need a chain, for the same reason a bow needs a string. Which isn't a terrible analogy, since an archer compresses the bow via the arrow and string, and then the bow feeds back through the string to launch the arrow.
Take the motion of the chain and bike out of the picture and you could look at the system as a lever arm that pivots at the hub axle, with one end of the lever bolted to a pivot on the pavement, the other end of that lever at the top of the cassette. The chain is a connecting rod running from the top of the cassette to lever the height of the chainwheel. The pivot of the chainwheel lever arm and the hub axle are connected by a compression spring (the stays). Press the crank lever and the compression spring at the stay compresses so the crank moves back. Relax the pressure on the crank and spring decompresses, pushing the crank pivot forward due to the way the chain rigidly connects them.
If anyone thinks this sort of thing is hard to picture, learn how a helicopter rotorhead works. It has a much more complicated 3D version of this kind of push and pull.