I would like someone to demonstrate, or link to such demonstration, that a stiffer frame is faster -- like for racing. I'm talking about stiffness resisting pedaling deflection (BB shell, stays, head/seat tube orientation). My current belief is that it's a myth that won't die. I don't dislike stiff frames, but I don't see why there would be any advantage whatsoever -- maybe minor tradeoffs, but no advantage.

We'll assume a flexy frame is at least stiff enough to prevent chain rub on the front der. I can appreciate a preference for the stiff feel -- which could increase an individual's comfort enough to ride faster, but that's the rider, not the frame.

I'm looking for engineering and physics here. I'd like to be educated if I'm wrong.

Arguments like "that's what the pros ride" and "the bestest bike maker EVAR makes stiff frames" will not be considered.