The whole 'planing' thing is the notion that some bikes store AND release energy in a beneficial way, and some might not.

But the basic idea is that the pedal stroke has higher and lower output. On the downstroke the rider can exceed the ability of the frame's stiffness to transmit all the force to the road. When that happens the BB deflects - but that isn't happening in a vacuum. The BB deflects because the chainstays (among other things) are twisting to allow the rear wheel to be drawn closer to the crank by the load on the chain pulling them together. At the bottom of the pedal stroke the power drops sufficiently for the chainstays to unwind, and the rear wheel pushes to the rear against the mass of the rider - who is still loading the crank. If the rear wheel is pushing to the rear it is forcing the rider and the rest of the bike forward.


What always surprises me in this kind of discussion is that people can see that the bike can flex from exactly the kind of force that drives the bike, but think that force can suddenly be channeled into some sort of BB flex battery that briefly stores that energy and then somehow throws it away. But this is an equal and opposite situation - so that stored flex energy goes back into the mechanical path that created it, and not into some other realm.

Just like how a pole vaulter stores energy in the pole by compressing it and getting it back by allowing the pole to unwind. There is no other place for the pole's stored energy to go - the vaulter never stopped loading the pole, they just decreased peak force. Just like the bottom of the pedal stroke, the vaulter's momentum is still a load, so the pole can only unwind in the opposite way it was compressed, so it gives it back to the vaulter.

Same thing is true of how a rubber ball bounces, or how a half full milk jug acts when you slide it across the counter top. There is only one way to use up energy stored in a system that is mechanically closed - back out the way it came in.

Of course, if the frame isn't made of a material that makes good springs - steel, ti, carbon and (to a lesser extent) aluminum - you aren't going to store and release energy very efficiently. Good spring material has a very high energy transfer efficiency, which is why pogo sticks and the like aren't incredibly tiring to use. They convert very little stored energy to things like heat or sound. A frame made of plastic might not flex more than a steel frame, but you would notice it feel differently because it would use up much more of that peak pedaling compression as heat and the spring rate of the plastic would be low enough to not feed back into the drivetrain at the right time.


The problem with an ultrastiff frame is that it can't store any energy because it won't flex. So the peak torque while climbing/sprinting out of the saddle has to either go to the road or becomes greater than the system can absorb and feeds back to the rider's legs - opposing their output, making it choppy and less smooth. My Cannondale felt great climbing, but I could scrub the tires more easily than other bikes because peak torque would go right to the road and exceed the traction of the rubber - which is definitely a loss of energy. My Merlin ti deflects under that peak load, distributing forces over a longer period of time, making the chance of breaking traction much lower. So it uses the spring of its construction to accomplish the same thing as an oval chainring - distribute pedaling force better across the pedal stroke. Which is the same concept as those silly spring cranks, just more efficient because there are no moving parts in the spring path.