Just as another thought to throw into the mix. If you look at the dynamics of helicopter rotors (complicated), perhaps there are some clues for what is needed for stunt props. One of the things that occurs to me is that for a control input (analogus to turning the stunt plane) there is a 90 degree phase lag in the control reaction of the rotor, if the rotor is freely hinged at the hub. This phase lag becomes smaller as the hub stiffness is increased, to about 65 degrees (as I remember) for a "rigid" hub.
As we go to larger diameter and more or less flexible props, wouldn't this same set of dynamics occur, to a different degree. I guess that this is the source of some of what is attributed to P effect, which is normally thought of as just a gyroscopic precession. Perhaps I am calling the same thing by two different names here, not sure, but I think the helo guys have shown it to be different than pure gyro precession. It means an opposite hand roll and yaw depending on which way the nose rotates, but how much?
John