another terrible ATPL PoF question

[Arm out the window]

Morning all (or night as the case may be where you are)

AC, I agree that an object's motion can be broken down into rotation about the CofG and translation, and for a thrown object, that's fine.
For the helicopter, we have forces applied at different parts of the airframe working together to produce a motion.
A turn around the nose or tail isn't a turn around the CofG because we apply these forces in various amounts to achieve the aim and can pivot the helicopter around any point we like.
Nick and Crab seem to be saying that it's the relative moment arm around the CofG of the side forces from the main and tail rotors that's important in determining tail rotor roll.
There's a certain logic in this, but it doesn't tell the full story. Nick's diagram doesn't have weight in it, and he says that lift acts through the CofG.
I don't agree, and I think my diagram has all the appropriate forces in it, and explains tail rotor roll as the balance of two opposing force couples:
1. tail rotor thrust and main rotor side force acting to roll the fuselage left; and
2. lift and weight (lift through the hub and weight through the now laterally displaced CofG) acting to roll the fuselage right.
These are the determining factors, and the size of the left roll is affected by the relative vertical positions of the main and tail rotors, because the force couple has a larger moment arm - doesn't matter where the CofG is or whether these forces are acting through it or not, there's two opposing couples, and the body will roll until they're equal in magnitude.
The upshot of all this is that the relative heights of the main and tail rotors is important in the equation, which Nick in particular seems to be bagging as a theory. The 'vertical position of side forces with respect to the CofG' idea that he and Crab are touting sort of says the same thing, but it doesn't go far enough in consideration of the force couples.