Rudders and yaw are hard broke. Hop into the 109F4 (the only one I really did some testing in) and gradually input right rudder. You'll start seeing some visual stuttering, then at near full rudder deflection you'll see what seems to be an artificial stop on the aircraft yawing motion. On my computer, the plane starts bumping up against this stop.
In addition, full right rudder and stall = left spin in the 109. This can't be correct, no matter how much torque the engine has. Just because the plane stalls doesn't mean the rudder loses it's effectiveness... Methinks there's at least one bug in the code. Control effectiveness loss in stalls (in conventional aircraft) goes ailerons first, elevator second, rudder third, unless the rudder is completely blanked out by the wing in the stall. The only example of that rudder blanking that comes to mind is in the cessna 150 (high wing) when the flaps are dropped to full 40 deg, and maybe in some 172 models. But in every other plane I know of, some rudder effectiveness is kept through the stall, resulting in a spin in the direction of the yaw when a stall is encountered with a yaw angle present.
Add the knife-edge climbs and loops being reported, that adds up to either a completely broken flight model (unlikely) or a bug or two.
------------------
eagl <squealing Pigs> BYA
Oink Oink To War!!!