The certification Vmca figure is required and determined to give the pilot a fighting chance of maintaining control in the initial management of a critical engine failure - typically during the initial takeoff phase of flight. As we all know - hopefully (?) - it has nothing to do with whether the aircraft can climb or not. If the V2/Vtoss were to be unduly compromised by Vmca considerations, then the probability of a safe outcome is minimal or zero.
Certification Vmca has no direct relevance to continued OEI flight at altitude.
Now, putting the certification Vmca (i.e. the number quoted in the flight manual) to one side .... if one were to do the Vmca testing with a variety of bank angles (while maintaining a straight flight path), then the observation would be that the Vmca dataset so determined (what I refer to as "realworld" Vmca .. as opposed to the very specific certification value) was strongly dependent on (i.e. varies significantly with variations in) bank angle. As bank is increased into the operating engine, Vmca decreases ... and as bank is increased away from the operating engine, Vmca increases. This is to do with the effects of sideslip and directional stability on yawing moments and, consequentially, the rudder deflection required to make up the difference if one is to achieve a straight flight path.
There is also the high hazard consideration of what might happen to control with high sideslip angles ..... this is a related but separate issue and dictates that, at lower speeds, one doesn't want big bank angles ... regardless of the direction of bank.
If the pilot intends to manoeuvre at lowish speeds, then the effect of bank on the realworld Vmca becomes extremely pertinent to the discussion if the aircraft is permitted to operate in a manner where there is any significant sideslip .... Vmca departure, especially in dynamic circumstances, can be a real "gotcha".
If the pilot is manoeuvring at speeds sufficiently high to remain well above the region where Vmca variation is going to affect the operation, and the bank angles flown (considering airspeed) are sufficiently modest to keep the sideslip reasonably controlled - (i.e. something not too far removed from co-ordinated flight), then I don't see that it really matters which way we choose to bank (turn).
As in most things to do with flying, there are a lot of factors at play and most times explanations of why this or that happens or this or that technique is the better .... necessarily presume simplifying assumptions which are not disclosed. This consideration, I suggest, is the genesis of many of the "old wives' tales" represented as rock solid truths in circumstances sufficiently removed from the conditions of the original tale such that the underlying assumptions inherent in the original may no longer be relevant or applicable ........
Rant over ..... have a nice day .....
