Glasgow
As an owner and pilot of my own GA aircraft I think that one aspect has not been examined in this thread. Whilst it is relatively straightforward to identify an engine failure in a piston twin in simulated situations (e.g. an examiner closing the throttle on an engine) it is much more difficult to identify in a real life emergency. The props' constant speed units will disguise a failing engine, and the surging from the failing engine will send the 'rudder ball' rolling fully from side to side with the plane yawing wildly. The pilot would have been left with very little direct information as to which engine had failed. The aircraft's performance figures for single engine operation, at max weight with failed engine feathered would have indicated a return to field. Other reports of the right hand engine's bang may have been all there was on which to make a decision.
The AAIB never shrink from apportioning blame. If the AAIB do not blame the pilot then that is good enough for me.