In a moment of enlightenment brought on my this thread I now realize I can cheat and cause my stoker to go slight OOP by hesitating for some fraction of a pedal stroke. Of course this won't be precise, but it's an easy enough experiment (on a daVinci). In our situation where I expect I'm providing 2/3+ of the power when climbing it would probably be advantageous to have the stoker behind to help smooth out the power at the bottom of my pedal stroke. Climbing standing would seem to favor the stoker somewhat ahead. Maybe we need DI3+ with variable stoker phase under computer control.