Oh well, there goes my theory:
Over a randomly selected long time period, I'm running 56% of what I should be. I guess there are different types of allin, technically I'd have to go in a calibrate by effective stack size of the allin.
Just for fun in dollars: