Frameless -- That's a good question, and a good point you make. I don't think that this has been systematically investigated yet. If you have any speculations or hard data regarding what the answer is, please post it up. That said, I think that the question you're asking is more applicable to acute (short-term) studies, which are good for generating hypotheses. Long-term studies measuring the macro-outcomes (performance & body composition) would be the ultimate test of such hypotheses.