MazelMan, check the bottom of my earlier post. It includes the original km curve. *If* you assume that the only tampering was a simple switch of the two arms, then: Placebo outperformed the 3mg arm (though not by as much as it outperformed the "1mg arm").