Abstract Manley and Wasserman's PROP account of dispositions has been influential in the recent debate about the nature of dispositions. In this paper, I will bring under scrutiny one crucial step in Manley and Wasserman's reasoning leading to the PROP account. The step is one where they abandon what they call ‘MOST’, the idea that x is disposed to M when C iff x would M in most cases where C obtains. I will argue below, however, that this step is invalid.