In this paper I offer a new, probabilistic argument for the mind‐brain identity thesis, put forward by U.T. Place, H. Feigl, and J.J.C. Smart in the 1950s. After considering the epistemic, or conceivability based arguments against physicalism, I build an argument to the effect that naturalistic dualism ‐ the view that phenomenal properties do not metaphysically supervene on physical properties, but they are nomically connected – is probabilistically incoherent. The conclusion will be that phsyicalism, in the form of the identity thesis, is almost surely true.