This paper offers a new perspective on Boltzmann’s replies to the Loschmidt reversibility objection that draws on Boltzmann’s notion of molecular disorder, as presented in his Lectures on Gas Theory. Rather than adopting the assumption of molecular chaos or Stosszahlansatz, I offer the hypothesis of disorder, which takes into account changes in molecular-disordering upon dynamical evolution, and is suggested as a precondition for the application of the laws of probability to gas systems. This hypothesis can hold asymmetrically in time, unlike the notion of molecular disorder that is a claim about the instantaneous state of a system. This proposal is further investigated by simulations of the Kac ring model.