The stochastic method of ground‐motion simulation assumes that the energy in a target spectrum is spread over a duration D_T. D_T is generally decomposed into the duration due to source effects (D_S) and to path effects (D_P). For the most commonly used source, seismological theory directly relates D_S to the source corner frequency, accounting for the magnitude scaling of D_T. In contrast, D_P is related to propagation effects that are more difficult to represent by analytic equations based on the physics of the process. We are primarily motivated to revisit D_T because the function currently employed by many implementations of the stochastic method for active tectonic regions underpredicts observed durations, leading to an overprediction of ground motions for a given target spectrum. Further, there is some inconsistency in the literature regarding which empirical duration corresponds to D_T. Thus, we begin by clarifying the relationship between empirical durations and D_T as used in the first author’s implementation of the stochastic method, and then we develop a new D_P relationship. The new D_P function gives significantly longer durations than in the previous D_P function, but the relative contribution of D_P to D_T still diminishes with increasing magnitude. Thus, this correction is more important for small events or subfaults of larger events modeled with the stochastic finite‐fault method.