A number of empirical studies have tested the spatial mismatch hypothesis by examining the commuting times of blacks and whites. This note points out that the link between spatial mismatch and commuting times may be weak when employment probabilities decline as the distance from job site to residence increases. A simple spatial model of urban employment is developed in which a fixed number of agents live in the central city. Two examples are presented in which increased spatial mismatch may either increase or decrease the average commuting time of central city minorities, depending on the rate at which employment probabilities decline with distance.