Smokers' responses to the Smoking Consequences Questionnaire-Adult (A. L. Copeland, T. H. Brandon, & E. P. Quinn, 1995) were used to determine whether smoking outcome expectancies are moderated by subjective evaluations (desirability ratings) in their effect on smoking-related variables. Hierarchical regression in a reanalysis of data from A. L. Copeland et al. indicated that the product of probability and desirability ratings accounted for a significant amount of variance in smoking rate, nicotine dependence, saliva cotinine, and posttreatment smoking status above that accounted for by probability and desirability ratings alone. Results indicate that desirability ratings serve as moderators to probability ratings in explaining current smoking, nicotine dependence, and continued smoking. Effects were modest in magnitude but suggest that it may be important to address subjective evaluations of outcomes in smoking cessation and relapse prevention efforts.