Theoretical and empirical studies of consumer scheduling behavior usually ignore that consumers have more flexibility to adjust their schedule in the long run than in the short run. We are able to distinguish between long-run choices of travel routines and short-run choices of departure times due to an extensive panel dataset of commuters who participate in a real-life peak avoidance experiment. We find that the participants, who obtain monetary rewards for not traveling along a camera-observed highway link during morning peak hours, value travel time higher in the long-run model, as changes in travel time are more permanent and can be exploited better through the adjustment of routines. Schedule delays are valued higher in the short-run model, as scheduling restrictions are typically more binding in the short run.