This study investigated whether positive or negative affect has an independent association with exercise self-efficacy. Participants (N = 324) age 75-85 were classified as high or at-risk performers, and three exercise-self-efficacy items (scored 1-10) were assessed. For at-risk performers, positive affect was significantly associated with confidence in the ability to perform strength and flexibility (b = 0.83, SE = 0.23, p = .001) and aerobic exercise (b = 0.59, SE = 0.28, p = .04) and with the perception that exercise would not worsen preexisting symptoms (b = 0.73, SE = 0.24, p = .001). Among high performers, nonsignificant associations were found for positive and negative affect and exercise-self-efficacy. For at-risk performers, higher positive affect was associated with an increased odds ratio of 2.72 for scoring 10 on the muscle strength and flexibility item, 4.08 on the aerobic item, and 2.94 on the item assessing preexisting symptoms. The results suggest that improving at-risk older adults’ positive affect might increase their participation in exercise.