Background: Abundant evidence demonstrates a relationship between physical activity and mental wellbeing. However, the strength of the relationship is not consistent. Factors contributing to variation in the strength of association are not well understood and, therefore, it remains difficult to optimize physical activity to ensure the strongest possible relationship with mental health. Self-determination theory suggests that more autonomously motivated behaviors lead to better mental health outcomes, when compared to more controlled behaviors. Therefore, we examined whether autonomous and controlled motivation moderated the relationships between physical activity and affective wellbeing within two domains (i.e., leisure-time and active travel). Methods: Between February and April 2014, adolescents (N = 1632, M age = 12.94 years, SD = 0.54, 55% male) wore an accelerometer across seven-days and completed self-report measures of leisure-time physical activity and active travel. They also completed two measures of motivation (towards leisure-time physical activity and active travel) and an affective wellbeing measure. Results: Structural equation modeling revealed that greater self-reported leisure-time physical activity was associated with greater positive affect (β = .29) and less negative affect (β = −.19) and that motivation did not moderate these relationships. Self-reported active travel had no linear relationship with affective wellbeing, and motivation did not moderate these relationships. Accelerometer-measured leisure-time physical activity had no relationship with positive affect but, had a weak inverse association with negative affect (β = −.09), and neither relationship was moderated by motivation. Accelerometer-measured active travel had no association with positive affect; however, autonomous motivation significantly moderated this association such that active travel had a positive association with positive affect when autonomous motivation was high (β = .09), but a negative association when autonomous motivation was low (β = −.07). Accelerometer-measured active travel had no association with negative affect. Despite some significant moderation effects, motivation did not consistently moderate the relationship between all physical activity variables (leisure-time and active travel, and self-report and accelerometer) and affective outcomes.