Social factors are important determinants of youth physical activity (PA), but the longitudinal association between parental behaviours and adolescent PA has not been clearly assessed. This prospective study examined average and lagged associations between perceived parental support and control with adolescents' moderate to vigorous PA (MVPA); and assessed the independent associations between specific parental support and control behaviours and adolescents' MVPA. Data from three cycles of the MATCH study, when 374 participants were 12, 16 and 17 years old, were included in this analysis. At each cycle, participants self-reported questionnaires on perceived tangible parental support, intangible support, and control behaviours as well as number of days per week attaining at least 60 min of MVPA. Mixed effect models were used to assess the longitudinal relationship between parental behaviours and MVPA. Cross-lagged panel design was used to assess the association of parental behaviours during early adolescence with MVPA during late adolescence. Overall parental support (coef. = 0.46, P < 0.0001), tangible support (coef. = 0.37, P < 0.0001), encouragement (coef. = 0.12, P = 0.025) and transportation (coef. = 0.25, P < 0.0001) were positively associated with MVPA, whereas parental control was a negative predictor of MVPA (coef. = - 0.18, P = 0.003). Perceived parental behaviours appeared to have long term associations (5 y.) with MVPA as parent support (coef. = 0.40, P = 0.006) and co-participation (coef. = 0.33, P = 0.017) reported around age 12 were positively associated with MVPA measured 5 years later. Parental support for PA, particularly in the form of tangible support, may be a key factor to include in interventions aiming to promote PA during adolescence. In contrast, parents should be encouraged to avoid control behaviours as these appear to lead to lower MVPA among adolescents.