ObjectivesTo investigate the longitudinal associations of sport participation with self-esteem in youth, and whether this relationship differs between team compared to individual, and non-esthetic compared to esthetic sports. DesignProspective data on sport participation at age 10 and self-esteem at age 13 years was used, collected from 2970 participants within Generation R, a population-based cohort study in the Netherlands. MethodsHours of sports per week were measured with a questionnaire, completed by the primary caregivers. Participation in sports was classified into team vs. individual and non-esthetic vs. esthetic sports. Self-esteem was assessed by the Harter's Self-Perception Profile. Linear regression analyses, adjusted for multiple testing were conducted to test the association between sports and self-esteem subdomains. Then, differences in self-esteem among youth participating in different types of sports were explored by two separate sets of linear regression with team and non-esthetic sports as the reference groups. ResultsMore hours of sports were associated with higher self-esteem (β = 0.081, pfdr < 0.001). This association was driven by the subdomains social acceptance (β = 0.057, pfdr = 0.004) and athletic competence (β = 0.242, pfdr < 0.001). Participation in individual or esthetic sports was associated with lower self-esteem compared to the team (β = −0.196, pfdr < 0.001) or non-esthetic (β = −0.172, pfdr = 0.002) sports. ConclusionsSport participation, particularly team and non-esthetic sports, may be beneficial for self-esteem among youth, driven by aspects like socializing and one's ability to do well at sports. The aspects of different types of sports could be considered for developing improved sport-related environments and effective physical activity interventions.