The effectiveness of outdoor smoking bans on smoking behavior among adolescents remains inconclusive. This study evaluates the long-term impact of outdoor school ground smoking bans among adolescents at secondary schools on the use of conventional cigarettes, e-cigarettes (with/without nicotine) and water pipes. Outdoor smoking bans at 19 Dutch secondary schools were evaluated using a quasi-experimental design. Data on 7733 adolescents were obtained at baseline, and at 6 and 18-month follow-up. The impact of outdoor smoking bans on ‘ever use of conventional cigarettes’, ‘smoking onset’, ‘ever use of e-cigarette with nicotine’, ‘e-cigarette without nicotine’, and ‘water pipe’ was measured. Multilevel logistic regression analysis was used. At schools with a ban, implementation fidelity was checked. At schools where a ban was implemented, at 18-month follow-up more adolescents had started smoking compared to the control condition. No effect of implementation of the ban was found for smoking prevalence, e-cigarettes with/without nicotine, and water pipe use. Implementation fidelity was sufficient. No long-term effects were found of an outdoor smoking ban, except for smoking onset. The ban might cause a reversal effect when schools encounter difficulties with its enforcement or when adolescents still see others smoking. Additional research is required with a longer follow-up than 18 months.