Teen courts have grown rapidly in the United States despite little evidence of their effectiveness. A survival analysis of 635 teen court and 186 regular diversion participants showed no significant differences in recidivism, although program completers were half as likely to reoffend as noncompleters. Older offenders survived significantly better than younger ones, and girls better than boys. For the full sample, increasing the number of sanctions resulted in earlier reoffending. This effect disappeared when noncompleters were removed from the analysis, suggesting that increasing sanctions may lead certain teens to drop out and/or reoffend. Implications for policy include screening younger juveniles out of teen courts and reconsidering their panacea status. (Contains 3 tables and 1 figure.)