Before designing epidemiologic, genetic, or treatment trials in pediatric stroke, we should learn from adult trials which preceded. Adult trialists state that there is a need for improved animal models to mimic human disease. Dose-response curves with blinded outcome measures would improve preclinical data. Functional and histologic outcome measures would improve the animal model. In regard to human Phase 2 medication trials, safety, delivery, end points, and surrogate markers are necessary. The detection of biologic activity can be defined in Phase 2B trials. For Phase 3 trials, the experiment needs to be simple with global outcome measures as end points. New statistical designs such as futility analysis may improve and streamline trials in both adults and children. Clinical trials are a long process, and care needs to be incorporated at every step. Why is there a propensity of failed studies for acute ischemic stroke in adults? The reasons include (a) the animal model fails us, (b) the Phase 2 trials are ineffective in defining dose, and (c) the Phase 3 trials are poorly done. Because clinical trials, more often than not, fail to give positive (effective treatment) results, it seems reasonable to attempt to learn from others' past experiences before initiating pediatric stroke trials.