The issue of whether video games with aggressive or violent content (henceforth aggressive video games) contribute to aggressive behavior in youth remains an issue of significant debate. One issue that has been raised is that some studies may inadvertently inflate effect sizes by use of questionable researcher practices and unstandardized assessments of predictors and outcomes, or lack of proper theory-driven controls. In the current article, a large sample of 3034 youth (72.8% male Mage = 11.2) in Singapore were assessed for links between aggressive game play and seven aggression or prosocial outcomes 2 years later. Theoretically relevant controls for prior aggression, poor impulse control, gender and family involvement were used. Effect sizes were compared to six nonsense outcomes specifically chosen to be theoretically unrelated to aggressive game play. The use of nonsense outcomes allows for a comparison of effect sizes between theoretically relevant and irrelevant outcomes, to help assess whether any statistically significant outcomes may be spurious in large datasets. Preregistration was employed to reduce questionable researcher practices. Results indicate that aggressive video games were unrelated to any of the outcomes using the study criteria for significance. It would take 27 h/day of M-rated game play to produce clinically noticeable changes in aggression. Effect sizes for aggression/prosocial outcomes were little different than for nonsense outcomes. Evidence from this study does not support the conclusion that aggressive video games are a predictor of later aggression or reduced prosocial behavior in youth.