Yoga programs within schools have become more widespread but research regarding the potential effect on academic achievement remains limited. This study cluster-randomized 112 students within a single New York City public high school to participate in either school-based yoga or physical education (PE) for an entire academic year. The primary outcome was mean annual grade point average (GPA). Psychosocial variables (self-regulation, executive function, well-being, and mindfulness) were examined as mediators. The study's primary hypothesis that yoga would improve academic performance was not supported by intent to treat analysis; however, a significant interaction was observed between class assignment and class participation. Among students with higher participation, those assigned to yoga classes had a significantly higher GPA. For example, at 49 classes of participation for both groups, students assigned to yoga classes had an estimated 2.70 higher mean GPA (effect size = 0.31) than students assigned to PE.