An extensive body of research describes the stark social and academic disparities that characterize large, comprehensive public high schools. Advocates have argued that smaller high schools, with their necessarily more constrained curricula, have the potential to improve student academic outcomes and their equitable distribution. Using propensity score matching within a multilevel modeling framework, we explore this assertion in New York City, which has been at the center of small school reform. We find a small, positive impact of attending a new, small high school on progress through the math curriculum for low-achieving students, but no effect for average-achieving students, and a negative impact for higher achieving students. Despite any positive findings, the typical small school student still failed to complete Algebra II/Trigonometry--the lowest level course deemed "college preparatory" by the district.