This article examines the effects of academic tracking in secondary schools on educational stratification and considers how that tracking may affect levels and dispersions of academic achievement and high school graduation rates among social groups. Data from the High School and Beyond survey of students who were sophomores in 1980 show that placement in the college track substantially benefits growth in mathematics achievement and the probability of high school graduation, even when measured and unmeasured sources of nonrandom assignment to tracks are taken into account. Track assignment reinforces preexisting inequalities in achievement among students from different socioeconomic backgrounds. However, track assignment and differential achievement in tracks partially compensate blacks and girls for their initial disadvantages and makes racial and sexual inequalities smaller than they may have otherwise been. The article provides qualified support for the view that students are assigned to the tracks that provide the greatest reward to their measured background characteristics.