Delaying tracking, extending students' exposure to a general academic education and increasing their time on task on basic competences (reading, mathematics) could improve academic outcomes. To test the hypothesis that delayed vocational streaming improves academic outcomes, this paper analyzes Poland's significant improvements in international achievement tests and the restructuring of the system which expanded general schooling. Estimates using propensity-score matching and difference-in-differences estimates show that delaying vocational education and increasing time on task have a positive and significant impact on student performance on the order of a standard deviation.