Background: Virtually all high schools offer a range of courses to allow students to enroll in four years of high school mathematics. However, only two thirds of U.S. high school graduates took mathematics courses each school year. Purpose/Research Question: This study addresses three research questions: First, how do students' math course enrollment and motivational beliefs (i.e., self-efficacy in math, math utility, interest in math, and college expectations) differ by math track? Second, what is the relationship between students' motivational beliefs and their decision to take four years of math? Third, to what extent does this relationship vary by math track and whether a student passes or fails a math course? Much of the relevant prior literature approaches these relations primarily from an individualistic psychological perspective, viewing motivation as a student-level attribute that similarly effects students' decision-making process. By contrast, our analyses take a more contextual approach, focusing particular attention on the ways in which students' math track placements shape their academic approaches and moderate the link between motivation and course-taking. Research Design: This study uses secondary restricted-access data from the nationally representative Education Longitudinal Study (ELS: 2002). Students were surveyed and tested in mathematics during the base year (2002). In the follow-up (2004) year, data collectors requested academic transcripts for all participants along with follow-up student surveys and an additional math exam. Findings: Our results coincide with previous motivation research that shows that students opt to take additional math courses when they are interested in math, consider themselves skillful in math, and have high college expectations. But the motivational predictors of math course enrollment vary with students' initial math placement. For above-track students, interest in math is the strongest indicator that they will take four years of math, followed by self-efficacy in mathematics and college expectations, respectively. In contrast, for both low-track and on-track students, the strongest indicator of taking four years of math is college expectations. Conclusions: Our study focused on students' motivation and course enrollment, but this does not diminish the importance of tracking, curricular rigor, and teacher pedagogy. This study provides an additional way to improve inequities in math course enrollment, which is by making explicit recommendations for enhancing students' motivation. Understanding which particular beliefs have the greatest influence on specific student groups allows educators to appropriately allocate limited resources and increase math course enrollment. This would likely be more effective than a one-size-fits-all approach.