We examined recent process models of accumulated disadvantage with an intersectional lens in order to provide a more complete picture of how disadvantages across punishment and math trajectories can accumulate over time and disparately affect marginalized race-gender groups. Using structural equation modeling (SEM) with a nationally representative longitudinal study of high school students (HSLS-09), we found that punishment trajectories were influenced by math and vice versa, as well as that these relationships differed across math performance and various aspects of math attitudes, including efficacy, utility, and identity. Furthermore, we found that gender, race, and race-gender groups experienced significantly different relationships. When considering the intersection of punishment and math disadvantages, these differences appear to not only accumulate disadvantages within punishment and math trajectories but also across them for marginalized race-gender groups. This was especially true for Black males. We conclude with a discussion of implications for policy and practice.