There is a longstanding stereotype that segregated high schools serving a large majority of minority students are unsafe. But it is unclear if this stereotype has any merit. I use a QuantCrit theoretical framework along with data from the nationally representative Education Longitudinal Study and multilevel logistic regression models to explore how school racial segregation is related to student reports of victimization in high school, as well as student reports of peer disruptive behavior. Results indicate that attending a segregated minority high school is not related to increased chances of victimization, but students who attend such schools face a more disrupted learning environment.