Family connection has demonstrated protective effects on violence perpetration, victimization, and witnessing in the general U.S. adolescent population. However, several studies examining the impact of family connection on violence exposure in adolescents living in low-resource urban environments have failed to demonstrate similar protective effects. We interviewed male youth in low-resource neighborhoods in Philadelphia recruited through household random sampling. Adjusted logistic regression was used to test whether a supportive relationship with an adult family member was inversely associated with violence involvement and violence witnessing. In 283 youth participants aged 10 to 24 years, 33% reported high violence involvement, 30% reported high violence witnessing, and 17% reported both. Youth who identified at least one supportive adult family member were significantly less likely to report violence involvement (odds ratio [OR] = 0.35; 95% confidence interval [CI] = [0.18, 0.69]) and violence witnessing (OR = 0.46; 95% CI = [0.24, 0.88]). Youth with two supportive parents, and those with supportive mothers only, also demonstrated significant inverse associations with violence involvement. Supportive parental relationships were inversely but not significantly related to witnessing violence. The findings suggest that supportive parental relationships may not prevent youth in low-resource neighborhoods from witnessing violence but may help prevent direct violence involvement. Next studies should be designed such that the mechanisms that confer protection can be identified, and should identify opportunities to bolster family connection that may reduce adolescent violence involvement among youth in low-resource urban environments.