At many higher education institutions, admissions decisions often rely on standardized test scores and high school grades; yet, they are less reliable predictors for applicants falling slightly below cutoff points, what we call borderline applicants. Since borderline applicants are often from underrepresented backgrounds and diverted to 2-year institutions, this may potentially jeopardize efforts to increase campus diversity. Using a mixed-methods approach, two studies investigated an "admissions experiment" designed to increase campus diversity by admitting 34 borderline applicants into a summer bridge program. Study 1 compared program participants' performance to two comparison samples of regularly admitted students (N = 912). Compared with a matched sample, borderline students performed better after the first semester and comparably after the first year. Study 2 identified program components that helped or undermined participants' college adjustment and performance. Results suggested several program improvements that might enhance underrepresented students' performance and experiences on 4-year campuses.