Introduction: There has been a recent resurgence of interest in arthroscopic primary repair of proximal anterior cruciate ligament (ACL) tears. Patient selection is critical but not much is currently known on what predicts the possibility of repair. Goal of this study was therefore to assess predictive factors for the possibility of arthroscopic primary ACL repair. Methods: In this retrospective case–control study, all patients undergoing ACL surgery in a ten-year interval were included. Patients were treated with primary repair if there was a proximal tear and good tissue quality, or otherwise underwent ACL reconstruction. Collected data were age, gender, BMI, injury-to-surgery delay, injury mechanism and concomitant injuries. Receiver operating characteristic curves were used to find cutoff values, and all significant dependent variables were used in multivariate logistic analysis to assess independent predictors for the possibility of primary repair. Results: Three hundred sixty-one patients were included, of which in 158 patients (44%) primary repair was possible. Multivariate analysis (R2 = 0.340, p < 0.001) showed that age > 35 years (Odds ratio [OR] 4.2, 95% CI 2.4–7.5,p < 0.001), surgery within 28 days (OR 3.3, 95% CI 1.9–5.7, p < 0.001), and BMI < 26 (OR 1.9, 95% CI 1.1–3.3,p = 0.029) were predictive for the possibility of primary repair, and lateral meniscus injury presence decreased the likelihood of repair (OR 0.5, 95% CI 0.3–0.8, p = 0.008). Conclusion: In this large cohort study, it was noted that 44% of patients had repairable ACL tears. Primary repair was more likely to be possible in older patients, patients with lower BMI and when surgery was performed within four weeks of injury.