Glenoid bone loss has been reported to occur in as many as 86% of patients with recurrent shoulder stability. This systematic review evaluated the amount of glenoid bone loss associated with recurrent shoulder dislocation or subluxation and with worse patient-reported outcomes after arthroscopic Bankart repair. We hypothesized that the percentage of glenoid bone loss associated with recurrent instability after arthroscopic Bankart repair is lower than the previously proposed critical value of 25%. The systematic review included 528 patients with glenoid bone loss from 3 clinical studies. The percentage of bone loss was the value quantified and reported in each study. Failure was defined as recurrent dislocation or subluxation. The percentage of glenoid bone loss associated with recurrent shoulder dislocation or subluxation after arthroscopic Bankart labral repair was analyzed with receiver operating characteristic (ROC) curve analysis. Recurrent dislocation or subluxation occurred in 23.7% (125) of 528 patients in the pooled study cohort. There was a significant difference in age between those in whom the arthroscopic Bankart repair failed and those in whom it did not (22.9 versus 24.3 years; p = 0.009). The ROC curve analysis demonstrated that ≥16.0% glenoid bone loss was predictive of recurrent shoulder dislocation or subluxation (Youden index = 0.59, sensitivity = 80%, specificity = 80%). In patients who did not sustain a recurrent dislocation or subluxation, the ROC curve analysis demonstrated that 20.0% glenoid bone loss was predictive of a Single Assessment Numeric Evaluation (SANE) score of <85% (Youden index = 0.93, sensitivity = 93%, specificity = 100%). The critical amount of glenoid bone loss associated with an increased risk of persistent instability was found to be less than previously reported. Glenoid bone loss of ≥16.0% was found to place patients at higher risk for recurrent shoulder dislocation or subluxation after treatment with arthroscopic Bankart repair alone. Therapeutic Level IV . See Instructions for Authors for a complete description of levels of evidence.