BACKGROUND Shoulder instability is a common problem in American football players entering the National Football League (NFL). Treatment options include nonoperative and surgical stabilization. OBJECTIVE This study evaluated how the method of treatment of pre-NFL shoulder instability affects the rate of recurrence and the time elapsed until recurrence in players on 1 NFL team. METHODS Retrospective cohort. METHODS Medical records from 1980 to 2008 for 1 NFL team were reviewed. There were 328 players included in the study who started their career on the team and remained on the team for at least 2 years (mean, 3.9 years; range, 2-14 years). The history of instability prior to entering the NFL and the method of treatment were collected. Data on the occurrence of instability while in the NFL were recorded to determine the rate and timing of recurrence. RESULTS Thirty-one players (9.5%) had a history of instability prior to entering the NFL. Of the 297 players with no history of instability, 39 (13.1%) had a primary event at a mean of 18.4 ± 22.2 months (range, 0-102 months) after joining the team. In the group of players with prior instability treated with surgical stabilization, there was no statistical difference in the rate of recurrence (10.5%) or the timing to the instability episode (mean, 26 months) compared with players with no history of instability. Twelve players had shoulder instability treated nonoperatively prior to the NFL. Five of these players (41.7%) had recurrent instability at a mean of 4.4 ± 7.0 months (range, 0-16 months). The patients treated nonoperatively had a significantly higher rate of recurrence (P = 0.02) and an earlier time of recurrence (P = 0.04). The rate of contralateral instability was 25.8%, occurring at a mean of 8.6 months. CONCLUSIONS Recurrent shoulder instability is more common in NFL players with a history of nonoperative treatment. Surgical stabilization appears to restore the rate and timing of instability to that of players with no prior history of instability.