Abstract Purpose The purpose of this study was to assess the role of CT scan in preoperative evaluation of glenoid and humeral head bone loss in cases of recurrent anterior shoulder joint dislocation using on track and off track concept and whether it can be used to predict engagement during arthroscopy. Patients and Methods In the period from June 2019 to October 2022, this study was conducted in Ain Shams University hospitals including 40 cases with recurrent anterior shoulder dislocation. CT scan with 3D reconstruction was done to calculate the width of the glenoid track using the best fit circle method as well as the Hill Sachs interval. We used the on-track off-track method to predict engagement and these results were compared with the findings related to engagement seen during arthroscopy Results Using the on-track /off-track technique on CT images, 22 of the 25 engaging (off-track) lesions was predicted correctly (sensitivity, 88%).While 13 of the 15 cases that did not engage (on-track) was correctly predicted (specificity, 86.7%). Overall, the accuracy of the on-track off-track method was 87.5% with a positive predictive value (PPV) of 91.7% and negative predictive value (NPV) of 81.3%. The on-track /off-track method was found to be a significant predictor of engagement as there is no statistically significant difference (p = 0.18) between this method and arthroscopic findings. Conclusion Our study showed that using the on-track off-track method by CT scan can accurately assess the glenoid and humeral head bone losses seen in patients with anterior shoulder instability for predicting the presence of engagement, or off-track, lesions. This technique can be used for preoperative evaluation to help in deciding the type of surgical procedure performed on patients with anterior shoulder instability.