Background: A Hill-Sachs lesion is a common injury associated with anterior glenohumeral instability, and a Hill-Sachs lesion that engages with the anterior glenoid rim is 1 factor related to recurrent instability. In a cadaveric study in 2007, a new concept, the "glenoid track," was proposed to evaluate the risk of engagement of Hill-Sachs lesions with the glenoid. Purpose: To investigate the glenoid track in vivo using a custom-developed noninvasive motion analysis system. Study Design: Descriptive laboratory study. Methods: Using a wide-gantry magnetic resonance imaging (MRI) scanner, the right shoulders of 30 healthy volunteers were examined. The MRI scans were taken of the right arm in 7 static supine positions from 0 degrees to maximum abduction, keeping maximum external rotation and horizontal extension. Using the custom motion analysis system, 3-dimensional models of the scapula and humerus were created from the MRI data. Then, the movement of the humerus and scapula was calculated using voxel-based registration of each model, and the motion of the glenoid on the humeral head was analyzed. Results: The models demonstrated that glenoid contact shifted from the inferomedial to the superolateral portion of the humeral head. The mean widths of the glenoid track with the arm at 60 degrees, 90 degrees, 120 degrees, and 150 degrees of abduction were 20.7 +/- 4.5 mm, 19.4 +/- 3.9 mm, 18.9 +/- 2.7 mm, and 18.7 +/- 2.5 mm (89%, 83%, 82%, and 81% of the glenoid width), respectively. The width of the glenoid track at 60 degrees of abduction was significantly greater than those at 90 degrees, 120 degrees, and 150 degrees of abduction (P = .0472, .0148, and .0083, respectively). There were no significant differences among widths measured at 90 degrees, 120 degrees, and 150 degrees of abduction. Conclusion: The existence and widths of the glenoid track were confirmed in vivo.