During the past few decades, both the diagnosis and the treatment of rotator cuff injuries have improved. Despite this, uncertainties remain. For example, when a patient first presents to the clinician’s office, we do not have good criteria for deciding whether he or she should be treated nonoperatively. This is the case partly because rotator cuff injuries are common and often minimally symptomatic, especially in the elderly, sometimes making it difficult to determine whether the symptoms in those patients who do come in with shoulder complaints are concordant with (or caused by) findings on imaging tests. And, among those patients who have symptoms, some are able to cope with them. Therefore, most clinicians recommend that treatment begin with nonoperative measures and surgeons see those whose symptoms fail to resolve. However, other clinicians recommend surgery in many of their patients with a cuff tear for fear of the tear getting more severe over time (ie, tear progression). While most surgeons have experience with an individual returning to their office several years after diagnosis of a small tear that has now become massive, our knowledge of tear progression is limited to a small number of shoulders over short time periods of a few years [3, 9] (Fig. 1).