Polce EM, Kunze KN, Fu M, et al. - This study was intended to train supervised machine learning (SML) algorithms to predict satisfaction after total shoulder arthroplasty (TSA) and set up a clinical tool for an individualized assessment of patient-specific risk factors. Between January 2014 and February 2018, researchers conducted a retrospective review of primary anatomic and reverse TSA patients. They assessed a sum of 16 demographic, clinical, and patient-reported outcomes for predictive value. The data suggest that best conducting the SML model indicated excellent discrimination and adequate calibration for prognosticating satisfaction following TSA and was applied to develop an open-access, clinical-decision making tool. Nevertheless, in different geographic locations and patient populations, rigorous external validation is essential prior to the assessment of clinical utility. The data demonstrate that given that this tool is based on partially modifiable risk factors it may improve shared decision making and permit for periods of targeted, preoperative health optimization efforts.