Background: Variable clinical outcomes of tibial tuberosity transfer surgery have been reported. Hypotheses: The biomechanical outcome of surgery is patient-specific; no single procedure produces superior results for all patients. Use of patient-specific computer models can optimize choice of procedure. Study Design: Computer simulation study using clinical data. Methods: We used patient-specific multibody models of the patellofemoral joints of 20 patients with a diagnosis of patellar subluxation and osteoarthritis. Four tibial tuberosity transfer procedures (two anterior and two anteromedial) were simulated for each patient and compared with their preoperative model. Results: When results for all patients were averaged, all simulated operations produced a statistically significant decrease in surface-wide mean contact stress, although no significant difference was found among them. Conclusions: The simulated surgical outcomes were patient-specific: no single procedure was consistently superior at decreasing peak or mean stress and each procedure produced a potentially detrimental outcome, an increase in either mean stress or peak stress, in at least one patient. Clinical Relevance: Computer simulation may serve as a valuable tool for tailoring procedures to specific patients.