Hip fracture related to osteoporosis and metastatic disease is a major cause of morbidity and mortality. An accurate and precise method of predicting proximal femoral strength and fracture location would be useful for research and clinical studies of hip fracture. The goals of this study were to develop a structural modeling technique that accurately predicts proximal femoral strength;​ to evaluate the accuracy and precision of this predicted strength on an independent data set;​ and to evaluate the ability of this technique to predict fracture location. Fresh human cadaveric proximal femora with and without metastatic lesions were studied using computed tomography scan-based three-dimensional structural models and mechanical testing to failure under single-limb stance-type loading. The models understated proximal femoral strength by an average of 444 N, and the precision of the predicted strength was +/- 1900 N. Therefore, the ability to predict hip strength in an individual subject is limited primarily by the level of precision, rather than accuracy. This level of precision is likely to be sufficient for many studies of hip strength. Finally, these models predict fractures involving the subcapital and cervical regions, consistent with most fractures produced experimentally under single-limb stance-type loading.