Objectives. Periprosthetic femoral fractures (PFFs) have a higher incidence with cementless stems. The highest incidence among various cementless stem types was observed with double-wedged stems. Short stems have been introduced as a bone-preserving alternative with a higher incidence of PFF in some studies. The purpose of this study was a direct load-to-failure comparison of a double-wedged cementless stem and a short cementless stem in a cadaveric fracture model. Methods. Eight hips from four human cadaveric specimens (age mean 76 years (60 to 89)) and eight fourth-generation composite femurs were used. None of the cadaveric specimens had compromised quality (mean T value 0.4 (-1.0 to 5.7)). Each specimen from a pair randomly received either a double-wedged stem or a short stem. A materials testing machine was used for lateral load-to-failure test of up to a maximal load of 5000 N. Results. Mean load at failure of the double-wedged stem was 2540 N (1845 to 2995) and 1867 N (1135 to 2345) for the short stem (p < 0.001). All specimens showed the same fracture pattern, consistent with a Vancouver B2 fracture. The double-wedged stem was able to sustain a higher load than its short-stemmed counterpart in all cases. Failure force was not correlated to the bone mineral density (p = 0.718). Conclusion. Short stems have a significantly lower primary load at failure compared with double-wedged stems in both cadaveric and composite specimens. Surgeons should consider this biomechanical property when deciding on the use of short femoral stem. Cite this article: A. Klasan, M. Bäumlein, P. Dworschak, C. Bliemel, T. Neri, M. D. Schofer, T. J. Heyse. Short stems have lower load at failure than double-wedged stems in a cadaveric cementless fracture model. Bone Joint Res 2019;8:489–494. DOI: 10.1302/2046-3758.810.BJR-2019-0051.R1