Clinical sequelae have been associated with corrosion of modular junctions in total hip arthroplasty. These sequelae, including local adverse reactions to corrosion by-products, have been observed with metal-on-metal and conventional metal-on-polyethylene systems. Retrieval analysis can quantify material loss at modular interfaces, permitting correlation with prosthetic and patient factors. The purpose of this study was to quantify and characterize material loss at the head-stem junction of retrieved metal-on-polyethylene systems exhibiting imprint corrosion to identify parameters associated with corrosion in vivo and to establish a model of error for ex vivo material loss measurements. A coordinate measuring machine (CMM) was used to measure volume loss, maximum linear depth (MLD), and taper angle of 18 head bores and four stem tapers of the same manufacturer. The CMM used a 3-mm diameter ruby stylus to trace 72 axial lines across the taper surfaces. Five heads were bivalved for white light profilometry to validate MLD measured by the CMM. An error budget model, comparing interferometry with CMM measurements, was mathematically calculated and experimentally measured, providing error bounds for interpreting CMM-based material loss data. Median volume loss and MLD on the head bores were 1.63 (−2.56,12.49) mm³ and 27.94 (1.85,74.93) µm, respectively. Median head taper angle was statistically smaller than stem tapers angles, indicating distal engagement (p = 0.009). Maximum material loss was found to be on the portion of the head bore closest to the lesser trochanter, in an area of high stress and greatest contact with stem taper. This suggests continuous disturbance of the protective oxide in this region, promoting corrosive action. Stress may play a role in promoting corrosion, but further experimentation is necessary to fully characterize imprinting processes. A model of error was created for the interpretation of CMM measurements. The magnitude of material loss closely matched the interferometry results.