A novel elasto-geometric model is introduced that simultaneously estimates joint compliances and geometric error parameters by employing the loaded double ball bar apparatus. The model parameters are estimated from tests at different force levels by distinguishing between errors that change with the applied force (compliance effect) from those that do not (geometric effects). At lower forces, the geometric errors are dominant while at higher forces compliance errors dominate. Using all data to build a single global geometry and compliance set of parameters (global constant compliance model), the radial volumetric variations due to geometric errors and compliance are estimated at 0.019 mm and 0.046 mm, respectively, making compliance dominant by more than three times. The impact of dominant and non-dominant equivalent global compliance CXXX, CYYY, CXYX, CCXY, CCYY, and CCCY on the loaded circular test readings at the highest force level of 742 N are predicted to be around 0.045, 0.034, 0.00058, 0.0022, 0.0014, and 0.0045 mm peak-to-peak, respectively. The impact of loaded geometric parameters EXX1, EYY1, EYX2, EXY2, EC(0Y)X, EXt0, and EYt0 on the loaded circular test readings is predicted to be around 0.019, 0.014, 0.0074, 0.012, 0.00017, 0.0076, and 0.0012 mm peak-to-peak, respectively. The dominant global compliances are CXXX and CYYY at 0.0619 and 0.0461 μm / N , respectively.