This paper concerns, the effect of (unknown) residual stresses in the near‐surface region of a sample on outcomes of an indentation plastometry technique for obtaining stress–strain relationships, using relatively large spherical indenters. The technique is based on the iterative finite element method (FEM) simulation of indentation, so as to infer the true stress–strain curve of the material (with the target outcome being the profile of the residual indent). It is expected that residual stress levels (if significant compared with the yield stress) are likely to influence the outcome, and indeed this forms the basis of many attempts to measure residual stresses in this way (using a known stress–strain curve). However, there are important issues of sensitivity here, which affect the reliability of such procedures and are also relevant to the accuracy of stress–strain curves inferred from measurements made on samples containing unknown levels of residual stress. Herein, both experimental work on samples with (equal‐biaxial) residual stresses created by the application of external loads and extensive FEM modeling to explore various scenarios are covered. The main conclusion is that the sensitivities involved are in general very low, particularly with relatively deep indentation. Inferred stress–strain curves are thus likely to be quite accurate, even in the presence of relatively high levels of residual stress. Conversely, the measurement of residual stress levels via this procedure is likely to be rather inaccurate, although the reliability is improved using shallow penetration (low ratio of depth‐to‐indenter radius). It is also noted that tensile residual stresses tend to influence outcomes more strongly than compressive ones.