Current formulations proposed by Eurocode 8 part 3 for the inelastic deformations of existing reinforced concrete members are assessed aimed at finding out whether they are representative of beams with different cross-sectional aspect ratio, in particular for wide beams (WB) contrasted with conventional deep beams (DB). The current approach shows that WB present larger ultimate chord rotation but lower chord rotation ductility than DB despite the similar curvature ductilities, due to lower plastic hinge lengths in WB. Results of the large experimental database on the basis of current model are disaggregated for DB and WB. Predicted chord rotations appear to be significantly biased for both types, especially at ultimate; the actual model provides conservative values for DB and non-conservative values for WB, which suggest that plastic hinge length may be even greater for DB in comparison to WB. Hence, some different feasible modifications of both pure empirical and fundamental formulations for chord rotations are proposed, in order to reduce the bias and thus increasing the robustness of the model against cross-sectional shape variability.