The constantly increasing demand of the world for energy has led to a remarkable advancement in the offshore geotechnical engineering field that is closely related to the oil and gas production. In this industry, floating production facilities are connected with the seabed through mooring lines that are attached to anchor foundations. In this context, suction anchors have been increasingly installed to ensure good mechanical performance under variable operational loads. Therefore, they should be able to withstand the horizontal, vertical and torsional (HVT) loads stemming from the inclined loading imposed by the mooring chains. The undesired torsion is induced in the problem when the anchor is not perfectly aligned with the floating facility due to imperfections during installation. This work constitutes a comparative study between three distinctive numerical approaches that revolve around the undrained capacity of misaligned suction anchors founded in normally consolidated clay. Its main aim is to unravel the strengths and weaknesses of the well established methods of analysis used nowadays. The effectiveness of the two current adopted methods is shown with reference to a benchmark study that involves a thorough 3D finite element investigation on the impact of misalignment on the capacity of suction anchors following a total stress analysis (TSA) within Plaxis 3D. Even though this type of analysis (TSA) is widely employed nowadays, it is considered as a rough approximation of the actual soil behaviour under undrained conditions. This statement acted as an incentive to conduct an effective stress analysis (ESA) of the same problem. For this purpose, different calibrations of an advanced Cam - Clay type constitutive model (the Soft Soil model) were realised to capture the soil conditions of the benchmark study, which embraced a simplified Tresca failure criterion. Additionally, a simpler 2D numerical approach of the same problem was realised with the use of SPCalc, which is a special purpose tool that accounts for 3D effects through the application of side shear and reduction factors. In this case, misorientation is modelled by the reduction of the soil - anchor interface resistance. Failure envelopes are derived for the two current approaches and compared for three different length-to-diameter anchor ratios (1.5, 3 and 6), and for misalignment angles in the range of 0° to 10°. The outcomes of this study give a better insight into the practical implications of certain modelling assumptions. In particular, it is shown that the influence of misalignment on the undrained HV capacity may be in fact non-negligible, while the expected conservatism arising from simplified 2D analysis is quantitatively pointed out. Even though ESA is thought to enhance the mechanical soundness of undrained modelling, it creates additional difficulties in terms of soil parameter calibration. Therefore, the close relationship between HVT failure envelopes and the modelling of clay strength in the effective stress framework is critically discussed with respect to the preliminary results of the benchmark study.