Abstract Precompression stress ( σ p ) is commonly used as an indicator for the load carrying capacity of a soil and its stress history. Several calculation methods have been proposed to determine σ p , but they have proven to be not interchangeable. The aim of this study was to perform a functional evaluation of seven existing methods to test which method would be most recommended to assess soil strength in cultivated fields. This was achieved by investigating the relation of the obtained σ p of undisturbed soil samples to other soil properties and stress history. Additionally, the robustness of the methods was evaluated by checking their sensitivity to perturbation of the measured strains. The two most commonly used methods, the Casagrande method and Gompertz model fit method, were found to be the least suitable to calculate σ p . These methods lack robustness and do not yield a significant relation with the relevant soil properties or stress history. Better results were obtained with Intersecting linear regressions or at the inflection point of a fourth‐degree polynomial.