The selection of a proper sorbent for a given application is a complex problem. The design and efficiency of adsorption processes require an equilibrium adsorption model. Linear transformation is one of the methods available to estimate the adjustable parameters of isotherm models but possesses limitations compared to nonlinear regressions. A different approach to calculate predicted equilibrium isotherm values leading to an alternative nonlinear regression is presented in this paper and compared with usual regression methods. Adsorption isotherm data of gaseous THT onto three activated carbon materials constitute an experimental basis for the discussion. Assessment of the goodness-of-fit of the Langmuir model is supported by different selected test functions. The new nonlinear approach did not obtain the best results for each test function, but raises questions about the inherent combined error in regression procedures.