Various methods that link a representative pore‐throat size to permeability k and porosity ϕ have been proposed in the literature for rock typing (i.e., identifying different classes of rocks and petrofacies). Among them, the Winland equation has been used extensively, although when it was first proposed, it was based on experiments. Because of its empiricism, the interpretation of the parameters of the Winland model and their variations from one rock sample or even one rock type to another is not clear. Therefore, the main objectives of this study are (1) to propose a new theoretical approach for identifying rock types that is based on the permeability k and the formation-resistivity factor F and (2) to provide theoretical insights into, and shed light upon, the parameters of the Winland equation, as well as those of other empirical models. We present a simple, but promising, framework and show that accurate identification of distinct petrofacies requires knowledge of the formation factor, which is measured routinely through petrophysical evaluation of porous rocks. We demonstrate that, although some rock samples might belong to the same type on the k‐vs.‐1/F plot, they might appear scattered on the k‐vs.‐ϕ plot and, thus, could seemingly correspond to other types. This is because both k and F are complex functions of the porosity, whereas the porosity itself is simply a measure of the pore volume (PV), and does not provide information on the dynamically connected pores that contribute to both k and F. We also show that each rock can be represented by a characteristic pore size Λ, which is a measure of the dynamically connected pores. Accurate estimates of Λ indicate that it is highly correlated with the permeability.