Sorption isotherm models are frequently used to understand the interactions of water with cellulosic materials. Many commonly used models, such as the Guggenheim-Anderson-de Boer, Hailwood-Horrobin, and Dent isotherm models, are based on idealized physical systems. In the scientific literature, these models are often used to predict properties of wood such as the monolayer capacity, which is a representation of the number of available sorption sites in the material. Reporting such properties assumes that an idealized system adequately represents actual wood-water interactions. These models can be reduced to the same form, where the ratio of water activity, aw, to equilibrium moisture content (EMC), is a second-order polynomial function of aw. Here we review twelve models that have a parabolic form, fit these models to high quality water vapor sorption isotherm data for wood at multiple temperatures, and calculate physical properties from the model parameters. Moreover, variability in model predictions arising from measurement uncertainties in aw and EMC values is propagated using a Monte Carlo method. Predicted physical properties are tested against independently measured values of hydroxyl accessibility, relative amounts of primary and secondary water, and enthalpy of sorption. None of the models accurately predicts any of the physical properties, even when the reported measurement uncertainties are increased by a factor of three. Therefore, it must be concluded that parabolic sorption isotherm models are not valid for water in wood cell walls.