Drying of colloidal suspensions often yields intriguing, complex and generally undesirable crack patterns. Despite a consensus on the main underlying physical mechanisms that lead to their formation, quantitative prediction seems at present out of reach. Notably, a prerequisite for this is the determination of the mechanical stress field. Here, our point of view is to provide a sound and well-established continuum mechanics model that can be used safely and easily by a large community and particularly by the fracture mechanics one. We show by comparison with beam deflection experiments, that (i) the use of Biot's linear poroelasticity theory allows for the prediction of the stresses that lead to the crack formation and that (ii) the setup allows for the identification of some poroelastic constants. This evidenced that the hypothesis of incompressible constituents is reasonable for the nanolatex suspensions considered here, greatly simplifying the constitutive equations and reducing the number of material constants from four to two.