This work proposes probabilities of multimatrix distributions and simplifies the affine shape theory under the setting of real normed division algebras and elliptically contoured laws. The contribution of the paper, valid for all real normed division algebras and elliptical distributions, involves: 1) computable probabilities for several matrices in multimatrix distributions2) computable probability for several multimatrix variate generalised Wishart distributions3) probability study of the multimatrix variate Pearson type II distribution4) a new setting for generalised estatistical shape theory5) a dynamic molecular docking based on matrix probabilities, under dependent samples of a reported inhibitor in two cavities of SARS-CoV-2. © Indian Statistical Institute 2025.