The article considers a way to compare large bulks of experimental data with theoretical calculations, in which the quality of theoretical models is clearly demonstrated graphically. The main idea of the method consists in grouping physical observables, represented by experiment and theoretical calculation, into samples, each of which characterizes a certain physical process. A further choice of a convenient criterion for comparing measurements and calculations, its calculation and averaging within each sample and then over all samples, makes it possible to choose the best theoretical model in the entire measurement area. Published theoretical data of the three-fluid dynamic model (3FD) applied to the experimental data from heavy-ion collisions at the energy range $\sqrt{s_{NN}}\,=\,2.7 - 63$ GeV are used as example of application of the developed methodology. When analyzing the results, the quantum nature of the fireball, created at heavy ion collisions, was taken into account. Thus, even at energy $\sqrt{s_{NN}}\,=\,63$ GeV of central collisions of heavy ions, there is a nonzero probability of fireball formation without ignition of the quark-gluon plasma (QGP). At the same time, QGP ignition at central collision energies above at least $\sqrt{s_{NN}}\,=\,12 GeV occurs through two competing processes, through a first-order phase transition and through a smooth crossover. That is, in nature, these two possibilities are realized, which occur with approximately the same probabilities.