[en] Lattice calculation by Monte Carlo methods is expected to compete against deterministic tools such as collision probability or the method of characteristics, provided that its calculation time can be shortened. In this paper, we investigate the possibility of using cross-section probability tables conjointly with optimized energy meshes in a multigroup Monte Carlo algorithm. Regarding the Monte Carlo method, probability tables cross-sections are frequently used in the unresolved range. In deterministic lattice codes, this formalism can serve during the self-shielding stage, typically in a subgroup method. We propose using probability tables cross-sections on the entire energy range in a Monte Carlo algorithm, together with an adequate energy structure. Numerical results are proposed using legacy benchmarks and comparison with multigroup, deterministic and pointwise, Monte Carlo solutions is discussed. We demonstrate that using probability tables in Monte Carlo multigroup calculations with only 295 or 361 groups can lead to accurate results, as far as the probability tables are consistently evaluated. (authors)