In order to optimize materials properties, in many cases a deeper understanding of the relationship between the chemical-atomistic structure and the physical properties of the solid and fluid phases of the material is necessary. Monte Carlo simulation is a tool that allows the reliable calculation of thermodynamic properties of strongly interacting many-body condensed matter systems. Given a model of effective interatomic or intermolecular interactions (drawn either from quantum-chemical-type interactions or from analysis of suitable experimental data), macroscopic bulk properties of a material can be simulated, as well as interfacial phenomena and certain kinds of slow dynamic processes (of relaxational or diffusive type). After a brief review of the foundations of this approach in statistical mechanics, the wide potential of this method is illustrated with examples taken from magnetism, metallurgy and amorphous polymeric materials. Strengths and limitations of this atomistic approach towards modeling materials properties are discussed and directions of future research are spelled out.