The accuracy of rockfall trajectory simulations depends to a large extent on the calculation of the rebound of falling boulders on different parts of a slope where rockfalls could occur. The models commonly used for rebound calculation are based on restitution coefficients, which can only be calibrated subjectively in the field. To come up with a robust and objective procedure for rebound calculation, a stochastic impact model associated with an objective field data collection method was developed and tested in this study. The aims of this work were to assess the adequacy of this approach and to evaluate the minimum amount of field data required to obtain simulation results with a satisfactory level of predictability. To achieve these objectives, the rebound calculation procedure developed was integrated into a three-dimensional rockfall simulation model, and the simulated results were compared with those obtained from field rockfall experiments. For rocky slopes, the simulations satisfactorily predict the experimental results. This approach is advantageous because it combines precise modelling of the mechanisms involved in the rebound and of their related variability with an objective field data collection procedure which basically only requires collecting the mean size of soil rocks. The approach proposed in this study therefore constitutes an excellent basis for the objective probabilistic assessment of rockfall hazard.