As a severe disaster in the whole world, hail poses significant threats to the life of human and other animals, and leads to great economic loss. To prevent potential damages caused by hail, it is important to clearly understand and accurately predict the mechanical response of natural hailstones to impact, which is, however, extremely challenging due to the randomness of the hailstone’s geometric configuration, spatial location, spatial attitude, motion state and so on. In this work, randomness was incorporated into these parameters based on statistical analysis of experimental data, and a new approach was then proposed to generate a spatial field consisting of sparse and random hailstones by using Monte Carlo algorithm. Based on the statistical data of falling hailstones in the literature, a random hail field was established via the proposed approach and then used to simulate the response of stochastic hailstones impacting on a rigid target using the finite element method. The random response of hail impact was well captured by simulation results, and the convergence and reliability of the proposed modelling approach was validated by a comparison between simulation results and the experimental data in the literature. It was finally discussed the promising potential of the application of the proposed model.