The paper describes the application of a computer intensive methodology to evaluate antenna susceptibility to random physical damage for a dipole and a bow-tie antenna. For each randomly generated damage pattern antenna parameters are simulated and a statistical comparison is performed. Computer intensive statistical methods like the Bootstrap, used here for postprocessing, make fewer initial assumptions than classical statistical methods when modelling a problem, and are robust against small deflections from these assumptions. The bias-corrected and accelerated (BCa) Bootstrap method is used for confidence interval estimates. The dipole antenna is found to be more robust to numbers of small (similar to 100th of a wavelength) circular holes penetrating the structure than the bow-tie antenna.