estimating probabilities by running random experiments many times. It's one of the most powerful techniques in data science.