The aim of this paper is to present some of the features of the non-Poissonian statistics of the spread of a disease like COVID19 to a community of chemical-physicists, who are more used to particle-based models. We highlight some of the reasons why creating a ``transferable'' model for an epidemic is harder than creating a transferable model for molecular simulations. We describe a simple model to illustrate the large effect of decreasing the number of social contacts on the suppression of outbreaks of an infectious disease. Although we do not aim to model the COVID19 pandemic, we choose model parameter values that are not unrealistic for COVID19 and we hope to provide some intuitive insight in the role of social distancing. As our calculations are almost analytical, they allow us to understand some of the key factors influencing the spread of a disease. We argue that social distancing is particularly powerful for diseases that have a fat tail in the number of infected persons per primary case. Our results illustrate that a "bad" feature of the COVID19 pandemic, namely that super-spreading events are important for its spread, could make it particularly sensitive to truncating the number of social contacts.