Modelling is a precious source of information in science. With models, we can simplify an otherwise messy reality in order to understand the fundamental driving forces of a system, like an epidemic, and we can try to predict the course of events in complex scenarios where there is a great degree of uncertainty. In short, models can be used to explain and predict phenomena. Yet models interact with expert opinions in two fundamental ways. They are sometimes in competition with expert opinion, and they are sometimes heavily dependent, for their proper working, on expert opinion.

In this paper I will illustrate the different ways in which a model interacts with expert opinion. I will focus on epidemiological models. I will explain how, in epidemic modelling, getting the expertise right is as important as getting the model right. I will briefly present epidemiological models with a focus on the specific contribution of expert judgment to the choice and use of these models. I will compare expert judgment with statistical judgment, highlighting the limits of the former. I will analyse the interconnectedness of modelling and expert judgment in epidemic simulations based on a case report and, finally, I will suggest some strategies for ameliorating the interaction between modelling and expert judgment.