Agreed. That is an important point. Let epidemiologist do the modelling, and use some back-of-the-envelope calculations to convince yourself of the orders of magnitude. It's super annoying that everyone who knows what a jupyter notebook is now starts blogging about their own modelling efforts.

Atm it only takes a straight line through a couple of points to predict things for most countries, so I'd say yes.

The interesting question is whether they could help to implement useful political interventions based on which parameters matter. The answer seems to be yes in many cases.