Does the simulation timestep matter? (for Covid simulations we naturally pick days, and this works well. But obviously “years” would be bad. What’s the right timestep for weather simulations? Climate? Stock market?)

Can you argue that this is a garbage-in-garbage-out simulation, since the parameters are based on human behavior? What does that say about all of the other Covid simulations we read about in the news?

How would you develop trust in a particular disease-spread model?