Covid-19 models and projections have been receiving criticism, usually from conservatives, for being "inaccurate" in some respect, or subject to continual revision based on new data. Critics like to harp on relatively miniscule discrepancies in order to undermine the credibility of the whole project.
Yet the models have been plenty accurate in one important respect, which happens to be the *most* important respect in terms of making public health decisions. We know it’s an insanely infectious and fairly deadly disease that multiplies exponentially if left unchecked.
As for predicting the final death toll:
1. The outbreak isn’t over yet, so we obviously can’t yet assess whether and which death predictions have been accurate.
2. The disease is brand-new, and we don’t yet fully understand it: including the very important question of how many people have already been infected That we don’t know about. Because we don’t yet have antibody tests up and running for everyone. That information is necessary to arrive at an accurate “death rate.”
3. The future progression of the disease depends upon actions governments and individuals take in the future, which are out of the hands of public health officials.
4. We don’t know if and when a real cure/vaccine will be found and scaled into production.
A few reasons for very wide confidence intervals in these projections.
tl;dr: Some want science to give us all the answers right away, but unfortunately, science doesn’t work like that.
