A major explanatory factor for COVID death rates should be population-weighted population density. This is difficult to calculate but since the start of covid, I have found two sources that claim to calculate this - an academic source here by country and by state and almost by accident a British Columbian professor by state. The BC professor dataset provides better explanatory power for states, so that is displayed below. By country, I only had access to the academic source. It would be interesting to find out why there is a difference between the data in these two providers.

The explanatory power of population-weighted population density for death rates is around 30% - that is we can explain 30% of the variation in death rates with this one variable, making it likely the most important of the non-behavioral variables.

A look at US states shows why this is a useful way to look at the covid performance. States like NY and NJ had extremely high death rates but that is expected because of the density of the population. States like Wyoming and Alaska are on the other end of the spectrum.

US States