Switching to proportion is actually the wrong thing to do. If you look at the data from a place with high vaccination rates (such as King County), even low prob

Switching to proportion is actually the wrong thing to do. If you look at the data from a place with high vaccination rates (such as King County), even low probability of a break through case among the vaccinated will translate to high proportion. Edge case that demonstrates the fallacy of looking at the proportion is: let's say you have 100 people, 90 of them vaccinated, 10 not vaccinated. Let's then say that vaccinated have 10% of getting infected, and the unvaccinated have 50% of getting infected. You'll then get 9 vaccinated people that got sick, + 5 unvaccinated people that got sick, for an almost 66% proportion! But that doesn't change the fact that as a vaccinated person, your chances of getting infected are 5x smaller compared to the unvaccinated.