[jfnixon]

I think this is backwards. We have been using infection rates as a proxy for death rates, since deaths are a lagging indicator. It is certain we are missing a huge percentage of cases, and expanding testing will uncover those. It is completely unclear that uncovering more mild and asymptomatic cases will result in more deaths, so new cases are a bad proxy for future deaths.

[pmiller2]

Maybe in an ideal world, but we're not actually measuring infection rates. Dead is pretty objective, and easy to measure, and, we know that the number of deaths will roughly track the number of infections. That's why things like deaths and hospitalizations are what the state of California is watching closely.