[mantap]

It might have been correct to say there's no evidence in March (which is not a reason to not do something, we would still be in the Stone Age if every action we took required evidence.). There's plenty of evidence now as we have data for both going into and coming out of lockdown.

[SpicyLemonZest]

There's plenty of weird, contradictory evidence. Many places have come out of lockdown early, been told they're facing certain doom ("Georgia's Experiment in Human Sacrifice" [1]), and then been quietly forgotten when the predicted consequences don't come. It's hard to believe that lockdowns don't do anything at all, but I don't think anyone can honestly say we have definitive proof they were necessary.

[1] https://www.theatlantic.com/health/archive/2020/04/why-georg...

[akimball]

I can. R0~5.7 has dropped below 1.0 in many mask-averse stay-at-home regions, which I consider compelling evidence of efficacy in an adverse environment, under common-sense priors informed by the medical literature.

[SpicyLemonZest]

"This number is below 1.0" is not, by itself, an argument that some particular social policy was necessary or effective. An argument that lockdowns were necessary would at a minimum need to address the questions of "would a less strict policy have sufficed" and "will the long-term outcome after lockdowns end be different".

[etherealmachine]

Georgia has 164 deaths per million residents, versus 82 in South Carolina next door and 87 in California. Are you sure you still want to call that a bad prediction?