"Received: 22 March 2020" - direct from the study cover page at https://www.nature.com/articles/s41586-020-2404-8_reference....

States imposed lockdowns between March 21 and early April. Someone want to explain to me how a study submitted before the lockdowns were issued (nevermind had time to actually have an impact) "found" those results?

It must be a typo. March 22nd was a Sunday, where May 22nd was a Friday. It seems unlikely that it would take over 2 months to accept it, considering so many other articles are being rushed to publication in fractions of that time.

ETA: also, charts in that publication also show data collected from April, so unlikely they had data from the future.

Study suggests, not finds. Simulation models can't "find" facts. Observation finds facts.

Didn't the study of the simulation model find facts about the simulation model?

So far.

Exactly. I believe the 60 million infections delta was estimated for April 6th (figure 4), so the counterfactual would be considerably worse as of today.

"So far" meaning "might save more later" or "only postponed, not prevented"?

By now we have learned that masks + distancing is enough to stop this (Drive R0 below 1, it will die out in time.) If we were serious about doing these things the cases--and deaths--would be prevented, not merely postponed.

The latter. That's always what "flattening the curve" has meant. The total area under the curve (i.e. total unique infected individuals) stays about the same.

"Flattening the curve" is not the only possible outcome.

A cure, an inoculation, or segmenting geographic regions until localized herd immunity (like New Zealand just did) are all examples of cutting the curve short.

Vaccines are far away from approval for widespread use, and New Zealand's situation as an island nation is not applicable to something like 98% of the World's population and quite frankly, irrelevant to the discussion.

Simulacrum.