> the Imperial College paper estimates 2 million dead in the US with no response and ~1 million dead with a limited response that doesn’t involve a full shutdown.

The paper points that all these estimates are based on extremely biased data.

Testing a random sample of the population to get a reliable estimate of the scale of the epidemic is dirt cheap compared to the drastic measures currently being taken.

It's incomprehensible to me that this wasn't one of first steps taken.