re: Coronavirus has now killed more Americans in 1 month than flu killed in entire year

quote:

---

They keep doing antibody tests and keep finding 30%+ people have had the fricking thing.

---

They don’t “keep finding” anything close to that.

And the antibody tests have a high false positive probability (due to the base rate), and the one used in the study linked is by far the most inaccurate test I’ve seen (about 10% positive and negative error rates).

And none of the studies have any generalizability to even their local populations (e.g., not random; recruit people more likely to infected; volunteers are even more likely to be infected than those recruited). And that study was done in an area that has a case rate 10 times the national average anyways, using a small sample.