That number was the upper bound of at least one epidemiological model. According to https://www.independent.co.uk/news/world/americas/us-politics/coronavirus-de

That number was the upper bound of at least one epidemiological model. According to https://www.independent.co.uk/news/world/americas/us-politics/coronavirus-death-toll-trump-white-house-memo-february-peter-navarro-a9452761.html https://www.independent.co.uk/news/world/americas/us-politic..., the possibility was included in White House briefings. Should the media not report that fact? I'm also fairly certain all such report noted the large uncertainty of such models, especially at the beginning when almost no data was available. And I should point out that extrapolating current death rates in New York City (2%) to the entire US comes out to 600,000 deaths. If we further include that total excess deaths in NY are about twice the number of confirmed COVID deaths, and we're above 1,000,000 deaths. I don't believe we will see such numbers: NY's density and International connections are probably contributing factors that just don't apply to Idaho. But as a worst-case scenario, those 2 million are not quite as ridiculous as you seem to think.