Pulling up short of 75,000 deaths means, in other words, an incredibly abrupt conclusion to the pandemic, with deaths going all the way to zero very soon and staying there permanently. That is not going to happen. For about the last two weeks, the country has been on a roughly flat trajectory of about 2,000 deaths per day. If it stays on that plateau through August 4, it would mean not 12,000 more deaths, but 180,000. And the pandemic wouldn’t simply end on August 4 just because the modeling does.
Of course, we may well not stay on that pace, but decline. How quickly? According to a New York Times survey of five major models, published last week, all of the models project a quite rapid decline — as rapid as the ascent was. As for how quickly that decline would begin, a model based at the University of Texas, which has won praise as an alternative to the IMHE, now says with 100 percent certainty that the country has passed its peak — this despite the fact that just on Wednesday we reached a new peak, and despite the likelihood that no more than 5 percent of the country, at most, has been exposed to the disease.
So, what is happening? Why is it that nearly all efforts to project the future shape of the pandemic seem unable to see more than a week or two into the future? And why, even in that time frame, are they almost unanimous in projecting a precipitous decline that is almost every day contradicted by the number of new deaths?
There are two big explanations. The first is that even under present conditions, in which the spread of infection is being dramatically constrained by shutdowns, the disease is not behaving as we expected.
Read it all.