Source: Calculated from NY Times Github COVID data repository, data reported through 2/4/2021.
New COVID-19 cases per 100/000 / day are just shy of half the level they were at the 1/8/2021 peak.
I can’t tell you how much I’d like to make a big deal out of the circled area on the second (log-scale) graph. But I can’t.
The rate of decline starting 2/1/2021 is -23%/week, modestly steeper than the decline 1/11/2021 – 1/31/2021 of -17% per week. That’s about the right magnitude, and plausibly the right timing, to be the increasing rate of decline that I expect to see per the algebra of my last post. Dump a bunch of immune individuals on top of a pandemic that is already shrinking, it should shrink noticeably faster.
But I can’t say that. That conclusion is not confirmed by the state-level data. The U.S. Midwest is the poster child for herd immunity, and it’s not happening there. Instead, this sharper decline is more-or-less a fluke. It’s due to a true sharper decline in New York, the impact of prior data reporting issues now washing out of the system, and (probably) just random variability of the data.
See log-scale graph below for complete absence of increased rate of decline in the Midwest. Instead, those states show an almost-amazingly constant and consistent rate of decline (most lines run straight and parallel, few lines cross.) That argues much more strongly for seasonality of COVID being the main driver of this decline.