Post #1015: Trends to 2-16-2021

Posted on February 17, 2021

Source:  All graphs are calculated from NY Times Github COVID data repository, data reported through 2/16/2021.

We’re now down 68% from the peak.  Just 32% left to go.



More interesting to me, the rate of decline seemed to pick up on Valentine’s day.  The tail-end of the blue line seems to have a slightly steeper slope.  I thought that was just a data reporting artifact, but in so far as one can tell based on two days’ data, that now appears to be a real decline.



So let me do a little wishful thinking here, and sketch this out as if it were the start of a curve with ever-increasing downward slope.

This is the same log-scale curve above, but starting on January 1, 2021.  By eye, it looks like there are two modest inflection points (places where the line bends).  That divides the curve into three pieces — January, first half of February, and now the second half of February.

Let me remind you of two key reasons why I keep harping on this, and I keep looking for this.

First, the simple arithmetic says that if you use vaccination to force-feed immune people into a stable pandemic situation, then the pandemic should accelerate to an end.  It shouldn’t just slowly peter out, it should nose-dive.  I first laid that out in Post #984.

Unfortunately, there is still nothing in the cross-section of states to suggest that this is correct.  If it’s true, we ought to see that effect first in the states with the highest fraction of the population immune to COVID.  And so far, we’ve seen nothing of the sort.




The second is that if we see a sufficiently swift rate of decline of new cases with the existing variant of COVID-19, then we’ll be robust to the switch to the new, more-infectious U.K. variant.  My estimate, based on the U.K. variant being “40% more transmissible”, is that we’d need to see about a 45%/week rate of decline in new cases soon (before mid-March, say), in order to avoid a fourth wave of COVID-19.  That’s laid out in Post #1000.  If the current rate of decline continues, that seems a lot more feasible then it did just one week ago.

Finally, it’s T-3 and counting for the post-Super-Bowl explosion of cases.  Based on my estimate of 12 days’ lag, those cases — if they exist — will start creeping into the daily case counts by the end of this week.