Post #985: Trends update: Like watching paint dry

Posted on January 30, 2021

Source:  NY Times Github COVID data repository, data reported through 1/29/2021.

Above:  National data (thick line) and six regions, seven-day moving average of new COVID-19 cases / 100,000/ day.  Top graph is in “natural units” (i.e, actual COVID case counts per 100,000 per day).   Bottom graph is in logs.  The main reason for using logs, in this case, is that a constant rate of growth graphs as a straight line.

When things get this boring, you really have to dig to find something to write about.  In this case, it’s that the rate of decline in new cases / 100,000 / day appears to be more-or-less constant.

So:  Paint is drying uniformly.  Not exactly prime click-bait, but it’s the best I could come up with.

Waiting for the log line to break downward

On the top graph, you can see how the line is beginning to curve a little bit, on the down-slope.  Just starting to look a little bowl-shaped.  As so you are tempted to say, the reduction in case counts looks like it’s slowing down.  But that would be wrong.

But as you can see from the log graph — where there is no such curve — the line, when graphed to show the growth rate, seems pretty close to a straight line.

What I’m waiting for, per the last post, is for that log line to start curving the other way.  Instead of bowl-shaped upwards, as the top line is tending toward, I’m waiting to see the bottom line going bowl-shaped downward.  That’ll mean that the pace of decline is speeding up, to the point where you can see it.

Under a stylized picture of a pandemic, if the true “herd immunity” level requires 70% of the population to be immune, that natural speedup of the rate of decline would occur slowly and far into the future.  You shouldn’t be able to see anything happening within a one-month time frame.

But if, as I claimed in prior posts, we’re already at “remission herd immunity”, the vaccination of 12-ish percent of the U.S. population, this month, beyond that level, really ought to result in a visible slowdown.  A downturn in the log graph line.

So, as the paint dries, sometime in February I’m hoping for visible evidence that it’s starting to dry faster.  That might happen just by chance, and it might happen due to the inherent seasonality of coronavirus.  But if we’ve achieved “remission herd immunity”, it should happen purely as a matter of pandemic arithmetic.

U.K. and South African strains may disrupt this neat picture.  Or maybe not.

At some point in the future — March seems to be the consensus — all bets are off on the new-case growth rate, due to the new strains of COVID.  March is when the U.K. strain is expected to take over here in the U.S.  Plus now we have a South African strain.

The majority opinion says that the more infectious nature of those new strains of COVID pretty much guarantees a fourth wave of some sort, unless we make some material changes to avert that.  Which probably should start by putting everybody in N95s or equivalent.

But there are  a couple of things in our favor.  And so, while the smart money is betting on a fourth wave, that’s not a done deal..

One factor is the seasonality of coronavirus (Post #714). (There are now better overviews than the one pictured below, but the picture gets across the gist of it correctly.)

Winter will be starting to lose its grip by March, in much of the country.  As you can see above, incidence of the typical human coronavirus (red line) is trending down by that time.  (Or, at least, it was in Stockholm, the source of data for the graphs above.)  So the trend will be our friend.

The other factor that I think is under-appreciated is how “leveraged” these additional vaccinations should be.  Again, if I’m right about “remission herd immunity”.  In terms of the math, the combination of the high number of prior infections plus all our masking/distancing/whatnot means that we have already done the equivalent of denying the virus access to >70% of the population.  As long as we keep all that up, the vaccinations will be piling on, on top of that.

In effect, it’s now time for us to benefit from the past year’s suffering.  Instead of vaccination starting from zero (as with, say, childhood vaccine for measles), it’s starting off from a base of 40%-or-so of the population already being immune.

Anyway, for February, that’s what I’m hoping to see, as the paint is drying.  Just a little curvature, in the downward direction, in logs.

At least it gives me something to do, as we wait to see when or whether the next wave will begin.