Post #1210: COVID-19 trend to 8/11/2021

Posted on August 12, 2021

The U.S. now averages 38.4 new COVID-19 cases per 100,000 population per day.

Measured in percentage terms, the growth in new cases per day is slowing.  On the log-scale graph below, the line is clearly starting to curve downward, to the right.  Today, the week-on-week increase is just under 30%.

Data source for this and other graphs of new case counts:  Calculated from The New York Times. (2021). Coronavirus (Covid-19) Data in the United States. Retrieved 8/12/2021, from”  The NY Times U.S. tracking page may be found at

But in terms of the actual number of cases (not the percent increase), the growth rate remains fairly steady.  On the graph below — which shows cases, not the log of cases — you get a bunch of straight lines off to the right.

Starting around August first, the U.S. new-case rate has risen by about 8 additional new cases per week.  (E.g., it’s 38.4 now, and based on that growth, you’d expect to see 46.4 a week from now).

If that rate of growth holds, it will take just over four weeks to match the U.S. peak new case rate in the third (wintertime) U.S. wave of COVID-19.

(Nothing about the theory of epidemics suggests that we should see “arithmetic” growth of this nature.  I’m just saying that, empirically, that’s what we’ve been seeing of late. )

Today Louisiana is at 121 cases / 100K/ day, Florida is just over 104 cases / 100K / day.

Only ten states, total, have new case rates above 50/100K/day, and all of them are in the South / South Central regions of the country.  (As you probably could have guessed, from yesterday’s map).

By eye, it’s possible that Utah and Nevada are peaking, and so would make three states total (MO, UT, NV) that appear to have peaked for this wave.  Maybe “stabilized” might be a better word, as, at best, these are very broad peaks.  Too soon to tell, really.

The only thing that makes it worth bringing up is that these reinforce the notion of nine weeks to peak.  These all started their wave about the same time as Missouri, and all three states appear to be reaching a peak about nine weeks after the start of the wave.  I have argued before that, purely as a matter of observation, these waves tend to take about the same time in every state, they just start and end at different times.

Ah, for grins, let’s put Louisiana and Florida on the same graph.

Conveniently, the fifth wave in FL and LA started three weeks later than in MO, NV, UT.  Invoking the 9-week rule, I can make a quick-and-dirty prediction that both Louisiana and Florida should peak on or around September 1.

We’ll see how that one stands up.

(As I earnestly engage in this mumbo-jumbo, I have to remind myself that I once held down a responsible job requiring a fair degree of scientific rigor.  (As opposed the more-or-less mindless pattern-matching that I’m doing here.)  At the minimum, I should do this for all 50 states and publish that. )

But absolute minimum due diligence requires noting that the start of U.S. wave dates to just after Memorial Day.  If I had to pick a date by eye, I’d pick 7/5/2021.  Which means the U.S. peak should maybe a week later yet, or sometime in the second week of September.

Free advice is worth what you pay for it.  But I would plan on still being in a situation of rising new case rates, for another few weeks, as we send our kids back to school.

What’s up across the pond?

I need to do one more thing today, which is to check in on Great Britain.  They were ahead of us on their own Delta wave.  So if I’m in the business of making half-baked forecasts, surely I need to see how they are getting along.

First, FWIW, depending on exactly when you date the start of their wave, the British Delta wave ran for just about nine weeks.  Somewhere around May 20 for the start, running up to a clear, sharp peak on July 20.

And then, right at the peak, Britain removed all existing restrictions on social gatherings.  July 19 was their “freedom day”.  On that date, across the U.K, the horrified screams of British epidemiologists were drowned out by the drunken cheering of pub patrons.

For two weeks, cases fell precipitously.  The epidemiologists seemed to have egg on their collective faces.  Then a funny thing happened.  Almost exactly two weeks after “freedom day”, new infection rates began to rise again.  The peak was July 20, the trough was August 3, two weeks later.

I’m far from clear on the timing of testing and reporting of new cases in the U.K.  But in the U.S., that two-week lag would be a pretty good fit for the lag between infection and the final reporting of the data.  In other words, if I had to guess, I’d guess that the current rise in infections literally began with people who were infected on British “freedom day”.

In effect, Great Britain coasted downhill for two weeks, reaping the last benefits of the lockdown.  And when those two weeks were up, they began to see the impact of removing all of their restrictions on social gatherings.

I might say something along the lines of “we should learn a lesson from that”.  But as far as I can tell, learning a lesson is no longer part of the American mindset.  Not even from our own hard knocks, let alone someone else’s.  Whatever we may say about Great Britain’s recent COVID-19 experience, at least they’ve experienced a peak in their Delta wave.  We’re still moseying toward ours.