Post #1031: Trend to 2/25/2021, losing ground.

Posted on February 26, 2021

There’s nothing new to report.  The daily new case rate is now up a little bit, relative to what might turn out to be the end of the U.S. third wave on 2/21/2021.

Short-term bottom in new cases per day.  Since 2/21/2021, new cases / 100,000/day (seven day moving average) has begin creeping up at the rate of 9%/week.  That’s due in part to Texas’s return to normal reporting, but instances of leveling-off and increasing cases are widespread across the states.  So only a part of this can be attributed to Texas.

Exact timing match between U.S. and Canada.  Despite the fact that the U.S.-Canadian border remains closed, the timing of the inflection points in the U.S. and Canadian curves is an exact match.  They too hit their minimum new daily case count on 2/21/2021.  I just think that’s odd enough to be worth mentioning.

The U.S. COVID-19 deaths data still remains grossly out-of-sync with the U.S. new cases data — far too high, relative to new cases.  (See Post #997).  But every time I look at that excessive death rate — because, let’s face it, who couldn’t use yet another thing to worry about regarding coronavirus — I find another state that is reporting a big backlog of cases.  So I remain convinced that this is a mere data reporting issue, and not an actual change in the behavior of the virus. 

Today’s find was that Virginia is in the process of reporting large numbers of deaths.  These are deaths that occurred, in most cases, months ago.  But if you tabulate by “report date”, as we all do, they show up in the current data.  (If you look by date of death, the data tail off at the end due to deaths that have occurred but have not been reported.)  This is a little hard to find on the Virginia dashboard, and can be accessed through their “key measures” page.

And so, I assume that if that’s still going on here, that’s probably still going on elsewhere.  There might be something weird going on with all the new strains being reported.  But the simplest explanation, supported by hard data, is widespread catch-up data reporting.

Another reason to think this is merely a data reporting issue is that I don’t see the same thing happening in Canada.  Here are Canadian new cases and deaths, and U.S. new cases and deaths.  The Canadian data look as expected: The deaths count follows behind and mimics the shape of the new cases count.  The U.S. data, by contrast, don’t look as expected.  There’s been a three-quarters drop in new cases, versus maybe a one-third drop in deaths.  The two curves are practically unrelated.



I’m pretty sure it’s mostly the same virus on both sides of the border.  That again suggests that the oddly high U.S. deaths data are due to reporting issues.