Post #1300: COVID-19 trend to 10/18/2021, holiday artifact.

Posted on October 19, 2021


Today we’re looking at the final artifact of the Columbus Day federal holiday.  Cases not reported on the holiday get reported on the next day.  If both of those days fall into the seven-day moving average “window”, that averages out, and we  get the correct new case rate.  But if only one or the other of those days falls into the seven-day “window”, we don’t get a correct estimate of the new case rate..

On the day that the holiday day itself passes into that window, the new case rate dips.  And on the day that it exits the window, the new case rate jumps. Like so:

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 10/19/2021, from”  The NY Times U.S. tracking page may be found at

Near as I have ever been able to tell, there is no way to fix that.  At least, there’s no simple way.

There’s little point to making much of today’s numbers because I can’t accurately determine what portion of today’s new case rate is just an artifact of holiday data reporting.


If I had to call it by eye, offsetting the various report-related wiggles in the lines, I’d say there’s been no change.  The Delta wave continues to recede in the South and Pacific regions.  But not in the Mountain, Midwest and Northeast regions.

Virginia new COVID-19 cases by vaccination status.

Source:  Calculated from Virginia Department of Health COVID-19 dashboard.

Finally, it’s well worthwhile to keep your eye on the ball.  The ball being vaccination status.

Above you can see the most recent week of data from Virginia.  Those proportions bounce around a bit from week to week, but the picture never looks qualitatively different from what you see above.  There’s a big gray area for new cases among the unvaccinated.  An there are small colored areas for cases among the partially or fully vaccinated.

Just to put that in perspective, the 62% of the Virginia population that is fully-vaccinated accounted for 3% of the new cases in that most recent week of data.

I can’t quite get my mind around what, exactly, that says about my risk, as a vaccinated individual.  The figures above almost certainly combine medical aspects (vaccines reduce likelihood of infection) and the behavioral aspects (the vaccinated and unvaccinated likely differ in terms of risk factors and risk-taking behavior).

When Virginia’s new case rate got down to the low single digits this summer, my wife and I (e..g) returned to going to the gym on a regular basis.  When new-case rates climbed, we stopped.

But now, based on this most recent week of data, the new case rate for the vaccinated is again down in the single digits.  Virginia shows it at about 6 new cases per 100K per day.  That’s really not materially different from what it was this summer.  (But the current rate for the unvaccinated is vastly higher than it was this summer, presumably owing to the spread of the Delta variant since that time).

On paper, all other things equal,  my risk of infection (as a vaccinated individual) would seem to be similar to what it was last summer.  But we share our public spaces, vaccinated or not.  Does this mean I should go back to behaving as I did during this summer’s low, or not?

I don’t know.  I don’t worry about getting infected.   But I don’t go out of my way to take unnecessary risk either.  I guess I’ll keep that up for a while longer yet and see whether or not we get a winter wave of COVID-19.