Post #1179: COVID-19 trends, rounding out the week

Posted on July 17, 2021


I’ve tracked the trend in daily new COVID-19 cases every day for the past week.  Today I might as well finish out the data week, even though the story remains unchanged.  There won’t be any new information on the trend in new cases until next Tuesday, as most states don’t report that information over the weekend.

U.S. total daily new COVID-19 cases rose 68% over the past seven days.  Judging from how erratic the data reporting has gotten, I don’t think that’s materially different from the 75% increase for the seven days ending yesterday.

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 7/17/2021, from”  The NY Times U.S. tracking page may be found at

John Q. Public remains oblivious.

The U.S. public doesn’t seem to be too worried about this.  At least, not enough to affect behavior.  We can turn to the Carnegie-Mellon CovidCast website, where they use a large-scale survey of Facebook uses to track some COVID-19 related measures.

Here’s a map showing mask use rates as of July 1, July 7, and July 15, 2021.  Darker = more, paler = less.  The map is almost uniformly paler, showing that use of masks continues to decline, even as this wave gathers steam..  Mask use hasn’t picked up even in the states significant outbreaks.  Let alone as a preventative measure in the states where new case rates are still low.

Source:  Carnegie-Mellon CovidCast website

Here’s a map of the fraction of people who went to a bar two Saturdays ago, and last Saturday. Bars are among the riskiest indoor locations, and typically were the first sites shut down and last sites re-opened during various lockdowns.  Darker is more.  Looks like bar business picked up a bit last week, compared to the week before.

Source:  Carnegie-Mellon CovidCast website

Driving factors

At this point, this is obviously attributable mostly to the rise of the Delta variant.  The greater infectiousness of that variant has to be the main reason behind the unprecedentedly rapid increase in cases.

In addition, I think seasonality is playing some role.  Last year, at this time, we managed to have outbreaks in several hot-climate states, at a time when Delta didn’t even exist.  That summer 2021 outbreak was attributed to more time spent in dry, indoor air-conditioned spaces.

What I think I’m seeing, as I look at the data, is that the combination of high Delta prevalence in a hot-climate state leads to a particularly high rate of growth.  Using the most recent update from CDC on incidence of Delta (this table, accessed 7/17/2021), I can compare the top eight states from last year’s outbreak (first chart below) to the top eight states in terms of the CDC’s estimate of prevalence of Delta as of about a month ago (second chart below).

If anything, I’d say the first chart (hot climate) looks at least as bad as the second one (high Delta).  But maybe I’m just seeing what I want to see.  And at this point, if I picked any random set of eight states it might look just like that.

Finally, conventional wisdom says that the worst Delta outbreaks have been occurring in areas with low vaccination rates.  That would certainly make sense, as this is almost entirely a wave of infections among the un-vaccinated.  Here’s a contrast of the states with highest and lowest vaccination rates.

While the absolute level of cases is far lower in the high-vaccination states, it’s not clear by eye that the rate of growth (slopes of the lines) in this last week has been any lower.  Best guess, even in the states with high vaccination rates, this more-infectious variant has no problem in spreading among the un-vaccinated.

Restated, while this wave is definitely due to the un-vaccinated, it’s infectious enough that there is no shortage of targets, even in states with high vaccination rates.

On a more personal note

Post #1163 is rescinded.  My wife and I talked it over.  Even though the measured new-case rate here is quite low, and we’re vaccinated, working out in a gym, un-masked, doesn’t seem prudent right now.

As of June 23, my best guess is that one trip to the gym in every 7.5 years would pose a significant risk of infection.  But now, if I just take current Virginia prevalence and account for data lags, keeping growth at the current level, the estimate is down to a little under once a year.  Or to put that differently, if we keep going to the gym, under current conditions, sometime this year, we’d expect to get a good, solid dose of the virus.

More to the point, if conditions persist, that mean-time-to-infection-event interval shrinks.  Two weeks from now, if the growth of new cases continues at the current pace, that will be down to one significant exposure event every four months.   Seems like it’s better to quit while we’re ahead.

A final factor is that his rec center is the site of several County-run day camps.  As a result, the facility has a lot of little kids running around in it.  Those kids are by definition un-vaccinated. The risk of exposure from them isn’t really quantifiable, but given the number of kids in the building, it’s probably non-negligible.

So we’re back to walking around the block and calling that exercise.  For now, at least.