Post #1333: COVID-19 trend post-Thanksgiving

Today we get an initial look at the post-Thanksgiving trend in daily new COVID-19 cases.  So far, this seems to be shaping up for a much milder winter wave than last year.

And, as a bonus, I’ll add that if Omicron follows the path set by Delta, it’ll be too late to contribute to this year’s winter wave.  If the U.S. sees its first few cases of omicron today, and if omicron spreads at about the same rate that delta spread, it will be mid-March 2022 before Omicron is the dominant strain in the U.S.

Lot of “ifs” in that last sentence, but that’s the best estimate I can come up with at the moment. Continue reading Post #1333: COVID-19 trend post-Thanksgiving

Post #1332: More transmissible than the Delta variant?

You read a lot of hype every time a new variant of COVID-19 is discovered.

Stories on COVID-19 variants are a nearly-ideal source of click-bait.  They combine a high fear content, a broad target audience, and a near-infinite repeatability.  The virus has an almost unlimited ability to spawn variants, and each one has the potential to do any number of scary things (avoid your immune system, avoid current vaccines, spread rapidly, and so on).  Report on some scientist talking about that potential, and you’re done.  Eyeballs captured.

These days, stories on COVID-19 variants rank right up there with school bus crashes and mass-murder events involving children as the lowest common denominator of fear-based journalism.

And yet … well, yeah, Delta.

And Alpha.  Everybody forgets about Alpha (a.k.a. the British variant) because it got its butt kicked so swiftly by vaccination.  And Delta.  But in its day, it was a threat to be reckoned with.  If the virus had stopped at Alpha, we’d be done by now.

Press reporting on COVID-19 variants is one of those ugly situations where almost all the information is irrelevant.  But some isn’t.  How do you filter out the signal from the noise?

Short answer:  Transmissibility, a.k.a. “R-nought”, a measure of how readiliy the mutated virus can spread.  That’s the minimum screen for filtering out the noise from the potentially relevant.

Why?  The way that a new variant takes over is by “out-competing” its rivals.  For that, it doesn’t matter how deadly it is, or how sick it makes you.  All that matters is that it can spread more effectively.  If it can, it will eventually become the dominant strain.  If it can’t, it won’t.

The reason the Alpha (British) strain took over from the original (Wuhan or native) strain was that it was more transmissible.  The reason Delta took over from Alpha was the same.

Like so, from this seemingly-plausible reference, where R0 = “R-nought”:

Professor Cheng said last year's Wuhan strain had an R0 value of around 2.5, the Alpha strain was about 3.75 and the Delta strain was about 5.

That means if we were living life like we were in 2019, one person with the Delta strain would likely infect five other people, compared to just 2.5 last year.

As a result, until I see those words — “more transmissible” — I just ignore it.  Because anything that’s not more transmissible than Delta is not going to displace Delta as the dominant strain.  No matter what else it does.

As a result, I rarely talk about new variants.  The last time I talked about a new variant was back on June 15 2021.  (Post #1160: COVID-19 Delta variant bodes for a 5th U.S. COVID-19 wave. )

This post puts a marker down for the latest variant out of South Africa, currently termed B.1.1.529.  This one ticks all my boxes for indicators of being  “more transmissible” than the Delta variant.  Per popular press reporting, a) it’s spreading rapidly in South Africa, b) the South Africans have ruled out mere chance (e.g., a few super-spreader clusters), c) it’s rapidly displacing Delta as the dominant strain, where it exists, d) it’s spreading rapidly in other countries where it has already spread to, and e) other countries (Israel, Great Britain) are already taking steps to try to contain it or slow its spread.

Anything else you may read about it seems to be speculation, at this point. It may evade current vaccines, evade the immune system, blah blah blah.  Sure, it may.  And it may not.  Nobody knows yet.

There isn’t even an estimate of it’s native transmissibility (its “R-nought”) yet.  So nobody has even done the analysis to take a guess as to where it ranks relative to prior strains.

We’re still at that stage where the reporting is confused, and the likely future of this new strain is uncertain.  All I can say is that after reading through the reporting this morning, this one seems to be worth watching.  Despite the lack of any hard estimate of transmissibility.  Despite the small number of cases so far.  As with the Delta variant, this one — likely to be given the Greek letter omicron as its designation — is worth keeping your eye on.

Post #1331: COVID-19 trend through 11/24/2021


This is the last look we’ll have at the trend until next Tuesday, owing to the combination of the Thanksgiving holiday and the subsequent weekend.

As of now, the story remains the same:  So far, it’s a slow and mild winter wave for the U.S. as a whole, compared to last year.

For me, beyond that, the only thing to have changed is the reduced rhetoric about massive increases in air travel and the inevitable “surge” (a.k.a., “explosion”) in post-holiday cases.  That seems largely lacking this year, compared to last.

Plausibly, our experts have learned from last year.  Equally plausibly, nobody cares any more.  Either way, as with (e.g.) the monthly count of U.S. military personnel killed in Afghanistan, the lack of that in the news is something to be thankful for today.

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 11/25/2021, from”  The NY Times U.S. tracking page may be found at

If we were to compare the regions in turn, only the Northeast is on the same track as last year:

Every other region is on some version of “not as fast”, “not as soon”, or “too early to tell”.

Finally, it looks like holiday air travel has very nearly returned to the pre-pandemic level.  Below, gray = 2019, blue = 2021.  The chart as aligned so that the day before Thanksgiving is the last day for each year shown.

Prior to Thanksgiving, it looks like air travel was running about half-a-million passengers shy of the 2019 level.  But the large uptick in Thanksgiving air travel this year (end of the blue line) brings this past week’s air traffic quite close to the 2019 level.

Source:  Calculated from the U.S. Transportation Safety Administration website.

Post #1330: William and Mary COVID-19 trend through Thanksgiving break


There’s still a low but persistent rate of new COVID-19 cases each week among William and Mary students.  It’s under one new case per day, on average.

The last entry below is through the Wednesday before Thanksgiving.

Source:  Calculated from the William and Mary COVID-19 dashboard.

Post #1329: COVID-19 trend to 11/23/2021, slow winter wave

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 11/24/2021, from”  The NY Times U.S. tracking page may be found at

Case counts are rising slowly, and are rising in ever region except the Pacific states.

Tomorrow we begin our holiday season.  I’ve set up the second graph below so that I can track case counts relative to what happened during last year’s holidays.  Thanksgiving is one day earlier this year.

Up to now, I’ve been characterizing this year’s winter wave as “late”, compared to last year, likely owing to warm October weather in much of the country.  Best guess, based on mean October temperatures for the U.S., and the typical Fall month-to-month rate of temperature decrease, maybe two or three weeks late.

That may be true, but after this past week of slow growth, that’s starting to wear a little thin / stretch a little far / pick your own metaphor, as a complete description of what’s happening.

If the 2021 winter wave were a duplicate of 2020, just a couple of weeks later, we’d see the orange line above eventually paralleling the blue line.  Instead, ,that gap between last year’s line and this year’s line continues to grow.

Restated, either this year’s winter wave is much later than last year’s, or we have a generally lower rate of growth in cases this year, compared to last year.

We also have a more-or-less a complete lack of a winter wave in three states that were leaders of the winter wave in the Mountain region last year (ID, MT, WY).

Now that I look at it, the three states that led the Midwest region in last year’s winter wave have also mostly failed to get with the program this year.  Shown below.  It’s not as clear-cut as the leaders of the Mountain states — there is a slight upward trend in two of the three — but it is nevertheless a strong contrast to last year’s experience.

I’m not sure what to make of all that, yet, but given that it’s the holidays, I’ll try to end on an upbeat note.

Maybe we are due for a repeat of last year’s wave.  The same as much of Europe is suffering through right now. Just a bit later.

On the other hand, owing to our exceptionally high rate of infection in the past, and to a middle-of-the-road vaccination rate, maybe immunity in the population is now high enough to allow is to avoid another awful winter wave.

(Awful?  People forget.  For nearly the entire month of January 2021, we had an average of more than 3000 COVID-19 deaths per day.  As a result, for that period, the total U.S. death rate increased by more than one-third above the norm.)

Near as I can tell, we ought to be pretty close to “herd immunity”, for something as infectious as the Delta variant of COVID-19.  That’s defined as having enough immunity in the population (via infection or vaccination) to be able to suppress the spread of a virus with basic reproduction rate (R-nought) of 5.0.  (Where the average infected person would go on to infect five others, absent any interventions or immunity).  The table below shows 77%, where we’d need a value of 80% or more to suppress spread of a such a virus with no other measures taken (e.g., without wearing masks).

I have to make a lot of assumptions in that calculation, as it depends on a lot of things that cannot be observed.  For example, it doesn’t account well for the rate at which immunity fades over time.  It doesn’t account for the potential for re-infections (Post #1326).  And so on.

But I think this is enough to show that we’re in the ballpark.  Right now, infectiousness of Delta is rising owing to changing weather conditions.  So it would not be a surprise to have some winter wave.  But we ought to be close enough to herd immunity now that maybe we won’t have much of a winter wave.

Let me put it this way.  Let’s assume Spring is going to occur at roughly the same time this year as it did last year.  At the minimum, the heart of this year’s winter wave will be shorter than last year, reducing the total he total time we’ll spend at those very high winter infection rates.  And maybe we have enough immunity that the peak of this year’s winter wave will be lower as well.  Further reducing overall impact.

No matter how I slice it, it seems like we’re in for a “smaller” winter wave this year.  Either shorter in duration, or shorter in height, or, ideally, both.

Maybe we can take some inspiration from our northern neighbor.  For sure, if cold weather is the issue, they’ve got that covered.  And, while they have a much higher overall vaccination rate, but we beat them in terms of cumulative infections per capita.  So we’re not that different in terms of total population immunity.

And yet, compared to last year, so far, there’s not much of a winter wave happening there. Whatever it is that’s suppressing the start of the U.S. winter wave, it apparently is not limited to the U.S.




Post #1328: COVID-19 trend to 11/22/2021


We’ve reached the part of the year where we’re only going to get glimpses of the actual U.S. trend in new COVID-19 cases.  Between now and the second week of January, we will observe a mix of:

  • the true long-term trend of new case counts,
  • the lack of data reporting on weekends,
  • the even greater lack of data reporting on holidays,
  • reduction in test-seeking (and case counts) over holidays,
  • any actual impact of the holidays on new COVID-19 infections.

Today is a case in point.  When I left this last week (Post #1326), there seemed to be a sharp uptick in cases.  Now I can see that was temporary.

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 11/23/2021, from”  The NY Times U.S. tracking page may be found at

The trend is still up, but at a much slower rate than appeared to be the case last week.

And that’s a good way to illustrate the potential data-reporting impacts of the holidays.  Holidays?  There were no holidays last week.  How could holidays have possibly affected this apparent slowdown in new case growth?

In part, the trend reflects a true slowdown in case growth, in states such as Michigan and many of the New England states.  But it also appears slower because I goofed:  I forgot about Veterans’ Day (11/11).   The non-reporting of cases for the Veterans’ day holiday, combined with the use of a seven-day moving average, provides a boost to the apparent growth rate on 11/18.  That’s the date on which the under-reported Veterans’ Day case count finally moves outside of the seven-day window, leaving only the additional cases that were reported on the day following Veteran’s Day remaining within the seven-day window.

If you failed to follow that last bit in every detail, that’s fine.  Suffice it to say that holidays scramble the numbers.  We’re entering the holiday season.  And so the numbers are going to be scrambled for the next few weeks.

We’re in the same situation as last year, absent the hindsight that allowed us to fill in the underlying trend after-the-fact.  Below is the graph of the U.S. new case rate, from last year’s holiday season.  Imagine trying to guess what the actual trend would be, at each step of the way.  That’s where we are for the coming holidays.  The two large “dips” in the line below — known only after the fact — are the large data reporting artifacts of Thanksgiving and Christmas/New Year’s.

The 2021 winter wave and popular press reporting.

First, I want to continue to highlight the behavior of ID, MT, WY.  These were leaders in the 2020 U.S. COVID-19 winter wave.  But this year, they appear to have peaked at the end of the summer Delta wave.  As yet, there’s no indication that they’ll have a winter wave this year.  I don’t know whether that’s a harbinger for the rest of the country, or whether they’re just a little late getting started.  Or whether it’s just something unique to those states, this year.

The South seems to be split along temperature lines.  States bordering on the Mid-Atlantic region and Midwest are showing slight upticks.  But south of Virginia on the Atlantic Coast, right on through the Gulf Coast, there’s no apparent increase in cases.  (This isn’t very different from last year, where those regions had late peaks of the winter wave.)

Finally, here’s the U.S., comparing the first and second pandemic years.  Aside from being a couple of weeks late, and missing part of the Mountain states, we still seem to be on track for a winter wave.

If you’ve stuck with this post, this far, you’ll probably understand why this graph above — the one that compares this year to the same time last year — is the only one worth looking at for the next six weeks or so.  The hope is that this year’s holiday “artifacts” are about the same as last year’s.  Which means that the only hope of making sense of this year is to compare it to the same period last year.

I’m guessing that most popular press reporting isn’t going to bother to do that.  But if you don’t see the comparison to last year, you really can’t make head-or-tail out of what the most recent trend has been.

For example, we should now expect to see a substantial decline in the new case counts, owing to the Thanksgiving holiday.  (Which never generated a “surge” in cases, despite what you may have read to the contrary (Post #1324, Winter Wave Buzzword Bingo)).

By itself, that decline is meaningless.  Only in the context of last year can we start making some judgments about where the trend is actually heading.  Look for the popular press reporting that does that, and you’ll have put your finger on the people who understand how to make sense of the numbers.  Ignore the rest.

Post #1326: COVID-19 winter wave gathers speed, and Missouri data on reinfections


Nobody cares about methodology.  That’s why I no longer bother to mention it when states make large one-day corrections to their data.  I undo whatever the state did, then get on with plotting the data as consistently as possible over time.

But today’s data boo-boo is  worth mentioning.  Missouri added about 6500 cases of re-infection to their counts.  There’s a formal definition of re-infection from the CDC, which says that a COVID-19 re-infection occurs when a person tests positive (via PCR or antigen test) at least 90 days after a previous positive test.  Restated, a positive COVID-19 test, more than 90 days after a prior positive test, counts as a re-infection.

This is noteworthy because it’s the first state I’ve seen that tracks the re-infection rate.  (It’s not as if I’ve looked for it.)  Without getting caught up in the exact details, Missouri is currently showing that about 1.5 percent of all COVID-19 infections in that state are re-infections.  (That is, re-infections divided by the sum of confirmed and probable cases = ~0.015.)

Source:  Missouri Department of Health and Senior Services.

There are a few caveats.

First, I wonder about the potential false-positive rate in that re-infection estimate, but I have no hard data on that.  Fragments of COVID-19 viral RNA and surface proteins can linger after a person has recovered and is no longer infectious.  (Standard PCR tests do not distinguish these “dead” viral fragments from virus that is still capable of reproducing.)  That’s why you are specifically directed NOT to get re-tested within 90 days of a confirmed infection, and why all institutions end quarantine based on a specified number of symptom-free days, and not based on a re-test (e.g., purely at random, here are directions from U.KY., Yale).

Second, I’m not sure you can accurately assess your odds of re-infection, based on the data above.  (That is, the odds of getting COVID-19 again, once you’ve already had it).  Among other things, I think you’d need to know what fraction of the entire population has already had one infection.

I bring up that second point because there’s a clear corollary:  Anything you might have read about low re-infection rates, early in the pandemic, is probably now out-of-date.  I mean, if few people had been infected at that point, then pretty much by definition you were going to see very few re-infections.  But now, it’s not so clear.  At any rate, that 1.5% above seems vastly higher than the fractional percents that appeared in the earlier research literature.

Without over-thinking it, I am a little surprised that the re-infection rate is that high.

But that’s only due to my ignorance.  As it turns out, re-infection by respiratory viruses is common.  We know that immunity fades over time.  I’ve certainly had flu more than once in my life.  But there are multiple strains of flu, so I’m not sure I’ve been “re-infected” with the same strain.  By contrast, researchers have looked for same-strain reinfections, and have found that a non-negligible fraction of people  get re-infected with more-or-less the exact same strain of flu on different occasions.  Apparently, it’s not even all that rare.

That said, the CDC continues to classify COVID-19 reinfection as “rare”.  But it’s not clear whether that’s merely because your risk of re-exposure is low (there just aren’t that many cases per 100,000 per day).  Or whether your likelihood of re-infection conditional upon re-exposure, is low (whether you really can’t catch it again, even if re-exposed).

So we can file this one away with all the other aspects of COVID-19 that I somehow thought must have been unique to this disease.  But that were in fact common across many viral respiratory diseases.  Things such as asymptomatic infections, and long-distance aerosol spread of disease, and so on.  And now, as it turns out, for your entire life, you were at risk of getting re-infected with more-or-less the exact same strain of flu virus.  Who knew?

Winter wave picks up speed.

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 11/18/2021, from”  The NY Times U.S. tracking page may be found at

Do I need to belabor this?  The only news is that the rate of increase seems to be accelerating.

The U.S. new case rate is up 29% over the past seven days. 

Three regions (Midwest, Mountain, Northeast) are already well over their peaks from the summer Delta wave.

In the case of the Northeast, there are now three states that seem to be “going vertical” in terms of new case growth.

The only clear ray of light is what I discussed yesterday, the trio of ID, MT, WY in the mountain states.  Historically, they ought to be seeing increasing new case rates.  But instead, they seem to have peaked a few weeks ago, at the end of their summer Delta wave, and have been heading down since.

All of this, a bit over two weeks after the unusual summer warmth faded in October, and cold fall weather settled over much of the northern U.S. as of the first of November.

So, you look at Europe, you look at our mediocre vaccination rate, you look at the change in the weather, and I guess that a winter wave isn’t a surprise.

Problem is, we’re all sitting here, fingers crossed, hoping it won’t be too bad. All the while NOT looking at Japan, China, Korea, where somehow, for the second winter in a row, those high-mask-use countries seem to be sidestepping a significant winter wave.

While new research comes out — in this case, a meta-analysis or review of prior individual studies — concluding that:

Mask-wearing is the single most effective public health measure at tackling Covid, reducing incidence by 53%, the first global study of its kind shows.

And yet America remains dumb as a post.  A bare-faced post.

Worse, thanks to Republican politicization of masks, the facts no longer matter.  It has long passed the point where the actual efficacy of masks or vaccines, or lack thereof, has any relevance.  We’re stuck with a significant fraction of the population that ain’t gonna do neither, and you can’t make them.  So nyah.

I wear a high-filtration (N95, or maybe KF94) mask.  It’s not as if one lacks for good choices (Post #1246).  I’ve gotten my vaccines and booster.  I’m getting ready to fire up my humidifiers (Post #895).   I’m staying out of situations that I perceive as high-risk public indoor spaces, and certainly keeping my distance from the dumbasses who can’t be bothered to mask up in public.

I encourage my friends to do the same.

And I hope that with this winter 2021 wave, we finally have enough of the remaining population get infected so that we achieve something approaching herd immunity.  And COVID-19 can finally fade into the background in the U.S.A.

End of screed.

The crazy thing is, near as I can tell, we should be getting close enough to herd immunity that one more broad wave of infections should push us over the line.  Here’s my estimate of where we stand, now, where the magic herd immunity number is 80%, given how this is calculated.  Best guess, we’re somewhere around 77%.


Even accounting for the numerous uncertainties and assumptions that have to be made to produce this estimate, my point is, it’s not beyond reach.  One more wave, like last year’s winter wave, and we should be there.  Or in the neighborhood of there.

Meanwhile, vaccination (including booster shots) continues apace.  And every day, more people acquire immunity via infection.  We’re slowly reducing the fraction of the population (in red below) with no immunity to COVID-19.  If we can push that segment below 20%, then, in theory, the pandemic will collapse for lack of fresh disease carriers.  And while it will almost certainly not disappear, we should have long stretches of time where the circulating rate is low enough for most of us to ignore.

To me, it just looks like it’s going to take one more winter wave to do that.  And at this point, that winter wave is looking like a near-certainty.

Post #1325: COVID-19 winter wave, Mountain states

Map courtesy of

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 11/18/2021, from”  The NY Times U.S. tracking page may be found at

The U.S. winter wave continues apace, except in the Mountain region.  As you can see, the northernmost mountain states now form sort of a “hole” in the northern tier of U.S. non-Pacific-coast states.


Last year’s winter wave was led by the Midwest and Mountain regions, both of which peaked sometime around Thanksgiving.  This year, by contrast, while the Midwest is now rising rapidly, the Mountain region seems to have stalled (circled above).

Within the Mountain region, instead of any uniform trend, it’s a mixed bag.  As shown below, three states appear to have peaked.  Those are the same states that are circled on the first map graph above.

The three northernmost Mountain states are now showing a downward trend.  Based on my counts, it doesn’t look like they’ve hit herd immunity.  So what’s going on?

Turns out, it looks like they’ve had their winter wave.  Tough to say why, exactly, but it looks like the end of their summer Delta wave simply ran into and effectively became their winter wave.  Note that there’s no end of the Delta wave for those states.  Cases continued to climb from July all the way through to early October.  And has been noted in earlier posts, Idaho already had to declare “crisis standards of care”, allowing physicians to ration hospital beds to those who seemed most likely to survive.

The upshot is that instead of a distinct winter wave with a sharp peak late in November, their summer Delta wave just continued, with the result of a broad peak spanning most of September and October.  If I had to guess, I’d guess they’re done for the year.

The rest of the Mountain states, by contrast, had a low peak in early September, consistent with the end of the summer Delta wave.  And they’ve (mostly) started on a new Winter wave.  Those states don’t appear to be done yet, for 2021.

Time and again, the pandemic has run right up to the point where hospitals are full, and then stopped.  It’s as if there’s only a finite subset of the population at risk of infection at any one time.  And once they’ve been infected, the pandemic runs out of steam for a while.

And so, in those three northernmost Mountain states, it surely looks as if the summer Delta wave worked its way through all the infection-eligible population.   For all intents and purposes, it looks like the pandemic has simply run out of steam, in those three states, for now.  And so, there won’t be a  winter wave cresting around Thanksgiving.  Instead, they had two months of high case loads under Delta, leading to (among other things) crisis standards of hospital care in Idaho.

Maybe that bodes well for the entire U.S.  Maybe we’re finally reaching the point where there’s just some modest subset of the population left to be infected.  And so instead of a rip-roaring winter wave, as we had last year, maybe we’ll all go the way of WY, MT, and ID.  Maybe the people still capable of being infected will end up so thin on the ground that we’ll have just a modest peak this winter.

Post #1324: Winter Wave Buzzword Bingo


Looks like the U.S. is stuck on repeat, in terms of a COVID-19 winter wave.  I suppose that in addition to the case counts, this probably means a replay of much of the nonsense reporting we saw last winter.

In the interest of efficiency,  in this post, let me try to handle some of the known nonsense in bulk, rather than piecemeal.  We may see novel forms of it evolve.  But at least I can efficiently dispose of the stuff that’s a straight-up repeat from last year.

To be clear, this post is all about things that are wrong.  But that nevertheless make great click-bait, or allow people to feel all righteous about something, or otherwise seem to stick around despite clear evidence that they are incorrect.

This underscores yet again the fundamental problem with America today:

  1. Every citizen feels the unalienable right to have a strongly held and firmly expressed opinion.
  2. Nobody feels even the tiniest obligation to do their homework.

Hence, this list of pre-debunked items from last winter.  I’ve already done the homework so you don’t have to.

1:  Post-holiday surge.

So far, we have never had a post-holiday “surge” in cases in the U.S.  Never.  Not even once.  Not post-Thanksgiving.  Not post-Christmas. Not post-Superbowl, Memorial Day, 4th of July, Labor Day. Not not post-fill-in-the-blank.  Never happened.  Never never never never never.

And yet, you will almost surely hear solemn and dire warnings about the expected post-holiday surge in cases this winter. 

When you hear that, bear two things in mind.

1)  Nobody except me has ever bothered to define what “a surge” means.  A “post-holiday surge” is a coordinated increase in daily new COVID-19 cases, above the pre-existing trend, occurring in all or most states, roughly two weeks following the holiday.  (That last clause is consistent with the average lag between date of infection and date of case reporting.)

2)  Consistent with that, I seem to be the only person who has actually bothered to look at the data, after the fact, to see whether or not a surge actually occurred as predicted.  Among the reasons that you’ll never see that in the newspapers is that you have to wait two weeks to see whether or not a surge occurs.  That’s far longer than the attention span of the American people and their news media.

The data-driven part of this is not exactly rocket science.  U.S. reported COVID-19 cases fall on the holidays because state health departments are closed. Then (most of) the backlog of cases gets reported, and then cases return to their prior trend.  Not just once.  But over and over, holiday after holiday.

We have never had new COVID-19 cases surge, or jump, or suddenly rise after a holiday, so some level beyond the pre-holiday trend.  Never.

Below, you can see that I looked for those post-holiday surges, really hard.  Rather than walk you through it, I’ll refer you to many prior posts that go through this in detail for Thanksgiving, Christmas, and most specifically, for the Superbowl.

And Columbus Day.  Labor DayLabor Day again.  You get the drift, I hope.

Near as I can tell, all of this “post-holiday surge” nonsense dates back to a single epidemiologist who incorrectly asserted that there had been a huge surge in cases following Canadian Thanksgiving.   Why?  Because case counts were higher the week after Thanksgiving than the week before it.  But a) it takes two weeks for new infections to enter the data and b) that same statement had been true for every week so far that winter, up to that point, in Canada.  (Because they were in their winter wave, and cases were increasing week-after-week.)

No matter how obviously stupid and illogical that “surge” conclusion was, that has stuck.  And every holiday since, you’ve heard all the talking heads tut-tutting about the dangers of holidays and the inevitable post-holiday surge in new cases.

Accordingly, my favorite post-holiday-surge graphic is this one, suitable for the many Americans who know nothing about Canadian culture:

Posted on December 13, 2020

Stop: Before you read any further, take 30 seconds, study the graph above, and identify when Canadian Thanksgiving occurs, based on the huge surge in cases that occurred two weeks afterwards.

The other graph that gets the point across is this one, where I’ve simply filled in the trend line.

Posted on January 13, 2021

Source:  Calculated from NY Times Githib COVID data repository.  Thick blue line is the U.S. average.

What we actually get with major holidays, first and foremost, is a steep reduction in reported cases because state governments shut down.  And, oddly, we also get small, permanent reductions in actual total cases.  Seems as if a lot of people just don’t bother to get tested over a holiday.

This is followed by a return to the prior trend, as all that gets sorted out.  What we never get is a surge in cases, above prior trend, two weeks after the holiday.

Finally, if you want to see what a surge actually looks like, try Post #920, Now That’s a Surge.  That’s just a summary of the past two decades’ worth of U.S. flu seasons.

If you wonder why we never get a surge, consider how one-sided the pundits’ discussion of the holidays is.  Two things happen on major holidays.  One, work stops.  Or at least, slows down for most of us.  Two, families gather.  All you ever hear about from the talking heads is the second aspect.  But, arguably,  the reduced risk from time off work may more-than-offset the increased risk from family gatherings.


2:  Massive increase in air travel for the holidays.

Ah, just see Post #939.  Here’s the graph of actual TSA passenger screenings at U.S. airports.  Spot the huge holiday-related surges, please:

This is the fourth quarter of the year, and the blue line is 2020.  The thin wavy line is the actual  daily data, and the thicker line is a seven-day moving average.  So, yes, there are more airline passengers in the holiday season.  But the increase is only about 100,000 per day above the baseline.  Against a U.S. population of 330,000,000, any modest rate of infections occurring on those flights is just rounding error.  I went through all that arithmetic in Post #939.

3:  Predictions of smooth sailing from now on.

So far, official predictions of the course of the pandemic have had poor accuracy.  I’m hardly the first person to note that.  Or to have shown that with some degree of formality.

In the first year of the pandemic, when you’re still trying to figure things out, that’s forgivable, I guess.  But now?  When we have more than a year’s worth of experience?  When a few simple factors such as indoor relative humidity appear to explain the Southern summer wave and Northern winter wave?  When state after state goes right to the point where hospital beds are completely filled, but rarely goes much beyond that?

By now, it seems like we pretty much ought to have a fairly good grip on the likely ups and downs.

Arguably, some of the worst predictions have come from CDC, from the consortium of modelers that they tasked to provide some sort of official prediction.  Let me just repeat the last one, which is that there would be no winter wave this year.  That prediction was made in September.

Source:  Cited just above

These predictions suffer from the same basic problem as “the surge”.  Our collective attention span is so short that no news agency can be bothered to look back and check the accuracy of the forecast.  And so, assess the forecaster’s history of inaccuracy.  Before presenting the next forecast.

As a result, the same forecasters who repeatedly provided poor predictions in the past remain free to provide equally poor predictions going forward.  And whenever those predictions match something that somebody wants to hear, you’ll see them in the popular media.  And the long track record of being incorrect will just be swept under the rug.

The CDC-organized group that made the projection above, in early September, is slated to redo their projections again in December.  Without a doubt, given the CDC imprimatur, whatever they say will be duly reported in the news.  And equally without doubt, their incorrect prior predictions will simply be ignored.

4:  Florida has fewer cases than fill-in-the-blank now, thus proving fill-in-the-second-blank.

I went through one of these in Post #1319.  The current Fox push seemed to be “thus proving that masks don’t work”. I’m sure at some point that second blank will morph into “vaccines”, or maybe “Biden”.  Whatever fits the need of the moment.

All of that is aimed at people who are too stupid and lazy to take literally one minute to look at the actual data.

All you have to do is Google “COVID Florida”, and Google will provide you with a graph of new COVID-19 cases in Florida over time.  Barring that, Googling “NY Times Florida COVID” will give you a graph of COVID-19 cases in Florida over time.  Either of which could then be compared to (say) California.  In either case, you’d see that most of the time, Florida has exceeded California in terms of daily new case count.

It’s not as if we lack for ready sources of the data, available with absolutely minimal effort.

It’s just straight up Rule 1 and Rule 2 above:

  1. Every citizen feels the unalienable right to have a strongly held and firmly expressed opinion.
  2. Nobody feels even the tiniest obligation to do their homework.

Normally, I would add a third rule:

3:  And nobody cares whether or not they are right on the facts.

But Rule 3 is a misstatement.  Near as I can tell, it’s far worse:  Nobody wants to see facts that might contradict a firmly-held belief.  It might make then have to think, and thinking is hard.

And  so, as outlined in Post #1319, any set of statistics that varies over time or in a cross-section becomes a ready propaganda tool.  Wait until the numbers are leaning your way, add whatever semi-attached figure you wish, and loudly proclaim that the data make your point.  When they lean the other way, say nothing.  And you can reliably count on the laziness and stupidity of your audience to do the rest.

It’s cheap, simple, and effective.  Of course it’s going to be re-used this winter.  To stop it, people would have to wise up.  And I predict that is unlikely to happen, based on our experience to date.