Holiday Fallout And COVID News Without a Narrative
Probably the least creative thing I do here is my monthly COVID report. I gather the data, group it, and spit it out in a way that I hope gives some insight. I just did this two weeks ago before the Thanksgiving break, but we’re in a period of this crisis in which I think weekly updates are appropriate. This week won’t be a region-by-region assessment but rather an less comprehensive overview of exactly what is happening and where.
COVID and the Holidays
What Is Happening Right Now?
Disney Shorts: Drip Dippy Donald
COVID and the Holidays
Let’s start with some big-picture concepts. The first is the factor of the Thanksgiving holiday on COVID numbers. When holidays hit, we tend to see a near-immediate drop in cases and deaths simply because reporting takes a holiday. The 4th of July was right in the middle of a surges, so that drop looked more like a pause, but we saw it clearly on Labor Day where the numbers dipped for a week in perfect synchrony with the holiday weekend. Most case tracking is using a 7-day rolling average, which helps smooth out the fact that case and death reporting drops on the weekends and picks back up at the beginning of the week. Adding a holiday to that mix means that people don’t want to go in to get a test, testing facilities and reporting departments are short-staffed. The result is a predictable bumpiness in the data.
We should absolutely expect that bumpiness around Thanksgiving. Already it looks like our case data has peaked and is dropping. If I had to propose an explanation in this pattern, it would be a surge in case tests as people get ready to see family for the holiday and then a drop in cases as no one wants to go get tested on Thanksgiving day.
This is exactly what we are seeing so far.
The next step is most likely going to be what looks like a surge in cases but is, in fact, cases that would have been detected on a normal Thursday waiting until Saturday to go in for a test.
One thing I do want to note is that there were panics about both the 4th of July and Labor Day (and Memorial Day) being super-spreader events where people get together and everyone gets COVID. None of these predictions have come to pass.
In fact, prognostications of doom based on holiday get-togethers have fallen so incredibly flat that my inclination is to say that the macro forces of weather, indoor socialization, and the natural curves of pandemic spread are such powerful forces that the micro effects of a single holiday are unlikely to make a significant difference.
What Is Happening Right Now?
I can’t do a region-by-region assessment every week, but I want to call out the things that I see that I think are important. During my week off, I was incredibly frustrated simply because, when I’m not personally going to grab the data myself, I have no idea what is going on. Reports that I see through traditional media channels are simply garbage.
I’ll try to hit some of the large scale things I’m seeing in the data and then toss some charts at you to explain why I think these things.
Let’s start with some good news: The COVID surge in the US will almost certainly start trending down in the next month. We can unambiguously see COVID cases trending down in the hardest hit regions (this includes the mountain states, plains states, the midwest, and southern states that had largely avoided COVID in the summer). This started happening before Thanksgiving and we currently see the “false dip” from the holiday week. Even so, we are likely past the worst of this wave in most regions.
This is not universal. The west coast has not yet hit a peak, thought their numbers are much lower than most other regions. I don’t expect them to go nearly as high simply because I believe the culture of caution on the west coast is substantially more robust than in other regions.
Keep in mind that COVID positives are the canary in the coal mine. First we see the positives surge. Then we see hospitalizations surge. Then we see deaths surge. That is how the data moves and that is how the focus of the news stories shift in response.
If you’ve seen reports of deaths surging in the plains states like South Dakota, this is true. South Dakota has seen the worst surge in deaths per population since June. It has been less than the surge we saw in New York, New Jersey, and Connecticut in April and May but, given that we have substantially better medical mitigations, this is not encouraging. It is merely an example of how hard this virus hit the NYC area when we were unprepared for it.
In other areas like the southern states, things are not bad. Chalk this up to better weather or the caution inherited from a difficult summer, but the lowest COVID rates are in the south. So if you, like me, run across a news article that informs you that you that the Governor of Florida is hiding in a cave while everyone in his state is dying from COVID, keep in mind that this is simply not true.
I’m sad to say that news is unreliable, but that is just the reality. Reading the news on COVID numbers actually disinforms me. It makes me think that there is something wrong with my code or that I’ve mislabeled something or otherwise gotten the data wrong. But I haven’t. The news (as I see it from my algorithmically selected news sources) is just trash.
Disney Short: Drip Dippy Donald
This short is an interesting animation and narrative workshop. Not much happens through the entire short, it’s really just about Donald trying to get to sleep. In one sense, this is a very boring story, but it’s something everyone can relate to. It’s about the exhaustion and tedium of simply not being able to get to sleep.
Though the narrative is thin, the gags are superb. There is a joke where the dripping of a faucet that is keeping Donald awake becomes a thundering echo literally shaking the whole earth. Played for exaggerated comic effect that’s how I feel when I can’t get my sleep.
California is probably going to be quiet for a while because while we've started to see slightly colder temperatures, we haven't had much real weather -- daytimes are still in the 60-70s. There have been a series of ridges driving most of the rain into Canada, and this pattern is expected to be maintained into late December. In the northern and Sierra foothill counties, there has been cold weather (with snow in the mountains), and a corresponding increase in cases. Death rates, however, continue to look like the March-early June period. And it's still ~74% 65+ for deaths. <18 is a grand total of 2 (out of 136,904 positives in that age group).
This is encouraging to read, I am very curious to know if this is the last surge that we're going to see before this is all over or if we'll get hit with one more on the west coast/north east before enough of the population is vaccinated. If we think that we're at around 25% immunity already due to the caseload that we've seen and we're gonna vaccinate around 50M people by the end of february in the US, combined with additional infections of around 500K per day (tested positives x 3), that brings us to 80M + 50M + 45M = 175M, or 50% of the US population. I might make this my prediction, if we can maintain where we are at now without losing control, we might be out of the woods (literally) by April.