COVID-19: The Schrodinger's Cat Crisis
No - cases aren't rising; the stats are still declining. Not that it matters
Once I grew up and read about Schrodinger’s Cat, I found it quite amusing. It’s a great way to explain the randomness of subatomic particles, even if it’s hard to grasp the implications at first. The premise is that a cat is put into a box, and then someone has to guess if it is alive or dead. Intuitively, you can’t really know the answer until you check inside the box [assuming it’s quiet]. The extra quantum twist is that until you look in the box, it is both alive and dead, and you only nail down one of the options when you learn the answer.
Sparing the details, COVID-19 has presented a perfect storm of complexity and confusion for politicians to take advantage of. The problem for constituents of every country is that the consequences of twisting the narrative during a pandemic can be far more tragic - and for our leaders, far easier to blame on something else.
On June 30th last year, I explained why Americans kept getting mixed messages from their leaders:
Mixed Signals about COVID-19 have fueled anxiety and political tension, when the truth somewhere in between. How deadly is COVID-19? As bad as it needs to be to win an election. A good politician never lets a crisis go to waste. A good COVID politician doesn't worry about defining the crisis - why put Schrodinger's cat in a perfectly good box when you've already decided what you'll say?
When the pandemic first arrived, the fear in Washington D.C. was palpable - the stunning bipartisanship of the initial stimulus package could only emerge if the threat was so obvious and bad that McConnell & Pelosi could find it more threatening than each other. However, it became immediately obvious that one side had more reason to embrace the pandemic than the other, and before the first stimulus checks were received Republicans and Democrats were feverishly digging trenches in the middle of the evolving scientific evidence.
Republicans knew that intentionally nosediving the economy was self-defeating as the November elections approached, and suddenly became passionately interested in any sign that the threat was overblown. Democrats watched Trump assume extra power and marveled at the ease of passing the largest stimulus in history. Suddenly, the effects of locking down the economy became acceptable, with little electoral downside: safe-siding all measures would surely save lives [oops…see below for actual statistics], a crashed economy with Depression-era unemployment provided ample emergency to tack on extra spending, all deaths could be blamed on President Trump regardless of the specific number, and if there ended up being a fall wave, they could blame that on him, too.
In an effort to provide both sides in an honest light, I must add that Trump’s tendency to encourage the medicinal use of chlorine and mock mitigation measures tended to undercut his own arguments.
Misinterpretation of statistics can sometimes just be timing. I just saw an article about how COVID-19 cases are rising in the US, published yesterday [Monday, 3/22]. The article mentioned that 27 states are seeing rises in case numbers, based on 7-day averages.
I'm sure that's true - except 7-day averages can't just include a single 7 day period. Most countries have charts with rhythmic 'spikes,' just like ours does. It took a while for me to realize that the spikes were tied to Mondays & Tuesdays. Half of the states report stats 7 days a week, but half don't, which is why there is a continual heartbeat rhythm to the daily charts:
The Johns Hopkins tracker is always updating, but US statistics are a day behind simply because people don't all die at a designated single time every day. Sometimes, on the weekends, the lag is a little longer. I always wait until Monday afternoon to add another week's totals to my chart, because otherwise the 7-day period doesn't include the latest weekend drop. Not including the latest Sunday will artificially inflate rolling averages when the trend is going down. This is exactly the same way I approached my 1/19 estimate that the true peak was over and the down curve was just beginning:
That was 2 days before the earliest news story and nearly a month before most of them [3rd pic]. However, most have been drawing the wrong conclusions about why [vaccines].
As you can see below, our current weekly peak and low point are each still trending down, and 5/7 days were lower than the previous week. The trend has slowed down, but that is more likely tied to a herd immunity floor. That tells me that the primary driver of the recent massive drop in cases during February was rising temperatures, not vaccines [as opposed to now].
Why not vaccines? Because a virus entering a population with 0% immunity is different than one trying to keep churning against a population at 50.066%, as of 3/24/21 [83,930,495 Americans who've received at least 1 vaccine dose + (29,918,936 confirmed cases / est. .375 test capture rate)]. Even as late as August 15th, cumulative seroprevalence hovered within 3% - 7% for most states (seroprevalence is the extent of infection within a population, as measured by the percentage of samples that tested positive out of all samples taken):
But not in late January / early February, when the rapid decline began. Other than the freak 7-day snowstorm [here in Texas], we've had 60-70 degree days, like much of the South. The 'massive drop' I referred to was the rapid Feb. decline in new daily cases, that has since leveled off:
83 million people with at least one dose is a lot of people, but the United States has a lot more people. There’s also no data to show how many vaccine recipients had already endured COVID-19 infection.
There’s a tendency to assume that anecdotal evidence is correct if the correct people say it, and that is probably the most disturbing aspect of every nations’ government responses to the pandemic. The most consistent error I’ve observed so far is tied to the relationship between vaccines and race - especially among black Americans. It is undeniable that black Americans have been catching and dying from COVID-19 at higher rates than any other group throughout the pandemic.
[NOTE: this portion is drawn from a conversation on Twitter; I wasn’t talking to myself, I think]
However, the lower rate of vaccination amongst some minority groups is being viewed as another symptom of the same problems as case/death rates, which is wrong. The numbers don't occur within a vacuum, and the narrative risks missing the true story. 1 factor is generalized: AA's are less trusting of vaccines, per surveys. The other is ignored. By chance, I updated my demo COVID-19 correlations yesterday; which I’ve done roughly quarterly:
Initially, I was trying to determine the impact of population density on virus transmission, but we shouldn't view vaccine implications. the same way. The key here is age, not race. White Americans are older, and became eligible for Vax's 1st. There's a .5 correlation swing at 65+. 3 of the states you mentioned first, VT/WV/ME, are the 4th, 3rd & 2nd oldest, and they are also among the most white. Combined they have 0.21% of the US's African Americans. As average age goes up, a state's proportion of black residents goes down; it is not sound to draw conclusions from Vermont's 8,412 black residents, assuming that they have statistical significance for the other 43.4 million across the country.
Ultimately, most of the disparity in vaccines comes from a younger population that wasn't as eligible for vaccines in the beginning. The broad consternation misses the point, and does nothing to lessen inherent misgivings in minority communities; using the numerical discrepancy to paint yet another aspect of the pandemic as racist is as unjust as it is effective. Black people continue to be far more likely to reject a vaccine [according to several polls, typically 20+ lower than other ethnic groups], and it seems unlikely that highlighting statistics that show low rates of access and acceptance will make them more trusting and willing to be given one.
Instead, it becomes a self-fulfilling prophecy.
Ultimately, intentional or unintentional misrepresentation of statistics is not a victimless crime, but there has been very little actual debate between the partisan politicians and media. Dr. Fauci, the CDC, WHO and many other medical authorities have engaged in similar tactics, presumably because debating responses during a pandemic would sow doubt amongst the public. No organization or government seems to have realized that withholding and tailoring the data to support a desired narrative or outcome can be far more damaging, if the public learns about it. Given the ability of doctors to use social media [especially local doctors with independent practices], pretending not to hear dissenting opinions [or discrediting the medical professionals who voice them] has simply shown how bureaucratic and timid public sector doctors are.
Continuing to favor the most safe-sided mitigation measures only protects one small sliver of modern societies - public sector doctors/officials and the politicians relying on them to limit casualties.
Bonus: Statistics showing the irrelevance of lockdown stringency on total excess deaths. I’ve argued for a long time that this should be obvious - any measure designed to prevent large-scale deaths from COVID-19 should focus on those most at risk. If 95% of victims were 50+, closing schools probably wasn’t going to make much of a difference.
I’m probably missing some important detail here, so just assume that these stats don’t matter. I’m not a doctor or a scientist.
Do it for the children.
A chart depicting the true relationship between case numbers and deaths; the average time from confirmation of infection and death is 14-17 days. Not that anyone would ‘mind the gap’ and adjust as needed, to make a snapshot look more or less favorable, of course. Especially during an election, protests, riots or other tense moments. Or when the appearance of control is desired.
For the children