Apologies to regular readers, but I got real busy the past few days. So just a mini-update on our model, which will be useful for making a couple of small points. Plus it’s Good Friday. Regular posting will resume on Monday, after Easter.
All the descriptions, caveats, warnings, explanations and everything else are in the regular updates. All data current 8 PM Thursday EST.
Look to the bottom for the important discussion on the blamestorm that’s coming. It’s on the horizon, dear ones. People are still panicky now, but it won’t last. When this is over, the loss of trillions of dollars is going to be noticed.
The overall reported totals:
Not much change from Monday, which is a good thing, and which is an indication we are post-peak. Total reported cases 2.06 million (small increase from Monday), total reported deaths 142,000 (up 10 thousand from Monday). Not surprising, since the reported deaths peak after reported cases, since you have to have it before you can be reported to die from it.
If there’s a third peak in the fall, or the Lord knows when, this model cannot see it.
Daily reported cases:
Do not expect that plunge to remain. I’d be shocked if it didn’t bump back up. Reports have a high variance. I’ve said it a hundred times, but I repeat it here: we are modeling reports of cases and deaths, and not actual cases and deaths.
The interesting thing will be to predict, using the model, the post panic date. When do the over-reactors and nervous politicians calm themselves? I don’t think you can get it from the global data. Since lock downs and the like are political, you have look at data inside political jurisdictions. See below for USA.
But if you wanted to use just the global, then maybe 7-10 days. Not until new cases and deaths cease being reported, but when politicians feel safe talking about something else. Like the economics of all this.
Daily reported deaths:
Same warming! Regular readers will recall the daily reported deaths have a strong up-down up-down up-down pattern, because of differences in reporting times. You can anyway see the lag in reported cases to reported deaths peak.
I’ll save the acceleration plots, which I don’t think we need anymore.
Here’s the reported deaths divided by reported cases. Still soaring upwards!
Note that we know we’re nearing the end when this thing starts to flatten.
Totals for USA only.
On Monday the model said “Total reported cases: 510,000. Total reported deaths: 18,500.” But now it says total reported cases: 590,000. Total reported deaths: 28,300. This is a huge jump in reported deaths. We see why in a moment.
First, daily reported cases:
Don’t forget this is across the entire USA, which has areas which start reporting and testing at different rates. Hence the double peak.
The real interest in the daily reported deaths:
What happened? Well, this Birx and others, as we have all heard by now, emphasized the way to report corona virus deaths was to “conservatively” (their word) report any death with as a death from. Incidentally, when some of us worried about this early on, we were called bad names. Ah, well. You can read the CDC guidance on death reporting here.
Dying with and not from necessarily produces higher totals. Our model is a model of reporting, and it cannot see abrupt changes in reporting methods. Tough cookies for our model. This change is not so much abrupt, but a re-emphasis of extant guidance.
On the other hand, we have the double-tap peak in reported cases, too, indicating the additions to testing.
Finally, the reported deaths divided by reported cases.
Same bump can be seen, and it’s also not flattening.
This is all we have. Using this, when does the American panic subside?
Maybe sooner than we think. This Anthony “Don’t Shake My Hand” Fauci is already walking back the predictions of reported deaths. He and his team went from 1 to 2.2 million, to 100 to 250 thousand even assuming full compliance with panic measures, to now admitting reported deaths will be “likely closer to 60,000“.
Words, you notice, which allow him to claim success if it’s 30,000.
If you were a cynic—Heaven forefend!—you might think they Facui-Birx tag team insisted on reporting death-with from death-from because they saw their initial totals were going to be ridiculously high.
The re-emphasis is a good way to boost totals. Not enough to make up the discrepancy to 175 thousand (the median of their full-compliance window), or even to 100,000, their low end.
As I said on Tuesday, anything under 100 thousand is a blown forecast. They cannot claim “social distancing worked even better than we thought!”
No. The low ball 100 thou was with “social distancing”. If under 100K, it’s a blown model. We have to look for other reasons for the corpse shortfall. Like maybe the same causes of what made the other pandemics and flu to go away are more important that the experts thought. Or we have to redefine what “social distancing” means.
Listen: those modelers have to justify the spending and loss of trillions of dollars. Not billions: trillions. Trillions! Not just on the printed-money stimulus, but the job losses, the upheavals, the chaos, fear, shortened lives, retirement funds gutted, and on and on. And on some more for years to come.
How’d you like to be the guy who said “We must full on panic!”? Especially when it’s more and more looking like your body count was way off. Brother, I would not want to be that guy. I can’t imagine being responsible for losing a couple of thousand.
What if I was the man held to account for crap-canning trillions? I too might be tempted to do anything I could to make myself look better.
To support this site and its wholly independent host using credit card or PayPal (in any amount) click here