Tuesday, May 5, 2020

The data on Kung Flu deaths is getting better

CDC just revised Wuhan Flu deaths down by 44%, to 37,000.  Here's why:

Click through to the CDC's web site - the data are very interesting.  Everything that I've seen reported by the idiot media has been a big grab bag of "Death by Kung Flu" + "Death with Kung Flu" + "We think it might have been Death with Kung Flu but couldn't be bothered to test to make sure".  Now CDC has broken this all out explicitly.  Looking at just the "Death From Kung Flu", things are looking pretty encouraging:

Source: CDC

We're on the other side of the mountain.  The graph looks the same if you plot "We think it might have been Death with Kung Flu but couldn't be bothered to test to make sure":

Source: CDC
How dodgy is this particular plot (the "Kitchen Sink" Kung Flu Deaths)?  About the same number of people died in the week of 4/25/2020 as before the flu broke out.  That is either extra crazy medical juju, or it's crummy data.  But even with data this bad, it's clear that we've crossed the mountain.


1. The data have been dodgy but we're getting better visibility.  Things are better - a lot better - than anyone has been saying.

2. Even using crappy data (where nobody dies of natural causes anymore) it's very clear that things are a lot better than anyone has been saying.

3. Prediction: when the data for this week are in, the plot trajectories will continue in the same direction, heading towards zero.

4. There is absolutely no data justifying maintaining the lockdown that has thrown 30 Million out of work.  The trajectory of that graph is also clear - each day another million people lose their jobs.

5. There is no data that shows that the lockdown is saving lives.  There is no data that shows that the lockdown is preventing ICUs from being overloaded.  You can say that after the lockdown were imposed the deaths peaked and then declined, but the lockdown went into effect a month ago - why the delay in turning the curve? Where's your control group so you can compare lockdown vs. non-lockdown deaths?  This is nothing but post hoc ergo propter hoc reasoning.  OK, maybe it's also bureaucratic ass-covering.

6. The lockdown is being implemented by fascists.  That by itself is sufficient reason to end it.

We're over the mountain.  It's over, we're just waiting for the Fat Woman to sing.  Given how wretched our government is, that means it will only be another month of lockdown.


LindaG said...

I couldn't find the click through to the CDC but this is a very interesting post.

Makes me even more annoyed that governor's keep extending the lockdowns.

Borepatch said...

Linda, the link is in the first sentence, and in the caption for both graphs.

What's interesting (and which I didn't include in my post which was already too long) is that there is State-by-State data at that same links. A bunch of States are in lockdown even when they've only had a handful of deaths (e.g. Delaware and Maine, just off the top of my head).

Yeah, the Governors are destroying their States.

Aesop said...

Oh, God, another one....

How is it "getting better" when you try to palm off provisional data that won't be finalized nor current for weeks, as though it were graven on stone tablet?!?!?!?

No, Borepatch. They didn't revise anything down.
It was one click to the site to reveal what those figures are, and why, printed on the same mother-farking page. Criminy sakes, man, it's in 40-pt letters, in it's very own block:

"Provisional death counts are based on death certificate data received and coded by the National Center for Health Statistics as of May 4, 2020. Death counts are delayed and may differ from other published sources (see Technical Notes). Counts will be updated periodically. Additional information will be added to this site as available. The provisional counts for coronavirus disease (COVID-19) deaths are based on a current flow of mortality data in the National Vital Statistics System. National provisional counts include deaths occurring within the 50 states and the District of Columbia that have been received and coded as of the date specified. It is important to note that it can take several weeks for death records to be submitted to National Center for Health Statistics (NCHS), processed, coded, and tabulated."

They go on:
Why these numbers are different

Provisional death counts may not match counts from other sources, such as media reports or numbers from county health departments. Our counts often track 1–2 weeks behind other data for a number of reasons: Death certificates take time to be completed. There are many steps involved in completing and submitting a death certificate. Waiting for test results can create additional delays. States report at different rates. Currently, 63% of all U.S. deaths are reported within 10 days of the date of death, but there is significant variation among jurisdictions. It takes extra time to code COVID-19 deaths. While 80% of deaths are electronically processed and coded by NCHS within minutes, most deaths from COVID-19 must be coded manually, which takes an average of 7 days. Other reporting systems use different definitions or methods for counting deaths.

Aesop said...

"Things to know about the data

Provisional counts are not final and are subject to change. Counts from previous weeks are continually revised as additional records are received and processed. Provisional data are not yet complete. Counts will not include all deaths that occurred during a given time period, especially for more recent periods. However, we can estimate how complete our numbers are by looking at the average number of deaths reported in previous years. Death counts should not be compared across jurisdictions. Some jurisdictions report deaths on a daily basis, while others report deaths weekly or monthly. In addition, vital record reporting may also be affected or delayed by COVID-19 related response activities.

IOW, you're looking at data from 1-8 weeks or more ago, and it hasn't been finalized.

My calendar says that would be as far back as March 5th, "or more", by their own admission, right on the same page.

Can we not at least agree that if we're going to link to something labeled "PROVISIONAL DEATHS", wherein how and why it's explicitly "provisional" data is explained in terms anyone can grasp, not once, but three separate times on the very page, that maybe that information ought to be read, and factored in, instead of focusing on "44,000 less deaths" as though the latter were settled history, when it's clearly identified as anything but??

This has repeating error has to be the month's Emily Litella Award Grand Prize Winner, and I've seen the same error repeated on at least three sites now that should bloody well have known better, had even two seconds' careful investigation been invested.

You've joined them in posting information clearly labeled as being the score at the end of the first inning, and called it the final score. Anyone can make such a mistake; pray, let it not become a pattern.

Borepatch said...

Aesop, then CDC needs to start publishing the percent uplift (by week) for late reporting. Historically, this can easily be tracked and can let us project what the final results are likely to be. They don't do this, even though it would be trivially easily to do so, e.g.:

one week ago: +5%
two weeks ago: +8%
three weeks ago: +10%


Until then, any projection of final numbers are nothing but conjecture.

And until then, the data are what the data are. Your projection may be different, but it's based on conjecture because it sure as shootin' isn't based on data.

Rick C said...

"You've joined them in posting information clearly labeled as being the score at the end of the first inning, and called it the final score. Anyone can make such a mistake; pray, let it not become a pattern."

But you've been touting the "record anyone who died as having died of COVID even if we didn't bother to test them" numbers all this time. They're no better to justify throwing 30 million--and counting--people out of work than the lower number is to say "panic's over, throw everything open again!"

Glen Filthie said...

Some folks are just stuck on stupid.

If it was a pandemic... would there even BE an argument? Sheesh.

Up here it's just ludicrous. They are starting to open things up a bit. The nature trails north of town are open but the dog parks are firmly closed. The casinos and bars are going to open, but the churches are closed. Way to flatten that curve, morons! I am beginning to think there is some fat old bitch of a cat lady, well into her third box of chardonnay making all these rules. And her cucks will defend this lunacy to their dying breath.

It's time to start telling stupid people to FOAD. You can't run a country like this.

Rick C said...

Everyone in rural counties in Montana needs to stay home for another year! But it's OK for Bill de Blasio to go for a walk in NYC, for Lori Lightfoot to summon a hairdresser to her house, and Neil Ferguson to have his (married) mistress over to his house even though she thought her husband had Winnie the Flu.

Aesop said...

Hey Rick, I'm surprised you're not in mourning.
Remember last week or so when you assured us that some spiffy splendiferous new model predicted this thing would peter out at 67K deaths in the US, by August 4th?
And I told you in reply they were smoking hopeium, and we'd get there by more like May the 4th?

Read 'em and weep, buddy:

So to recap, one of us hit not just the bullseye, but the "X" ring in the bullseye; and one of us missed it by a country mile.

Just a humble suggestion, mind, but maybe you oughtn't be the one telling anyone what numbers sound more sciencey.

alo 89 said...

I've been working the numbers for weeks. This CDC site, though it says the numbers are 100% for week 16 will almost certainly be revised upward. Every week the numbers are tweaked (going back to January). Howeverd news reports it does seem fairly clear that the CDC changes fruits pretty regular (not apples and apples) so all the charts are partly bullshit. That said, based on past revisions, I think were still gonna see a downturn between week 15 and 16. In final analysis the fellas supplying the numbers are a bunch of lyin' liars.

Borepatch said...

Aesop, even if you *double* the last week, it's still showing that we've crossed the mountain. And I've been banging the drum that alo 89 hits here about mixing apples with oranges - the Fed.Gov has so clearly screwed the pooch in their model projections - used to throw 30 M out of work - that there is probably a desperate attempt to find any dead body they can blame Kung Flu for.

Because otherwise they'll be swinging from lamp posts.

But even if you take the high number (67k) and double it (to 135k), that's about 6% of the model predictions. If I were a CDC modeler, I'd be looking to change the subject, too.

Ted said...

Context is being left out of all these numbers. Look a little further down the CDC report. for 2017 they have final numbers od deaths from all causes ranked.

1 Heart disease 650 K
2 Cancer 599 K
3 Unintentional accidents 169 K
4 Chronic lower respiratory 160k
5 Stroke 146 k
6 Alzheimer's 121 K
7 Diabetes 83 K
8 Influenzas/Pneumonia 55 K

so in a "Normal " year we see some 55,000 flue deaths a year , so even if the Kung Flue claims 100,000 we don't even equal Alzheimer's

And for this we shut down the economy????

Midwest Chick said...

There are craptastic data everywhere. Choosing an arbitrary February start date has skewed every single model. Different metrics, cherry picked base information, in some cases with political goals or federal dollars, have also had an effect.

What will really tell the tale is the widespread antibody testing that will, most likely, bear out reports that the Wuhan virus has been in the wild in the US since November/December and that herd immunity is very likely already in play.

As an aside, when someone uses the word humble, generally they are not-so-much.... I find that a comment’s word count and officious verbiage combined with vaguely veiled disrespect undermines the effect of an argument.

alo 89 said...

Wow. I get cited in a post in the ongoing debate between Aesop and Borepatch. Thanks. You made my day.

Jester said...

Is it too much to ask that we get actual stripped out data on the deaths that are being associated with the kung flu? Who has no known co morbid conditions that dies from it? Who dies of co morbid conditions, and what co morbid conditions do deaths getting tagged with this are out there? Can we get tracking on the infections and the lines of work? It seems awfully punky that some of us get to work and others get to shelter in place (Seems those non essential workers are getting really good at telling everyone that has been out there since day one how to live their lives for them.) I do agree with Aesop that we need to have a shit load of testing. Perhaps having GM or Ford making those fuckin things (Im being faceitious, I know getting things spun up to make diffrent items takes time but we now have a shit load of vents that are not being used instead of PPE being mass produced or test kits. And the evidence also seems to show that most people intubated are dead anyway..) instead of vents would be more useful. Since ya know as pointed out we don't have the workers capable of running the fuckers anyway. Why can't we have actual data of who dies from what and where, why and how instead of the blanket covid money train? The issue with these shut downs Aesop is that most the shit does not really make any sense first off. I can't get say a hair cut (Big woopie I don't care) but someone can prep my food and package it and bring it to my car? A mask on changes that for them but not someone else? And as pointed out we have the despots that are calling these innane rules not even fucking following themselves. You can't buy gardening seeds but you can wander the rest of a Walmart? I'm in no way denying the seriousness of this and the fact that our just in time method of hospital supplies has been exposed to be a total disaster. I just want actual data instead of piled up BS so, ya know folks can make good informed decisions. My issue with it all is this was to flatten the fucking curve. Well other than NY and NO where they encouraged people to party it up we have flattened the curve. Can't we now since the supposed thing was we need to be able to take the influx of people now open things up cautiously so that people can work? We sure as heck seem to have the hospital capacity.

Rick C said...

"Just a humble suggestion, mind, but maybe you oughtn't be the one telling anyone what numbers sound more sciencey."

Ignoring the beam in your eye, Mister two million deaths guy?

Rick C said...

What the Closers have led to is a judge in Dallas telling a woman he won't put her in jail if she apologizes for the temerity of not cowering in her home, and two guys who shot a security guard to death in Detroit for arguing with a woman who wouldn't wear a mask in a store.

Oh, and a SWAT team shutting down a bar with a handful of cops pointing rifles at everyone.

But at least we're saved from two million deaths in the US!

Oh, wait.

McChuck said...

@Aesop -
Please listen to yourself. You're arguing for fascism on the grounds that, "If it will save just one life".

We have impoverished 500 working class families (mostly young and healthy) for each death (mostly old and sick). As a society, we must not sacrifice the young on the alter of the old.

Ken said...

@McChuck, you're absolutely right about not sacrificing the young on the altar of the old, but that boat done sailed.

When AARP used to send postage-paid reply envelopes with their crap (I'm not 60 yet but if I squink my eyes just right I can sort of see it from here) I used to write "Get back to me when you stop supporting gun control and inter-generational looting" and send it back.

Can't say much more for "we the people" neither. Too many of the older members of my generation will cheerfully beggar their grandchildren for "free" boner pills.

Aesop said...


So your plea isn't that you didn't know you were misrepresenting the data, it's that you don't care? Noted. Perhaps we can revisit the casualty count today, in a month or two, when that data is finalized. Somehow, I don't think you'll want to go there then either, but time will tell.

Feel free to link to the time and place where I told you there would be, absolutely, bar anything, 2M deaths. If you can't or won't do that, you're talking out of the south side of your underpants. Like Joe Biden, the flaws of your memory don't become reality just because you believe them. I already explained to you some days and weeks past, how to calculate the final tally for this outbreak. So tell me how far it spreads, and show your work, and we can talk. So far, the last thing that sounded science to you was the death tally from last Sunday. So, is the outbreak over now, nationwide, everywhere? Or did you biff that guess?

I'm not doing any such arguing for facism, now, then, nor any other time. Read what I wrote. You're hearing a dog whistle no one is blowing.
Pointing out that someone has ham-fistedly fornicated provisional data into something it's not is called "pointing out the obvious".
Neither more nor less.

Having established that you have no further point on that topic, perhaps you'd care to explain why it's wrong to sacrifice the young on the altar of the old, but that doing the reverse is magically okey-dokie. Or was that not what you meant, every time you've advocated it?

Rick C said...

Here's some of the alarmist crap you were peddling, Aesop. I've helpfully put in numbers. No, you never said absolutely 2 million people were going to die, you just asked everyone who disagreed with your if they were OK with millions of deaths, and implied we should wait for a near-perfectly-accurate tests. And you're pretty free with the insults for people who question your meltdowns, in spite of Borepatch's and ASM's comments about how polite the discussion's been overall. It must be awful when the peons don't kowtow to your expertise. Meanwhile the IHME predictions you no longer like suggest 134K deaths by August, with a range of 95-240K which, yes, is worse than the average flu death (although not as bad as your exaggeration a few days ago; the CDC numbers for the last 11 years show an average of 37K deaths a year.)

"we can plan on casualties from turning people loose unilaterally at 4x-25X the current number of dead, which is about 54K today. We're due to blow right by Doofus Fauci's low-end 60K dead by Wednesday or so, still in the first two months of this thing, and while flattened, the death rate of about 2K/day seems to be holding steady in perpetuity. And we have 49 states itching to compete with NYFS for casualty king."

1.3 million deaths.

"So plug all that in, and tell me, with existing technology, whether or not you want to kill between 25X and 200X more people, or some smaller amount, by unilaterally ending lockdowns with no universal testing, and why or why not. If your answers is "Let's not do that", tell me how you do anything more sensible than testing everyone, with a validated and accurate (say >98%) test, to determine who is either clean, or over this and no longer contagious, before letting anyone of them back into the population-at-large pool to play."

1.1M - 9M deaths.

Borepatch said...

Good grief Aesop. Of course I care, duh.

I've been complaining about the poor quality of the data for a long time, but now I've shifted to posting about how the data is less bad. NOT perfect, just less bad. I posted about that again today.

We can extrapolate from that less bad data better than we could before. That extrapolation shows that the expected death rate is between 50x and 100x fewer than the models predicted. That prediction was used to throw 30M out of work and put a $10T hit on the economy. NOBODY voted on that, and nobody is voting on it now.

The situation os so wrong that anyone who wants to keep the lockdown in place needs to show some seriously rigorous work on why we should take another 10T hit and permanently damage the economy, Great Depression style.

Yeah, I care. I care a lot.

McChuck said...

@Aesop -
"Having established that you have no further point on that topic, perhaps you'd care to explain why it's wrong to sacrifice the young on the altar of the old, but that doing the reverse is magically okey-dokie. Or was that not what you meant, every time you've advocated it?"

I don't recall advocating that anybody be sacrificed, young, old, or middle aged. (Except communists. They can all die in a fire.) The old and sick should by all means hide in fear from this virus. But there is no reason for anybody else to do so. We should be sending our children to the homes of the infected so they can build up immunity.

But if, for some bizarre reason, someone must be sacrificed, it is far, far better to sacrifice the old than the young. This is axiomatic to civilization. "I believe that children are our future."