Wednesday, April 22, 2020

Dear Kenny,














"Poor Aesop - this fucks up his whole cha-cha, doesn't it? 
While Los Angeles County has reported a total of 13,816 coronavirus cases, early results from an antibody study conducted with the University of Southern California shows that hundreds of thousands more could have had COVID-19 in the past, officials announced Monday."
Sorry to disappoint you, but not so much.
How science actually works is a wee bit different than you think.
So first, let's talk about this survey :

First, the Good:
They actually allegedly controlled for age, something the Stanford survey notably didn't do. When you let subjects select themselves, instead of picking them randomly, you fuck up any validity of the conclusions. So this time, someone appears to have gone to all the lectures in their statistics classes.

Now, the Bad:
The sample size is way too small.
883 people gets you wild inaccuracy, which is why even KTLA could figure it out, and soft-sell the point that the sample is b.s.
Don't believe me, work it out for yourself online for free.
No math skillz required, which means even Common Core math grads could do this.
If you wanted 99% accuracy (considering major economic and medical consequences hang on decisions made) and you want to minimize the margin of error at ±3%, for a population the size of Los Angeles County (10M in the County proper, or 12,4447,000 people in the metro area, btw), you'd need 1842 people in your sample.
Anything less is statistically invalid. (That means it's functionally bullshit.)
A sample size as small as was actually used gets you only about 90% accuracy, which is fine for determining whether someone wants your brand of laundry detergent, but not even accurate enough for presidential polling, let alone life-and-death public health decisions.

It gets worse from there.
The survey guesstimates that the population already infected just in L.A. County could be 221,000 people - plus or minus 100%. When you have a survey that would pass the sniff test, let alone peer review, you don't issue a result that widely disparate. If they'd split the middle, called the infected percentage 4.2%, i.e. 330,000 people ±110,000, any illiterate bumpkin could tell you were pulling guesses out of your ass.
Like this survey, and the Stanford survey, both did.

The Ugly:
No one has demonstrated, or even alleged, that the test used in these surveys is specific for Kung Flu (SARS-Cov-2), as opposed to any or all coronaviruses, including those that cause the common cold. So for all anyone knows at this point (because none of this happy horseshit has been scientifically vetted), they could be proving that 2-5% of L.A. was exposed to one of the cold viruses this year, and not Kung Flu at all.
Whoops.

But we'll forget all that, and go with this, even though it's basically boob bait happygas by any rational point of view, and pretend it's true, for about a minute.

So 4% of the county's been exposed enough to grow antibodies. Okay.
So opening things up means the other 96% will or might now get exposed.
So in the worst case, 100% of the county gets it, 97% of them are fine, and you only kill off 25 times as many people as have died so far.

So L.A. County could get up to 16,650 deaths (surpassing NYFC!) instead of their current 666 deaths (lucky number!). Let's sign up for that, since it's worked out so well in Brooklyn and Queens.














That took me about 5 minutes to type up and explain to you, or anyone else, even someone with only a grade school understanding and no professional acquaintance with how medical statistics works. Because I've studied that, and done it IRL, and you probably didn't. I didn't stay at a Holiday Inn Express, or just hear it from someone, I've had to go out and actually do it. Design an entire survey, sample results, and collate the data. From time to time, I still do, as part of my job. I'm sure there are things you do, or have done, as part of your job, that make you similarly familiar with what's reality, and what's bullshit. Silicon Graybeard is an engineer. If he talks about machining things, or radio waves, or Ham radio, or all sorts of really technical sciencey shit I have a vague familiarity with, I figure he knows what he's talking about. Phil over at Busted Knuckles refurbs transmissions and shop machines, so if he's talking about that, I know he's probably got a handle on that. Even Angus McThag, for another example, has probably forgotten more about auto repair than I'll ever know, or have to, so I would tread very lightly before I assumed he was full of shit if he was talking about an engine rebuild.

But as soon as medical knowledge comes up, every swinging Richard in creation thinks they know this shit cold, because they watched CNN for 15 seconds, or heard the radio news burp out two lines that sounds smarter than they are. I lost track of how many people on movie and TV sets, whose expertise on camera exposure or set lighting or truck driving I would never question, suddenly became Harvard Medical School board certified cardiologists - right over my damned shoulder - the minute I, the only licensed medical professional around,  was talking with a patient on set about their chest pain. Everybody in creation thinks he's Dr. Fucking Kildare, because they watched an episode of House, Scrubs, or ER.

It doesn't work like that.

I know what I know, and I study a helluva lot more than just what I do for a living. Triply so before I burp it out in a post online, because, like anyone, I don't want to be wrong, or mislead someone where their life could be at stake. I work as a Registered Nurse, and I specialize in emergency nursing. I carry a dozen specific certifications required to work in that discipline, which means I know a helluva lot about everything, because that's what I see. Every age, every complaint, every sub-specialty. My patients range from 0 minutes old to 109 years old, so far in my career. And I'll cheerfully admit to being only smarter than average in the field, and still learning new things every damned day. As I regularly tell people at work, after doing nursing for 25 years, and the ER for 20, I'm starting to get the hang of it. And that's only in two of the busiest ERs on the planet, and another one or two dozen smaller ones. I've been doing trauma and disaster medicine since before nursing school, back to when Daddy Bush was president.

And unlike the bloviating fucktards who should know WTF they're talking about WRT this virus, but don't, I've already taken care of at least a few dozen Kung Flu patients, whereas most of the jackholes in question (bozos like Anthony Fauci, two-time winner of the Chip Diller "All Is well! Remain calm!" Award) haven't been at a patient's bedside in twenty years, so I have just a wee bit more skin in the game when it comes to understanding this virus, and getting it right.

If you want my full C.V., I'll happy provide it. I know the things I know, and it sure as hell encompasses a metric fuckton of things far beyond just nursing, many (most, in fact) that have nothing to do with how I make my living, because I did a lot of things before I got my nursing license, but unlike any number of loudmouth anonymous keyboard trolls, I'll dispute anything on earth with facts to back it up, instead of just blowing out assgas. I put up with shit on my blog about as well as you do on yours, and comments here are turned off now not because I can't stand to be disagreed with, but because I now don't have to do diaper patrol on people too chickenshit to sign their handle, own their bullshit, and take the heat when they step all over their dicks, without getting all butthurt and personally offended for being called on it.

If you'd like a detailed fisking of the rest of the erroneous bullshit (which, let's be honest, was pretty much all of it) in your latest blogpost about me, we can go back and do that too, and I can post the links and quotes to show you where your memory of what I said has not even a vague acquaintance with the facts of what I actually did say. But I don't think you want to go there, because you'd look even worse when we looked at the videotape, and examined the posts in question.

If you want to ignore how bad the Kung Flu pandemic is, and will probably continue be, that's your business: suture self. You and anyone else has complete personal freedom to be as big a jackass as they care to be, on that and many other topics, and it's no skin off my ass whatsoever. But you don't get to have your own facts about it, and grasping at every half-assed survey, with Jack and Shit to back up its validity, isn't going to serve you well in the long run. That's a nickel's worth of free advice, but the consequences of your actions are all yours. Well, except for anyone else you infect because you didn't know you'd gotten it.

But if you choose to talk shit on the internet, and troll from your blog, you can expect I'll happily repay you back, and leave your credibility the worse for wear.

"He's now predicting widespread death and destruction over the beervirus.
Time is proving him wrong."
Srsly? Ignoring the fact that I've predicted no such wild thing, we'll hit 50K dead from this virus in a few days (which is an entire worst-case flu season, except in just six weeks), the stock market lost $11.5TRILLION (some people would consider $11,500,000,000,000.00 a lot of money...), 20M people are now unemployed, hundreds of thousands of businesses, even major ones, may ultimately fail, and the entire planet is sliding into worldwide economic depression, which will take years to overcome, and yet you figure I 've overplayed my hand on this?


















You're a funny guy (your "Shit I post on Facebook" is screamingly keyboard-alert level comedy), you've lived in CA, so you aren't knee-jerk shit-talking when you bash the idiocy that goes in Califrutopia 24/7/365, you find a lot of stuff on the big wide Internet I'd miss, and you aggregate a lot of things from more places than I'd care to visit, much of it interesting and entertaining, at an internet cost to visitors like me of exactly $0. So my sincere thanks for the free ice cream.

But when you figure I'm fair game for thoughtless slander, because your memory is too hazy to get things right, you can step right off and go fuck yourself. I expect you'd feel the same way if the situation was reversed and I'd spent the time to call you out and start making shit up, and if you say not, then you're a hypocrite as well as a jackass. That either applies, or it doesn't, so you can decide which end of that stick you want to own.
 
Maybe stick to what you know, can find out, or analyze and prove. And maybe worry less about me, and more about how much you don't know about how much you don't know.

No comments:

Post a Comment