@RichardAEpstein @HooverInst got caught editing his past predictions again. It's happening so fast, I can't keep my piece updated, so here's a thread on how to deal with this kind of incurious grifting on COVID. Starting with my note https://twitter.com/RexDouglass/status/1243723120382316550

Just to clarify Epstein predicted in writing a # of deaths in the U.S. of
500
2,500
5,000
10,000
50,000

When 500 deaths passed, he said he meant to type 2,500. About the same time he edited the original piece to say 5,000 instead. Sometime in there, he edited the correction to make it look like he originally said 5k but meant 50k. Today that got retracted and blamed on an editor

This dumpster fire isn't what I want to focus on. I want to focus on another April 6 edit he posted which changed out the whole introduction and made a claim that the then peak of 9,655 (10k) was a sign the death count wasn't going to keep rising https://www.hoover.org/research/coronavirus-perspective-revised

In it, he laments that he's become a laughingstock because he made a too low forecast. That people are ignoring the substance of his arguments and focusing instead on one numeric mistake (maybe even a typo by some editor!). Let me alleviate that concern.

The charge isn't a bad forecast, it's intellectual dishonesty. My note documents 8 classes of unscientific and misleading arguments in his COVID pieces. The wrong forecast is much less important than the process leading to it being willfully incurious

The single greatest problem is the author does not actually care what is true. The business model is getting attention for being "contrarian," and being FOR something abstract like liberty or the economy. The university department would have to be called Applied Ideology.

Because the author's motives are insincere, the basic, first-year grad student level mistakes he makes can't engender a proper intellectual review and engagement. He already knows he's cherry picking, he already knows the argument is false, but he doesn't want to be accountable.

You might be familiar with the Gish Gallop or other similar debate tactics, where the goal isn't to be right it's to flood your opponent with wrong things beyond their ability to absorb and refute. Just by engaging a sophist, you've already lost.
https://en.wikipedia.org/wiki/Gish_gallop

But there are tactics to beat this. I'm fond of 4 in particular.

First, we have to use these as teaching moments. Epstein let me Trojan Horse a lot of basic social science methods lessons to audiences that might not otherwise see them. He's a concrete example to our students about how starting with an answer not a question leads to disaster.

Second, deconstruction (Jon Stewart on the Daily Show), is sometimes sufficient to show the in-authenticity of the author. When Epstein tells Reason he doesn't need a PhD because arguments are sovereign and then challenges the New Yorker to compare resumes, it's case closed.

Third, prediction or false construction (Colbert on the Colbert Report), is sometime sufficient. I concluded by predicting moving goal posts and more revisions, and the data generating process is producing just that. A growing mess of Wayback Machine spaghetti, means case closed.

Finally, scientific methods are like a vaccine to B.S. If you lay out the fundamentals once, they'll make the grifts obvious even when they appear over and over again wearing a different coat of paint. For example....

He's punted on all his false ideas about the state of the art having a constant R0, and now says he was originally concerned about 'logistic' models. He's straw-manning again against a field with many kinds of models and a distaste for simple curve fitting almost everywhere.

And we can tell he's faking a lit review because just before complaining about everything being a logistic model he links to a piece on Ferguson's agent based models that scared the UK. He deeply doesn't understand the literature he's critiquing. https://www.nature.com/articles/d41586-020-01003-6

That a select few are taking advantage of crisis to see what they can get away with, does not mean we need stronger gate-keeping. That millions of curious minds are applying a wide range of skills and trades to COVID-19 problems is a win for the species. We need to protect that.