Taleb wrong about masks , again

From Nassim Tooleb The Masks Masquerade: Incompetence and Errors in Reasoning Around Face Covering

Somehow this escaped my attention until only yesterday despite him writing it 5 months ago. As usual of anything written by Taleb, it is easy to refute, and went viral owing to Talweb’s huge popularity as a writer and all the media attention he has gotten for pushing a pro-mask agenda, not because of the strength and veracity of the arguments he presents.

Taleb’s idea of ‘research’ is posting screenshots and PDF preprints of Mathematica-rendered formulas, with no explanation of how the formulas work or why they are relevant, as opposed to actual proofs and logically and empirically-sound arguments. He uses the formulas to try to awe his readers into silence with his purported brilliance, as a substitute for actual reasoning skills.

It thus comes as little surprise Taleb’s CV is much less impressive than these formulas and arrogance on his part would seem to suggest, with just a handful of publications in bottom-ranked journals, published from 10-20 years ago. Apparently journal editors and referees are not nearly as impressed by Talab’s Mathematica-rendered PDF preprints as he and his mostly average-IQ readers are.

People who are good at exams (and become bureaucrats, economists, or hacks), my experience has been, are not good at understanding nonlinearities and dynamics.
The WHO, CDC and other bureaucracies initially failed to quickly realize that the benefits of masks compound, simply because two people are wearing them and you have to look at the interaction.

Let us say (to simplify) that masks reduce both transmission and reception to p. What effect on the R0(that is, the rate of spreading of the infection)?
Simply the naive approach (used by the CDC/WHO bureaucrats and other imbeciles) is to say if masks reduce the transmission probability to ¼, one would think it would then drop from, say R0= 5, to R0=1 ¼. Yuuge, but there is better.
For one should count both sides. Under our simplification, with p=1/4 we get R0’= p² R0 . The drop in R becomes 93.75%! You divide R by 16! Even with masks working at 50% we get a 75% drop in R0.

This may only be true is people wear masks continuously, but the problem is people do not wear their masks at home. Part of the reasons Covid spreads so virulently, especially in large, low/average-IQ households and communities, is because a single affected carrier will infect the entire family. East Asians and wealthier, smarter people, in general, tend to have smaller households than other ethnicities and low/average-IQ groups. This explains why Covid is still so prevalent in Turkey , Italy, and Spain despite masks being mandatory and mask-wearing being so commonplace in public, because people are infecting each other at church, in the Mosques, or at home.

Second error: Missing the Nonlinearity of the Risk of Infection
The error is to think that if I reduce the exposure to the virus by, say, ½, I would then reduce the risk, expressed as probability of infection, by ½ as well. Not quite.
Now consider (Fig 1) that probability must follow a nonlinear dose-response, an “S curve”. In the convex part of the curve, gains are disproportionately large: a reduction of x% of viral exposure leads to a drop of much more than x in risk of infection. And, patently we are in the convex part of the curve. For example, to use the case above, a reduction of viral load by 75% for a short exposure could reduce the probability of infection by 95% or more!

Taleb tries to silence skeptics with his self-proclaimed mathematical brilliance, without explaining how he arrives at his conclusions. He does not explain how the s-curve is derived, nor does the illustration he provide visually resemble an ‘s’ in any way. So, not helpful at all. Useless.

The s-cure has to do with the probability of contracting the virus. I will attempt to recreate what Taleb possibly had in mind but so poorly explained:

probability of infection = 1-(1-p)^d

p=probability of infection with a single dose

d= number of doses

Say, for example, that each exposure introduces a 1/10 probability of being infected. The probability of being infected after d-number of doses or repeated exposures is plotted.

40 doses corresponds to a 90% probability of infection. Cutting this in half as Taleb suggests, only reduces the probability to 65%, so unless I am misunderstanding him, he is wrong to say that is falls by more than half. Moreover, the curve resembles a semicircle instead of an S.

There are other derivations that give different shapes and probabilities, but Because Taleb provides no derivation or sources, it is hard to know how he arrived at his numbers.

“There is no evidence that masks work”, I kept hearing repeated to me by the usual idiots calling themselves “evidence based” scientists. The point is that there is no evidence that locking the door tonight will prevent me from being burglarized. But everything that may block transmission could help. Unlike school, real life is not about certainties. When in doubt, use what protection you can. Some invoked the flawed rationalization that masks induce false confidence: in fact there is a strong argument that masks makes one more alert to the risks and more conservative in behavior.

Again, another false equivalence and misconstruction of the anti-mask position. Masks and other measures are mandatory; locking one’s door to prevent burglaries is an individual choice. But why stop at masks? If the goal to to prevent the risk of infection at any cost, then this justifies the total erasure of individual rights in the name of public safety.

Paternalistic bureaucrats resisted inviting the general public to use masks on grounds that the supply was limited and would be needed by health professionals — hence they lied to us saying “masks are not effective”. They did not get the inventiveness and industriousness of people who do not need a government to produce masks for them: they can rapidly convert about anything into well-functioning protective face covering appendages, say rags into which one can stitch coffee filters… about anything. Nor did bureaucrats heed the notion of markets and the existence of opportunists who can supply people with what they want.

If you want to wear a rag or toilet paper around your face, no one it stopping you. Everyone keeps repeating this ‘Facci lied’ canard to suit their political agendas: both the right and the left do this. It was only beefily and in the early stages of the pandemic that masks were discouraged; soon after, the supply of masks quickly increased to meet demand, and masks were made mandatory in some states.

We have a) the salon story where two infected stylists failed to infect all their 140 clients (making the probability of infection for bilateral mask wearing safely below 1% for a salon-style exposure) — we know the probability of infection for non mask wearers from tens of thousands of data points and the various R0 estimations) plus b) the rate of infection of countries where masks were mandatory, plus c) tons of papers with more or less flawed methodologies, etc.

Again, going back to item #1, masks may only work if people wear them all the time, including at home. Or if people have small households and minimal contact with each other. Countries such as Turkey have mandatory mask wearing, yet the daily case and death count is at new highs, so this is evidence against the purported efficacy Talab’s ‘bilateral mask theory’.

“Libertarians” (in brackets) are resisting mask wearing on grounds that it constrains their freedom. Yet the entire concept of liberty lies in the Non-Aggression Principle, the equivalent of the Silver Rule: do not harm others; they in turn should not harm you. Even more insulting is the demand by pseudolibertarians that Costco should banned from forcing customers to wear mask — but libertarianism allows you to set the rules on your own property. Costco should be able to force visitors to wear pink shirts and purple glasses if they wished.

Again with the misconstructions. No ‘pseudolibertarians’ has ever said that Costco should not be allowed to require its customers wear masks, as it is Costco’s choice being that it is a private company; rather, libertarians and conservatives are opposed to the governments forcing individuals and businesses to require masks. So Taleb is too intellectually dishonest to even get that right.

Taleb (aka angry Arab man) uses the same “it’s a private company” argument that the left does in justifying tech censorship against conservatives, or anyone who makes fun of him (which is not that hard to do given how pathetic and lame his arguments and PDFs are). This is why Taleb supported the de-platforming of Stefan Molyneux a few months ago from Twitter and YouTube, because his followers were saying mean things about him so therefore he should not be allowed to use Twitter.