, 6 tweets, 3 min read
One comment about masks and nonlinearities that these imbeciles are not getting.

Reducing exposure to viruses by 30% thanks to an "imperfect" mask does not mean reducing risk of contracting the disease by just 30%. By convexity, it must be more than 30%, can even be 95%.
The proof is in
arxiv.org/pdf/1808.00065…
That's for an individual. If everyone wore *VERY BAD* masks, I guess the number of deaths would drop by a yuuuge factor, maybe 90%. Why? p is probability of infection, look at 1-p^n because everyone is reducing.

For 2 pple masks act a 2-way filter, 1-p^2. Yuuge.
We need to get the probability of getting sick at different concentration of droplets.
We could probably back it up (or get a handle on the convexity) from country data (Czech republic).
Intuition: reduction of 70% of particles is yuuuge: home masks should work. Plus if 2-way!
We have a hint: @dzviovich just sent me something useful: comparison of US vs CZ w/a single sign. difference: edict to wear maks.
We know masks work (from other combined statistical signals). Now let's back up the convexity of probability of infection to particle concentration.
Friends, another robust question to get the convexity.

How many people contracted COVID while wearing surgical masks? How many w/N95? How many with cowboy scarves?
(Assume for now homogeneity. We are looking for convexity, not measurement).
Missing some Tweet in this thread? You can try to force a refresh.

Keep Current with Nassim Nicholas Taleb

Stay in touch and get notified when new unrolls are available from this author!

This Thread may be Removed Anytime!

Twitter may remove this content at anytime, convert it as a PDF, save and print for later use!

# Try unrolling a thread yourself!

2) Go to a Twitter thread (series of Tweets by the same owner) and mention us with a keyword "unroll" `@threadreaderapp unroll`