HELPFUL PROBABILITY FRAMEWORKS AND MISCONCEPTIONS | Home

Punchline #1: Stop comparing Covid-19 with car deaths… 

Nassim Nicholas Taleb & Yaneer Bar-Yam explained this point very, very brilliantly: Sure, the risk of dying of Coronavirus is much lower than that of many diseases and even than being hit by a car, but one should not compare Coronavirus with the other incidents because Covid-19 is systemic and the others are idiosyncratic events. This means that if individuals don’t act against their own probabilistic interests (which is to ignore the virus), they will actually help Covid-19 spread, and the more vulnerable others will die. 

Punchline #2: It may not be rational for individuals to panic, but it’s rational for the society to panic. 

For Covid-19, we are dealing with “an extreme fat-tailed process owing to an increased connectivity, which increases the spreading in a nonlinear way. Fat tailed processes have special attributes, making conventional risk-management approaches inadequate,” wrote Taleb, Bar-Yam, and Joseph Norman back in January as they warned the public. “While repeated risks can be taken by individuals with a limited life expectancy, ruin exposures must never be taken at the systemic and collective level.” In other words, it may not be rational for individuals to panic, but it’s absolutely rational for the society to panic. 

Punchline #3: Naive empiricism and pseudo-evidence are dangerous for our society’s thinking. 

Taleb calls it “naive empiricism” to just throw out these seemingly evidence-based comparisons between Covid-19 and others, which is exactly the mistake we made during the Ebola crisis and time and again… Basic probability theory tells us that we should not compare raw data that don’t have the same variances! Something that is exponentially growing like Covid-19 cannot be compared with something static like the heart disease! Sure, you can compare them at face value to see which is killing more Americans at the moment, but you should not proceed further in deriving more conclusions regarding how we may need to proactively respond or not respond to Covid-19. 

“Learning a bit of statistics [but not the deeper theoretical caveats behind] is very dangerous to society,” as Taleb explains the danger of pseudo-evidence.