Ignore this post.
distillation of Taleb's core idea:
expected value estimates are dominated by tail events (unless the distribution is thin-tailed)
repeated sampling from a distribution usually does not yield information about tail events
therefore repeated sampling can be used to estimate EVs iff the distribution is thin-tailed according to your priors
if the distribution is fat-tailed according to your priors, how to determine EV?
estimating EV is much harder
some will say to use the sample mean as EV anyway, they are wrong
in the absence of information on tail events (which is v common)
you must judge the fatness of the left/right tails based on your priors
right tail is fat and left is thin: high EV
left tail is fat and right is thin: low EV (/ high magnitude in the negative direction)
both tails are fat: EV is highly uncertain
when calculating EV of a distribution that you have samples from:
the burden of proof is on you to show that you are operating in a thin-tailed domain
before you use the sample mean as an estimate of EV
else, you must rely on your prior beliefs on the fatness of the tails of the distribution
If Guyenet is right that olfactory/taste signals are critical to the maintenance of obesity, then we should expect people who take their meals exclusively through feeding tubes to be obese at rates well below baseline.