You cannot read two paragraphs by Nassim Taleb without running into something utterly idiotic.  Come to think of it, it's a lot like flipping channels on TV.  Maybe that is why he is so popular.  But I have yet to hear anyone quote anything of substance from Nassim Taleb, other than his life's work that can be compressed into one line, the realization that some successful money managers may simply have gotten lucky.

Here is Nassim's latest nonsense:

"Now, it is trivially easy to measure how much of this excess confidence there is in the human race. You even can do it at the dinner table of the next extended family gathering, if it is sufficiently large. Take room full of people. Randomly pick a number. It could be anything: the proportion of psychopathic stockbrokers in Western Ukraine, the sales of this book during the months with "r" in them, the average IQ score of business book editors (or business writers), the number of lovers of Catherine II of Russia, etc. Ask the members of the audience to each independently estimate a range set in such away that they believe that they have 98% chance of being right, and less than 2% chance of being wrong."

By requesting people put numbers to something they have no idea about, Nassim means to prove that people are over-confident.  They are prone to assigning known distributions to things which are unknown.  But in fact, it is Nassim himself - not a bunch of drunks at a dinner table - who spent the first half of his life imagining he knew the 99% confidence interval.  So he tells the people "Do what I did wrong."  And then, even though the guests may never have done it before or again in their lives (except to be polite and humor their oddball guest), he points at them and says "See, people are stupid!"

Even worse, Nassim has not demonstrated, as varied and contradictory as their answers may be, that any of them are wrong!  We can consider a trivial example to understand Nassim's error. Suppose Nassim is playing no-limit hold'em poker with $5 blinds, no antes, and he raises a million dollars all-in with 23 offsuit.  Nassim's opponent will be 99% confident Nassim has AA, AK, KK, or some hand other than what he has.  Over a lifetime of playing poker, this "over-confident" opponent will be right in this situation 99% of the time.  At the same table, there could be another guy who has met Nassim before, and knows what is going on.  Given his different knowledge, he could give a totally different set of hands from the first opponent - for whom Nassim is a stranger - and also be right.

Each individual may have a 99% chance of being right, if you consider his question to be a member of a class of challenges, where Nassim's is the only one they get wrong.  What time will my husband get home from work?  Between 6:00 and 9:00.  How many days of rain will we get this month?  2 to 20.  How many lovers did Catherine II have?  20-30.  People's lives are designed in such a way, that they are confronted with questions and challenges which they can handle.  In other words, while Nassim has selected a question at random, the questions which people ordinarily choose to face are selected for their ability to answer them.  Except when acting as voters - where they are forced to make decisions involving dynamics where they have no idea what the far-reaching effects will be - people pay a cost for being wrong, in 99.99% of the guesses they make.

What is my chance of getting hit by a car when I look both ways?  I don't think in distributions or apply numbers to things, but I know the chance is low enough - almost nothing!  Nassim could get a chimpanzee to take a piss in a Coke can, and then walk into a room pretending to be drinking from it.  He could then ask people "Give me the list of things which you are 99% confident are in this can."  All he has done, is tricked them into thinking a unique circumstance is an ordinary circumstance, by disguising it as an ordinary circumstance.  Ordinarily, people are not confronted with random questions that have no relevance in their lives.  If this were the case, they would all be dead.  Put differently, Nassim is the black swan at the dinner table.

The only thing Nassim has really discovered, is that situations in life are more varied than people's ability to discriminate between them.  A person used to crossing the street in New York, may instinctively look left when he crosses the street in London, and get hit by a car.  There are innumerable situations where a person thinks he has recognized a situation as something he is experienced with, or imagines his experience is complete.  But such situations are by far the exception, or else we would all be dead.  People's confidence every time they take a step on the sidewalk, is well placed.  To stop walking based on the possibility that you could be dreaming you are walking, and actually be standing on a lightpost, is folly.

For that small handful of pitfalls in life, thankfully we have parents and morals.  We enjoy instructions such as "don't drive over 100 mph," or "don't have screw hookers at the crack house."  Even though never before in our lives have we wrecked a car, or contracted AIDS.  People are not, in fact, disposed to betting their lives on a guess about how many lovers Catherine II had, it has never been a common human weakness.  People's over-confidence, by the very mechanism through which they arrive at confidence, can only be in regard to isolated situations.  There is not a naturally-evolving generalized over-confidence.  There are only contrived situations, which happen once and are very quickly recognized and corrected.  E.g. don't answer dumb questions from Nassim.

As we go through our days, we have no choice but to hope that someone is not setting a trap for us!  And 99% of dinner guests will not be out to prove we are dumb.  99% of guests seek to flatter, or to turn our answer into a funny joke or story of some kind, which is why we play along.  Nassim has not shown that people will not be right, 99% of the time they choose a 99% confidence interval.  He has not even shown that people have a habit of specifying confidence intervals.