Summary:
Nicholas Nassim Taleb at his best. A little angry on the state of the world, and knowing that he often knows better. To me the most memorable idea from this book is the allegory of the Turkey. It's a story about how a turkey tries to predict the future based on past data and fails. We are the turkeys.
At the beginning the turkey is happy and forecasting: "Oh wow, I am getting so much food, the times ahead are going to be epic". Until one day, all of a sudden, the turkey get's slaughtered and prepared for Thanksgiving. We are the turkeys. This is the central idea from the book in a nutshell.
Black Swan events happen and their consequences are enormous – often life-changing or even life-threatening. We have a blind spot for them though, because we like to see patterns and our brains are not very attuned to this idea of extremely powerful outlier events. Again. We are the turkeys, we use predictive models in the wrong way, and are extremely surprised when extremely bad things happen.
Main Takeaways:
Don't be the turkey. Remember that there are two different "worlds" out there: Extremistan and Mediocristan.
Extremistan is where the "fat tail" lives. The 1 in a million outlier that has more weight on the average than the rest combined. Wars are extremistan events, so are financial markets (and their crashes), or the amount of people reading a blog, the distribution of people across social networks and many many other things.
In Extremistan, inequalities are such that one single observation can disproportionately impact the aggregate, or the total.
When you sample Elon Musk by accident in your population statistics, you are going to heavily skew that average.
That being said, there are also a lot of things that are not in the domain of Extremistan. In fact, most of the things we are used to from an evolutionary point of view don't fall into the category of Extremistan.
They don't exhibit black swan style dynamics. The height of people, the amount of harvest we do every year, the amount of children people get, the amount of friends we have, how much we weigh or eat etc. This is probably why we have such a hard time grasping these kinds of concepts.
Because of this distinction into the two "countries" we have to look out for signs with which of the two we are dealing. And then adapt our behavior accordingly.
In Extremistan, extreme events happen. And they happen more often than you think, and their consequences can be more desastrous than you think. You have to prepare for that.
If you try to use the wrong model, while in Extremistan, your predictions are just like that of the turkey. In the worst case they can absolutely cost you your life.
There is this cartoon of how averages can hide something dangerous in plain sight, which comes to mind when thinking about these topics.
Talebs point is, that if you base your behavior on averages, in a domain where things are not normally distributed, your predictions are going to be devastatingly wrong as soon as the first black swan outlier event happens. You can drown in that 3ft on average deep river.
And Taleb's critique is, that the whole financial world operates like the turkey in the story. Economists regularly use normal distributions to judge how what is in reality a fat tailed, fractal distribution, is going to behave.
And then they routinely get surprised and wiped out by financial disasters that according to their models were so unlikely that they shouldn't have happened in the lifetime of the universe. And yet they did.
These are the ideas of Black Swan in a nutshell.
How systems could weather these kinds of events, and even become stronger as a result, is presented in another book of Taleb – Antifragile.