Skin in the Game: Hidden Asymmetries in Daily Life

by Nassim Nicholas Taleb

Probability, statistics, and data science are principally logic fed by observations—and absence of observations. For many environments, the relevant data points are those in the extremes; these are rare by definition, and it suffices to focus on those few but big to get an idea of the story. If you want to show that a person has more than, say $10 million, all you need is to show the $50 million in his brokerage account, not, in addition, list every piece of furniture in his house, including the $500 painting in his study and the silver spoons in the pantry. So I’ve discovered, with experience, that when you buy a thick book with tons of graphs and tables used to prove a point, you should be suspicious. It means something didn’t distill right! But for the general public and those untrained in statistics, such tables appear convincing—another way to substitute the true with the complicated. For instance, the science journalist Steven Pinker played that trick with his book The Better Angels of Our Nature, which claims a decline of violence in modern human history, and attributes this to modern institutions. My collaborator Pasquale Cirillo and I, when we put his “data” under scrutiny, found out that either he didn’t understand his own numbers (actually, he didn’t), or he had a story in mind and kept adding charts, not realizing that statistics isn’t about data but distillation, rigor, and avoiding being fooled by randomness—but no matter, the general public and his state-wor...