Black Swan events owe their name to a deceptively simple idea. People tend to think black swans can't exist, because all the swans they've ever seen were white; and with every white swan they see, their confidence in the impossibility of a black swan grows bigger. But a single black swan is enough to wipe out everything that was based on the assumption that all swans are white. This is, in essence, what happened during the 2008 subprime mortgage crisis[^1]. It was deemed impossible by many. Industry-standard models made the mortgage market look rock solid, when it was on the verge of collapse. Only a handful of outliers were able to see happening, no thanks to any model. >"You can expect blowups and explosive errors in fields where there is a penalty for simplicity." >(Taleb 2010) Overrelying on models makes one blind to the moving boundary between relevant and irrelevant information, between signal and noise. This raises a question: Can we make predictions about what we did not observe based on what was observed? Two puzzling observations are (1) how terrible we are at predicting the future and (2) how obsessed we are with predicting the future. > "All models are wrong, some are useful." *— George Box* [^1]: An excellent investigation on the topic: [Money, Power and Wall Street by Frontline](https://www.pbs.org/wgbh/frontline/documentary/money-power-wall-street/?)