Exploring Extremistan, Fat Tails, and Anti-fragility
Compare tail probabilities to understand black swan frequency
In normal distribution, a 6σ event is almost impossible (~1 in a billion). But in power law distribution, extreme events (black swans) occur frequently, and a single extreme event can outweigh all ordinary events combined.
A turkey is fed for 1000 days, every day "proving" humans are friendly
Past data cannot predict black swan events. The turkey built a false "humans are friendly" model from 1000 days of observation. In Extremistan, we need anti-fragile systems, not reliance on historical predictions.
Observe how power law exponent affects wealth concentration
Wealth follows a power law, typical of Extremistan. Average wealth is meaningless (Bill Gates walks into a bar, average becomes billion, median unchanged). In Extremistan, mean doesn't represent typical cases.
90% extremely safe + 10% extremely risky, avoid the middle
The barbell strategy preserves limited downside while maintaining upside exposure to positive black swans. Avoid "middle" strategies - in Extremistan, medium risk often hides enormous tail risks.
| Feature | Mediocrestan | Extremistan |
|---|---|---|
| Individual Impact | Doesn't affect whole | Dominates whole |
| Distribution Type | Normal Distribution | Power Law Distribution |
| Typical Examples | Height, weight | Wealth, sales |
| Predictability | History predictable | Black swans frequent |
Anti-fragile is more than resilient (resisting shocks) - it benefits from shocks. Like bones strengthening under pressure, muscles growing after tearing.