── ── Strategy
Black Swan
A Black Swan, in Nassim Taleb's framing, is an event with three traits: it is a rare outlier outside normal expectations, it carries extreme impact, and it is rationalized as predictable only in hindsight. Because such events dominate history yet resist forecasting, the rational response is not to predict them but to build for the tails you cannot foresee.
How it works
Recognize that in many domains, outcomes are dominated by a few extreme events rather than the everyday average, so models built on the normal range systematically miss what matters most. Forecasting harder doesn't help, because the defining trait of a Black Swan is that it sits outside the model that's making the forecast.
Shift effort from prediction to exposure. Ask not 'what will happen' but 'what happens to me if the unthinkable does,' then arrange your positions so that negative Black Swans can't ruin you and positive ones can benefit you. Manage the consequences you can control, not the events you can't.
When to use it
- Stress-testing whether a single rare event could destroy the company
- Resisting false confidence in forecasts and projections
- Building cash buffers and redundancy against unforeseeable shocks
- Deciding how much fragility to tolerate in a critical dependency
When not to use it
For routine, well-bounded decisions in stable systems, where invoking tail catastrophe just paralyzes ordinary action with improbable fear.
Worked example
The 2008 financial crisis
Before 2008, risk models treated a system-wide collapse in U.S. housing and mortgage-backed securities as so improbable it was effectively ignored. When it happened, the impact was extreme and global, and afterward countless analysts explained exactly why it was 'obvious' all along. It fit every Black Swan trait: unexpected, enormous, and retrofitted with hindsight.
Why it matters for founders
Founders build detailed forecasts and then bet the company on them, quietly assuming the future stays inside the spreadsheet. The discipline isn't predicting the shock; it's making sure one unforeseeable event can't end the story while leaving room for a lucky break to help. deciqAI's agents weigh tail exposure before acting, so plans survive the surprise the model never saw.
Install this skill (free, MIT)
npx skills add deciqAI/knowledge-skillsFAQ
Can you predict Black Swans if you try hard enough?
By definition, no; their hallmark is falling outside the models doing the predicting. The useful work is reducing your vulnerability to bad ones and increasing your exposure to good ones.
How is this different from a worst-case scenario?
A worst-case scenario is something you can imagine and bound. A Black Swan is precisely what you failed to imagine, which is why preparation focuses on robustness to the unknown rather than a specific named risk.