Avoid Predicting Small-Probability Payoffs
Taleb cautions against precise forecasts of rare-event payoffs when the available data cannot reveal their range or scale. Examine exposure and possible consequences rather than treating a fragile estimate as reliable.
THE IDEA
Rule: Avoid prediction of small-probability payoffs—though not necessarily of ordinary ones.
Rare outcomes can be difficult to estimate because there are few relevant observations and their scale may be poorly constrained. A model can fit ordinary events while missing the losses that matter most.
This does not make rare risks unimportant or common-event forecasts automatically reliable. Ask what evidence supports the estimate, how sensitive the decision is to an error, and whether exposure can be reduced without predicting the event precisely.
When to use it
When evaluating investment opportunities with long‐shot returns (e.g., venture capital, biotech drugs in early trials).
When designing risk management strategies for rare but catastrophic events (e.g., pandemics, financial crashes).
When allocating research time, asking which uncertain outcomes could change the decision, including rare events with large consequences.
When teaching decision analysis: emphasize the limits of forecasting and the value of robust decision rules over precise probability estimates.
FROM THEORY TO EVERYDAY LIFE
In practice
Startup Success Prediction
A venture capitalist assigns a startup a 1% chance of becoming a unicorn. Before relying on that figure, they examine the comparable companies, assumptions, and possible range of outcomes. They can also limit the amount at risk or preserve follow-on choices without pretending to know which company will become an outlier.
Insurance Underwriting
An insurer considering a rare catastrophe cannot assume that a short historical average captures the risk. It can compare scenarios, test dependence across exposures, and assess reserves or reinsurance while acknowledging uncertainty in the tail.
A MOMENT TO REFLECT
Put it to work.
Think of a recent decision where you tried to predict a low-probability outcome. How could you have redirected your analysis toward more predictable aspects?
Why might our intuition overestimate our ability to forecast rare events? Consider psychological biases like overconfidence or the narrative fallacy.
In your work or personal life, where is a rare-event payoff too uncertain to estimate precisely? What decision could remain workable across a range of outcomes?
Often confused with
CONTINUE THE CONVERSATION
Source texts
- The Black Swan The Impact of the Highly Improbable With a new section On Robustness and Fragility (Nassim Nicholas Taleb)