What Is Skewness in Finance? (Nassim Taleb's Definition)
Skewness is a measure of the asymmetry of a probability distribution around its mean. A symmetric distribution (like the normal distribution) has skewness of zero — its tails extend equally in both directions. A negatively skewed distribution has a longer left tail (rare, large losses). A positively skewed distribution has a longer right tail (rare, large gains).
Why Skewness Matters More Than Win Rate
In a symmetric distribution, the most common outcome (the mode) and the average outcome (the expected value) are roughly the same. A strategy that wins 60% of the time is probably profitable.
In a skewed distribution, these diverge wildly. A strategy can win 99% of the time (high mode) and still have a negative expected value (low average) if the 1% loss events are large enough.
Taleb's example: 999/1,000 chance of $1 gain, 1/1,000 chance of $10,000 loss. Expected value: −$9. Win rate: 99.9%. The strategy is a money-loser disguised as a money-maker by its win rate.
Negative Skewness in Real Strategies
Most strategies that appear to produce smooth, consistent returns are negatively skewed:
- Short-volatility strategies collect small premiums for being exposed to large market moves
- Yield-enhancement products offer above-market income by taking on implicit tail risk
- Carry trades profit from steady rate differentials until a currency or rate dislocation
These strategies win frequently and lose rarely. But the rare losses are large enough to wipe out the accumulated frequent wins — making the expected value negative despite the appealing track record during the calm period.
The diagnostic: any strategy with returns that are suspiciously smooth is probably negatively skewed. Ask what the strategy is implicitly short of — what rare, large event would produce the catastrophic loss that the steady returns are implicitly selling insurance against.
For the full framework, read Fooled by Randomness: How Luck Masquerades as Skill.