Skip to content
Menu
Finabase Learn

Glossary

Normal distribution (bell curve)

The symmetric bell-shaped probability curve set by a mean and a standard deviation, which many risk models assume for returns.

On this page
  1. The 68-95-99.7 rule
  2. Where markets differ
  3. Related terms

A normal distribution is a symmetric, bell-shaped curve that describes how often values fall at each distance from their average. Two numbers define it completely: the mean, where the peak sits, and the standard deviation, which sets how wide the bell is.

Finance uses it as a default model for returns because it is easy to calculate with. Many tools assume it, from value at risk to option pricing, which applies it to log returns.

The 68-95-99.7 rule

In a normal distribution:

  • about 68% of outcomes fall within 1 standard deviation of the mean;
  • about 95% within 2;
  • about 99.7% within 3.

Where markets differ

Real returns fit the bell curve near the middle and badly in the tails.

  • Fat tails. Extreme moves happen far more often than the curve predicts. A move beyond 3 standard deviations in one direction should occur about once in 740 periods; SPY had 3 such monthly falls in 404 months from 1993 to 2026, where the curve expects 0.5. See kurtosis.
  • Lopsided tails. Large falls are more common than large gains, which is negative skewness.
  • Changing volatility. The standard deviation itself rises in crises, so one fixed bell understates risk exactly when it matters.

Models built on the normal distribution therefore tend to understate tail risk. Fat tails: why extreme market moves are more common than models say measures the gap on real data.