VaultCharts
distribution Metric

What Is Kurtosis?

Kurtosis measures the “tailedness” of the return distribution—how often extreme outcomes (large gains or losses) occur compared to a normal distribution.

Quick Answer

Kurtosis measures the “tailedness” of the return distribution—how often extreme outcomes (large gains or losses) occur compared to a normal distribution.

What Does Kurtosis Measure?

Kurtosis is the fourth moment of the return distribution. Excess kurtosis (often reported by software) is kurtosis minus 3, so that a normal distribution has excess kurtosis 0. Positive excess kurtosis means fatter tails (more extreme events than normal); negative means thinner tails. For trading and backtesting, high kurtosis often indicates tail risk: rare but large drawdowns or spikes.

Formula:
Excess Kurtosis = E[(R - μ)^4] / σ^4 - 3 (μ = mean, σ = std dev)

Typical range: Excess kurtosis: -1 to +5+; financial returns often positive

How to Interpret Kurtosis

  • 1Excess kurtosis > 0: more extreme returns than normal (fat tails)
  • 2Excess kurtosis < 0: fewer extremes than normal (thin tails)
  • 3High kurtosis strategies may have “quiet” periods then large moves
  • 4Combine with skewness to understand full shape of return distribution

How to Use Kurtosis in Backtesting & Portfolio Analysis

Assess tail risk and likelihood of extreme drawdowns
Compare distribution of returns across strategies
Validate whether returns are “normal” for risk models
Understand why Sharpe or other metrics might miss tail events

Common Mistakes to Avoid

Ignoring kurtosis when strategy has rare but severe drawdowns
Assuming normal distribution when kurtosis is high
Using small samples where kurtosis is very noisy
Focusing on kurtosis without considering skewness

Backtest with Kurtosis in VaultCharts

VaultCharts includes backtesting with built-in and custom strategies. Analyze Kurtosis, Sharpe ratio, max drawdown, and more—all with your data stored locally.

Related Backtesting Metrics