Correlation and Portfolio Variance

Correlation measures how closely two assets move together. A correlation of +1 means they move in perfect lockstep; 0 means they move independently; -1 means they move in exactly opposite directions.
When combining two assets in a portfolio, the portfolio's risk (variance) is:
σ²(portfolio) = w₁²σ₁² + w₂²σ₂² + 2w₁w₂σ₁σ₂ρ₁₂
where ρ₁₂ is the correlation between assets. When correlation is less than +1, the last term is smaller than it would be if the assets moved together perfectly — portfolio variance is lower than the weighted average of individual variances. This is the mathematical expression of diversification's benefit.
A portfolio of 20 stocks, each with individual standard deviations of 30%, might have a portfolio standard deviation of only 15% if correlations are sufficiently low. The expected return is unchanged; the risk is halved.