Variance explained

What Variance measures

Variance is standard deviation's underlying building block — literally standard deviation squared. It measures the same thing (dispersion of price around its own average) but in squared units rather than the original price units, which makes it mathematically convenient for certain statistical and portfolio-theory calculations even though it's less directly interpretable at a glance than standard deviation.

How Variance readings are interpreted

Higher variance means greater price dispersion around the average — the same underlying signal as standard deviation, just expressed on a squared scale. Because squaring amplifies larger deviations disproportionately more than smaller ones, variance is particularly sensitive to the presence of a few large outlier moves within the lookback window.

Conventional Variance levels

• Rising variance: volatility expanding (same read as rising standard deviation, on a squared scale)
• Falling variance: volatility contracting
• Because of the squaring, variance changes are more dramatic in percentage terms than the equivalent standard deviation change — a doubling of standard deviation is a quadrupling of variance

Where Variance works, and where it does not

Functionally redundant with standard deviation for most practical trading purposes, since standard deviation is simply its square root and is more directly interpretable in price terms. Variance's real value shows up in portfolio-level statistical work (e.g. combining variances across multiple holdings) rather than single-security technical analysis.

Indicators commonly read alongside Variance

• Standard Deviation: Directly derived from each other — for single-security technical analysis, standard deviation is generally the more intuitive of the two to actually look at
• Portfolio construction: Variance (and covariance) are the standard building blocks of formal portfolio risk models, more so than in single-ticker technical screening

Known limitations of Variance

• Squared units make it less directly interpretable at a glance than standard deviation — a variance of 25 doesn't intuitively tell you "5 points" the way standard deviation does
• Highly sensitive to a small number of large outlier moves, which can make it swing more dramatically than the underlying price behavior might suggest
• For single-security technical screening, offers little beyond what standard deviation already provides

Variance in practice

• For everyday single-security technical analysis, prefer standard deviation — it's the more directly interpretable version of the same information
• Variance's practical value is greatest in portfolio-level statistical work, less so in individual ticker screening
• Included here primarily for completeness and for users doing more formal quantitative analysis who specifically want the squared form
• If unsure which to use, use Standard Deviation instead