All toolsVolatility

Realized Volatility Chart

Track how turbulent a market has actually been, annualised over a rolling window.

What is realized volatility?

Realized volatility measures how much an asset's price has actually fluctuated over a past window, unlike implied volatility, which is the market's forecast embedded in option prices. It is computed as the standard deviation of returns over the window, then annualised — typically by multiplying by the square root of 252 trading days. Rolling it forward produces a line that rises in turbulent regimes and falls in calm ones, quantifying realised risk directly from price history.

21-Day Rolling Realized Volatility (annualised)time-series

Illustrative 21-day rolling realized volatility, annualised, from a synthetic price path. Data for demonstration only.

Volatility comes in two flavours: what the market expects (implied) and what actually happened (realized). Realized volatility looks backward, measuring the size of the moves a price has genuinely delivered. This tool computes a 21-day rolling realized vol, annualised, so you can watch calm and turbulent regimes trade places over time.

How realized vol is calculated

Realized volatility starts from returns — usually the log returns between consecutive closes — over a chosen window such as 21 trading days, roughly one month. You take the standard deviation of those returns, which measures their typical dispersion, then annualise by multiplying by the square root of the number of trading periods in a year, 252 for daily data. The result is a single annual-equivalent percentage that makes windows of different lengths comparable.

Read more
The annualisation factor is where errors creep in. The square-root-of-time rule assumes returns are independent and identically distributed; it multiplies daily volatility by the square root of 252, weekly by the square root of 52, and so on. Mixing frequencies — annualising daily volatility with a monthly factor — produces nonsense. Using log returns rather than simple returns keeps the maths clean and additive across periods.

Realized versus implied volatility

Realized and implied volatility together form one of the most watched relationships in derivatives. Implied vol is the market's forward-looking price of risk; realized vol is the outcome. When implied sits persistently above realized, option sellers are being paid a risk premium; when realized suddenly exceeds implied, the market has been caught off guard and option buyers profit. The spread between the two, the variance risk premium, is itself a tradeable and closely studied quantity.

How Quadesto computes it

Quadesto computes rolling realized volatility from your price column: it derives log returns, takes their standard deviation over a window you set, and annualises with the square-root-of-time rule for your data's frequency. You choose the window length and can overlay implied vol for comparison. The rolling series above is built from a synthetic price path, and the chart embeds into volatility research.

Build this with your own data

Upload a CSV or connect a live source, and Quadesto renders this exact chart — styled, computed, and embeddable in your reports and newsletters. Free to start.

Create free account

Related tools

Volatility Smile Chart

Open tool

CBOE SKEW Index

Open tool

Bollinger Bands Chart

Open tool

Returns Distribution Chart

Open tool

Explore all free financial tools, browse the finance glossary, or see how teams use Quadesto.

Frequently asked questions

How is realized volatility calculated?
Take the returns of a price series over a window — usually log returns over about 21 trading days — compute their standard deviation, then annualise by multiplying by the square root of the number of periods per year, 252 for daily data. The result is an annual-equivalent volatility percentage.
What is the difference between realized and implied volatility?
Realized volatility measures how much price has actually moved in the past, computed from returns. Implied volatility is the market's forward-looking forecast, backed out of option prices. Realized looks backward at outcomes; implied looks forward at expectations, and the gap between them is a closely watched risk premium.
Why annualise by the square root of 252?
There are roughly 252 trading days in a year. The square-root-of-time rule scales volatility across horizons under the assumption that returns are independent, so daily volatility is annualised by multiplying by the square root of 252. This makes windows of different lengths directly comparable on one scale.
What window should I use for realized volatility?
It depends on your horizon. A 21-day window captures roughly one month and reacts quickly; longer windows like 63 or 252 days are smoother and slower. Shorter windows show regime shifts sooner but are noisier. Many analysts plot several windows to see both fast and slow signals.
Can I compute realized volatility on my own data?
Yes. Upload a price series to Quadesto, choose a rolling window, and the engine derives log returns, takes their standard deviation and annualises with the square-root-of-time rule. Overlay implied volatility to compare, then embed the chart in volatility research.