Historical Volatility

Historical volatility measures how much a stock, index, or other asset's price has actually swung around over some past period. It is a backward-looking statistic: you take a stretch of daily (or hourly, or weekly) closing prices, calculate the returns from one period to the next, and then measure how spread out those returns are. That spread is usually expressed as a standard deviation, then annualized so different time periods can be compared on the same scale, often shown as a percentage.

In practice this means a stock that moves 3% in either direction most days will show a much higher historical volatility number than one that creeps along at 0.3% a day, even if both stocks end the year at the same price. Volatility says nothing about direction — a stock crashing 5% a day and one rallying 5% a day can have identical historical volatility.

The nuance that trips people up is the name itself: historical volatility only describes what already happened. It is not a prediction. Traders sometimes assume a stock that was calm last month will stay calm, but volatility tends to cluster and can shift quickly around earnings, news, or broad market stress. This is different from implied volatility, which is a forward-looking figure backed out of option prices and reflects what the market currently expects, not what has already occurred.

Because it's just arithmetic on past prices, historical volatility is objective and easy to calculate, but the number you get depends heavily on your choices: how many days you look back, whether you use closing prices or intraday highs and lows, and how you annualize the result. Two people can compute "the historical volatility" of the same stock and get different numbers if their lookback windows differ.