Historical vs Implied Volatility

Historical volatility measures past variation. Implied volatility is an option-model input inferred from a market price.

Two different inputs

A realised estimate commonly takes the sample standard deviation of daily log returns and annualises it using √252 for trading sessions. The chosen window, return definition and annualisation convention affect the result.

Implied volatility is the volatility input that reproduces an option’s observed price within a specified pricing model. It depends on strike, expiry, quotes and model assumptions; a single asset does not have only one implied-volatility number.

Worked comparison

Suppose a hypothetical asset has 20% annualised realised volatility over its past 20 sessions, while a selected 30-day option implies 28%. The difference is eight percentage points. It does not establish an eight-point forecasting error: the periods, instruments and pricing measures differ.

To evaluate a forecast, record the option quote at the start, its exact future measurement period, and the subsequently realised value. Repeating this across a specified sample is more informative than comparing today’s IV with yesterday’s backward-looking volatility.

Why the gap is not an automatic trade

Options can price scheduled event risk and risk compensation that a backward-looking estimator misses. Skew means different strikes can imply different volatility. A strategy also faces bid-ask costs, hedging costs and tail exposure. A high IV-to-realised ratio alone does not demonstrate a profitable trade.

Which input Aiovel uses

Aiovel’s Gold model blends 20-session and 90-session realised estimates with an ATR range proxy. It contains no option-market volatility input. Its model/history agreement labels do not demonstrate calibration.