Should implied volatility be higher than historical volatility?

Usually, yes. Option sellers demand a premium over expected realized movement for bearing open-ended risk — the volatility risk premium. IV moderately above HV is the normal state; the informative signals are unusually wide gaps or the rarer inversion where IV drops below HV.

What does it mean when IV is lower than HV?

Options are pricing less future movement than the stock has actually been delivering. If the stock's behavior persists, long-option strategies are buying volatility at a discount. It can also mean the market expects a recent turbulent stretch to calm down.

What window is historical volatility measured over?

Any window, but 30, 60, and 90 trading days are standard. Shorter windows react faster but are noisier; comparing IV against several windows at once gives a more reliable read.

Historical volatility vs implied volatility

One is the record, the other is the forecast. The spread between them is where the edge lives.

Historical volatility (HV) — also called realized volatility — measures how much a stock actually moved over a past window (commonly 30 days), computed from its daily returns and annualized. Implied volatility (IV) is the market's forward-looking forecast, extracted from option prices. HV is a fact about the past; IV is the price of an opinion about the future. Comparing them tells you whether options are pricing more or less movement than the stock has been delivering.

How each is measured

HV is pure arithmetic: take daily log returns over the window, compute their standard deviation, annualize by √252. A 30-day HV of 25% means that over the last month, the stock's day-to-day swings were consistent with a ±25% annual range. There's no model or opinion in it — just the tape. IV, by contrast, is solved backward from option prices through a pricing model (see our IV explainer): it moves with supply and demand for options, not with the stock's actual behavior.

The normal state: IV above HV

Across most stocks, most of the time, IV trades moderately above trailing HV. This is the volatility risk premium: option sellers take on open-ended, convex risk and demand compensation for it, exactly as insurers charge more than expected losses. A stock realizing 24% with options implying 28% is normal. This persistent spread is why systematically selling fairly-priced options has historically been profitable — and why occasionally it costs the sellers a year of profits in a week.

Reading the gap

The information is in the extremes. IV far above HV (say 1.4× or more, with no event pending) means the market is paying for drama the stock isn't producing — premium sellers have the statistical wind at their backs, and long-option buyers need a regime change to win. IV below HV is rarer and more interesting: options are priced for calm while the stock is demonstrably moving more than that. Buyers of straddles and gamma get paid in that state if the movement simply continues. Always check the calendar first — a pending earnings date makes a wide IV-HV gap rational rather than exploitable (see IV crush).

A worked example

AMD realizes 42% over the past 30 days. Its front-month options imply 38%. That inversion says the option market expects the recent turbulence to fade — but you're being offered movement at a discount to the current run rate. If you believe the chop continues, long premium is statistically cheap. Conversely, if AMD realized 30% and options implied 50% with earnings three weeks away, most of that gap is event premium, not mispricing.

Common mistakes

Using one HV window. 30-day HV after a wild week overstates the norm; 90-day smooths it. Compare IV against 30/60/90-day HV together before concluding anything.

Expecting the gap to close instantly. IV can stay above HV for quarters. The spread is a statistical tilt, not a timing signal — combine it with IV Rank/Percentile for the timing dimension.

Ignoring why the gap exists. Pending binary events, low float, hard-to-borrow shares — sometimes "expensive" IV is correctly priced for a risk HV can't see yet.

Try it live

The numbers below are real — pulled from OptionScope's volatility dataset (CBOE IV plus nightly stats covering ~2,300 optionable tickers). Look up any symbol:

Explore the full OptionScope toolkit — free →