On This Page
Key Takeaways
- A volatility cone compares historical realized-volatility percentiles across rolling window lengths.
- Put trading-day horizon on the horizontal axis and annualized volatility on the vertical axis.
- Align each option expiration with the trading-day horizon used for the realized comparison.
- The worked 30-trading-day point has IV 31.5%, median realized volatility 21.5% and upper quartile 27.9%.
- Use the term structure, event calendar and selected strike premiums to turn that comparison into a trade.
What is a volatility cone?
A volatility cone is a chart of historical realized-volatility percentiles across rolling window lengths, used to judge whether option implied volatility looks high or low at each horizon. Trading-day horizon runs horizontally and annualized volatility runs vertically, so each expiration is compared with the movement the underlying has actually delivered over windows of the same length.
How is a volatility cone built?
At each horizon, calculate volatility for every rolling window in the dataset. Plot the selected percentiles and connect them across horizons. Overlay option IV at matching horizons to compare pricing with the underlying’s movement history.
Keep the same return estimator and annualization when comparing realized windows. For this example, use the sample standard deviation of daily log returns with its n - 1 denominator, annualized by √252.
Formula
RV_n(t) = √(252) × sample standard deviation of the last n daily log returns, and the cone band at horizon n is the chosen percentile of every rolling RV_n value in the history- n is the lookback length in trading days.
- Use daily log returns, subtract the sample mean and divide the sum of squared deviations by n − 1.
- Annualize the sample standard deviation by √252.
- At each horizon, calculate the selected percentiles across all rolling windows.
- The worked chart shows minimum, median, 75th percentile and maximum realized volatility, plus option IV.
- Map each option’s expiration to the matching trading-day count using the exchange calendar.
Volatility cone example
Use a worked XYZ dataset with 756 daily log-return observations. The specified rolling-volatility percentiles below form the cone. Overlay at-the-money option IV at matching trading-day horizons.
- 30-trading-day window count: 756 − 30 + 1 = 727
- 30-day realized percentiles: minimum 11.4%, median 21.5%, upper quartile 27.9%, maximum 58.2%
- 90-trading-day window count: 756 − 90 + 1 = 667
- 90-day realized percentiles: minimum 14.6%, median 22.3%, upper quartile 26.1%, maximum 41.0%
- 30-day option IV of 31.5% is 10.0 volatility points above the median and 3.6 above the upper quartile
- 90-day option IV of 25.0% is 2.7 points above the median and 1.1 below the upper quartile
- Maximum-minus-minimum band width: 46.8 points at 30 days and 26.4 at 90 days
The worked IV curve stands above the upper quartile at the 30-trading-day horizon and below it at 90 trading days. Compare the expirations and events within those horizons, then price the structures that express your view.
How to match option expirations to the cone
Match the selected option’s remaining exchange sessions with the cone’s trading-day horizon. Using the annual 252-to-365 conversion, 30 trading days corresponds to 43.5 calendar days and 90 corresponds to 130.4. Use the actual exchange calendar for the selected expiration.
How traders use a volatility cone
Compare the term structure and the events contained in each expiration. Inspect the selected strikes to see their IV and premiums. A calendar spread or diagonal spread connects two expirations; same-expiration spreads compare strikes within one horizon.
Check options flow in those contracts, compare entry cost, breakevens and price-and-IV scenarios in the options profit calculator, and record the selected horizon and entry inputs for evaluating the trade afterward.
Limitations of the volatility cone
- Compare current realized volatility with the bands. The cone shows the range realized volatility has covered, and a new regime can push it outside every historical band.
- Fix the inputs before you compare cones. The percentiles depend on the history length, the return estimator, the annualization and whether event days are included, and a different choice moves the bands.
- Use a long history for long horizons. Rolling windows overlap heavily: 756 returns produce 727 windows of 30 days, so the percentiles rest on far fewer independent observations than the window count suggests, most of all at long horizons.
- Check skew before you price a strike away from the money. The worked overlay uses at-the-money IV and does not show skew across strikes.
- Before calling an option expensive, check the catalyst calendar: an IV reading above the historical band can be pricing a scheduled event. An earnings report or other event inside the expiration can justify it.
Bottom line
A volatility cone shows whether option implied volatility looks high or low against the realized volatility the underlying has delivered over each horizon. Match each expiration to its trading-day horizon, keep the estimator consistent, and check the event calendar before treating a reading above or below the bands as a trade.
Clear answers
Frequently Asked Questions
Sources and Further Reading
- Burghardt and Lane, How to Tell If Options Are Cheap, Journal of Portfolio Management 16(2), 1990
- OIC: Understanding Volatility and Options Skew
- OIC: Volatility and the Greeks
- NIST/SEMATECH e-Handbook of Statistical Methods: Measures of Scale
How entries are sourced, calculated and updated: editorial methodology.