Historical Volatility Cone: Term Structure of Realized Vol

Key Takeaways

• A historical volatility cone shows minimum, maximum, and percentile bands of realized volatility for several maturities.
• Short maturities sit at the wide end of the cone; long maturities sit at the narrow end.
• The classical use is overlaying current implied volatility on the cone to spot mispricing.
• The most common error is comparing implied vol on a 30-day option to a 10-day realized window.

What It Is

A historical volatility cone is a fan-shaped plot. The horizontal axis is the lookback window or maturity in trading days, typically 10, 20, 30, 60, 90, 120, 180, and 252 days. The vertical axis is annualized realized volatility in percent. For each maturity, the cone displays the historical distribution of realized volatility values, usually the minimum, the maximum, and percentile bands such as the 25th and 75th.

The shape is wider on the left than on the right because short-window realized volatility fluctuates more than long-window realized volatility. Averages over longer windows are mathematically smoother, so the cone narrows toward longer maturities.

The Intuition

Saying that volatility is "high" or "low" requires a reference. A 30 percent annualized vol could be cheap for a small-cap biotech and rich for a utility. Even within one asset, 30 percent could be normal at the one-week scale but extraordinary at the one-year scale. The cone provides that reference visually for each maturity.

Option traders care about the comparison because the implied volatility quoted in option prices is itself a forecast. If a 30-day option's implied vol sits at the top of the 30-day realized vol cone, the market is pricing in more turbulence than the asset typically delivers. That can be a sell-volatility setup or a sign that the market knows something the cone does not.

How It Works

Building a historical volatility cone has four steps. First, compute daily log returns of the asset:

r_t = ln( close_t / close_t-1 )

Second, for each maturity N in trading days, compute the rolling standard deviation of returns over N bars and annualize using 252:

HV_N_t = stdev(r over last N bars)_t * sqrt(252)

Third, for each N collect the distribution of HV_N values across the historical sample. The minimum, maximum, and percentile statistics define the cone shape at that maturity.

Fourth, plot the statistics as bands across the maturity axis and optionally overlay the current implied volatility of options at the matching tenor. Burghardt and Lane recommended matching time horizons precisely. Comparing implied vol of a 30-day option to a 90-day realized vol cone is statistically meaningless.

Worked Example

Suppose you build a 10-year historical volatility cone for a large-cap index ETF. At the 10-day maturity the cone might span from 5 percent at the floor to 60 percent at the ceiling, with a 25th percentile at 8 percent, median at 13 percent, and 75th at 22 percent.

At the 90-day maturity the cone narrows. The floor might sit at 9 percent, the ceiling at 45 percent, the 25th percentile at 11 percent, median at 15 percent, and 75th at 20 percent. The same level of 30 percent vol is well within the 10-day cone but near the top of the 90-day cone.

Now you observe that 30-day at-the-money implied volatility is quoting at 16 percent. On the 30-day slice of the cone, suppose the median is 14 percent and the 75th percentile is 19 percent. Implied vol at 16 percent sits between the median and the upper quartile of historical realized vol for that horizon. By the cone framework, the option is not screaming rich and not screaming cheap. A trader would look for additional catalysts before sizing a directional volatility trade.

Common Mistakes

1. Mismatching tenors. Comparing a 30-day implied to a 10-day realized window is the classic error. Both legs of the comparison must use the same time horizon for the inference to hold.
2. Using too short a history. A two-year cone may miss every previous shock regime. Burghardt and Lane recommended at least five years and preferably more, with sensitivity checks on the window length.
3. Ignoring overlapping samples. Rolling 60-day windows produce overlapping observations that are not statistically independent. Confidence intervals computed naively will be too tight. Practitioners often subsample to non-overlapping windows for statistical tests.
4. Treating the cone as a forecast. The cone is a description of past behavior, not a prediction of the next quarter. Regime shifts can move a market outside its historical cone, and the cone will only catch up later.
5. Ignoring jumps and gaps. Asset returns are not perfectly log-normal. Cones built on plain standard deviation understate tail risk in markets prone to jumps. Some practitioners use Yang-Zhang or Parkinson estimators to capture intraday range.

Frequently Asked Questions

What is a historical volatility cone in simple terms? A historical volatility cone is a chart showing how realized volatility for an asset has typically ranged over different time windows. It narrows from short windows on the left to long windows on the right and lets you place current volatility in historical context.

How does a historical volatility cone affect investment decisions? Options traders overlay current implied volatility on the cone to judge whether option premiums are rich or cheap for that horizon. The relative position guides volatility-selling versus volatility-buying setups.

What is a real-world example of a historical volatility cone? On the S and P 500, the 10-day realized volatility cone routinely peaks above 40 percent in crisis years like 2008 and 2020, while the 252-day cone has rarely exceeded 30 percent over the same period. The shape is what defines normal versus extreme.

How can investors use a historical volatility cone effectively? Match maturities precisely, use a long enough history to include at least one stress regime, and treat the cone as a relative-value gauge rather than a forecast. Combine with options skew analysis for a fuller picture.

How is a historical volatility cone different from Bollinger Bandwidth? Bollinger Bandwidth measures the width of one set of bands on a price chart. The historical volatility cone compares realized volatility across many lookback windows in one view. The cone is a term structure; bandwidth is a point estimate.

Sources

1. Montreal Exchange. Trading Volatility Using Historical Volatility Cones (Burghardt and Lane). https://www.m-x.ca/f_publications_en/cone_vol_en.pdf
2. CME Group QuikStrike. Pricing, Volatility, and Strategy Tools. https://www.cmegroup.com/tools-information/quikstrike/pricing-volatility-strategy-tools.html
3. CME Group QuikStrike. QuikVol Tool. https://www.cmegroup.com/tools-information/quikstrike/pricing-volatility-strategy-tools/quikvol-tool.html
4. IVolatility Education. Historical Volatility. https://www.ivolatility.com/education/historical-volatility/

Disclaimer

This article is educational content only and is not financial advice. Nothing here is a recommendation to buy, sell, or hold any security. Consult a licensed advisor before making investment decisions.