Historical vs Implied Volatility: Understanding the Gap

Key Takeaways

• Historical volatility (HV) is an annualized standard deviation of past log returns; IV is the vol backed out of current option prices.
• The volatility risk premium (VRP), IV minus subsequent realized vol, has averaged roughly 4 vol points on the S&P 500 since 1990.
• A common mistake: comparing IV and HV over mismatched windows, a 30-day IV must be benchmarked against 30-day forward realized vol, not 60-day trailing HV.
• VRP flips negative during crashes when realized vol explodes past what IV priced; premium-selling strategies must size for those drawdowns.

What It Is

Historical volatility (HV), also called realized volatility, is the standard deviation of past log returns over a chosen window, annualized by the square root of the number of periods in a year (typically 252 for daily data). It is a pure statistic computed from the price series.

Implied volatility (IV) is the volatility figure backed out of current option prices. It reflects what traders are paying today for exposure to the distribution of outcomes between now and expiration.

HV looks backward. IV looks forward. They are measured in the same units, annualized percent, which lets you compare them directly.

The Intuition

If the options market were a perfect forecaster, IV would equal the realized volatility that follows. It usually does not. On equity indices, IV tends to run several volatility points above subsequent HV. That persistent gap is the volatility risk premium (VRP).

The reason is simple. Option buyers, especially hedgers buying index puts, are paying for protection against tail outcomes. Option sellers demand extra compensation for bearing that tail risk. The premium sellers collect on average is the VRP. It is not a free lunch: sellers eat the fat left tail when crashes actually happen, which is exactly why the premium exists.

Tracking the HV-IV spread gives you a view into how anxious the market is today versus how turbulent the recent past has actually been.

How It Works

Both measures use the same unit, but they are computed from different data sources.

HV = stdev(log_returns over last N periods) * sqrt(252)
IV = volatility that makes BS_model_price equal to market_option_price

The volatility risk premium is typically defined as implied volatility minus the realized volatility that unfolds after it:

VRP = IV_today - HV_realized_over_next_N_days

A positive VRP means IV was richer than what actually happened. A negative VRP means realized volatility overshot what the market had priced in. Realized VRP is measured after the fact, because you need the subsequent realized path to compute it.

On the S&P 500, research from AQR and others documents that the long-run VRP, measured as VIX minus subsequent 30-day realized S&P 500 volatility, has averaged roughly 4 volatility points since 1990. Since 2020, it has run higher on average.

VRP is not a single static number. It widens in calm regimes, compresses in stress, and flips negative during shocks like 2008 and March 2020, when realized volatility exploded past the levels IV had priced in.

Worked Example

Suppose today's 30-day IV on SPY is 18 percent. That is the market's current price for the next month of expected volatility. Over the next 30 days, SPY's actual daily log returns print a sample standard deviation of 0.88 percent, which annualizes to:

HV_realized = 0.0088 * sqrt(252) = 0.1397, or about 14 percent

The realized VRP for that month is:

VRP = 18 - 14 = 4 vol points

Option sellers who sold 18-vol premium and watched it realize at 14 kept the 4-point spread, minus transaction costs and tail-hedging costs. Option buyers paid it.

Now assume the next 30 days contain a rate shock. Daily standard deviation prints 2.5 percent, annualized HV is about 40 percent, and IV during that window climbs from 18 to 35. Realized VRP for the window flips sharply negative. The sellers who held through the event gave back multiple months of collected premium in a single window. That two-sided payoff is the core of VRP as a risk premium.

Common Mistakes

1. Comparing IV and HV over mismatched windows. A 30-day IV predicts volatility over the next 30 calendar days (roughly 22 trading days). Comparing it to a 60-day trailing HV is apples to oranges. Match your HV window to the remaining life of the IV you are measuring.

2. Treating HV as the "true" vol and IV as the forecast error. IV is an efficient market price that includes a risk premium, not a mean estimate of future realized volatility. Expecting IV to equal HV on average misunderstands what option premium is pricing.

3. Expecting the VRP to hold every period. On average, IV runs above HV, but the distribution has a left tail. VRP flips negative during crashes and regime breaks. Strategies that harvest VRP systematically need to size for those drawdowns, or they will blow up in the one event they were meant to survive.

4. Using a single HV number without considering path. Two months can have the same annualized HV but very different shapes. A month of steady 1 percent daily moves and a month with one 10 percent gap down surrounded by flat days can both print the same standard deviation. Gap risk matters for option positions even when the summary HV looks normal.

5. Assuming the IV-HV spread is a signal on its own. The spread is informative but not a mechanical trading rule. It widens before events, compresses after them, and is distorted by earnings, index rebalances, and macro catalysts. Read it alongside IV rank, the term structure, and the underlying's trend.

Frequently Asked Questions

Q: What is the difference between historical and implied volatility in simple terms? Historical volatility tells you how much the stock actually moved in the past. Implied volatility tells you what the options market thinks it will move going forward. Both are in annualized percent, so they can be compared directly.

Q: How does the HV vs IV gap affect investment decisions? When IV significantly exceeds recent HV, options are historically expensive and premium sellers have an edge. When HV exceeds IV, options are cheap and premium buyers may be better positioned.

Q: What is a real-world example of the volatility risk premium? SPY IV at 18% with subsequent 30-day realized vol at 14% produces a 4-point VRP for that month. The option seller who sold at 18 vol and watched it realize at 14 kept the spread as profit.

Q: How can investors use the HV-IV spread in their process? Match your HV lookback to the option's remaining life. A consistently positive spread favors selling premium; a negative spread warrants caution before entering short-vol positions.

Q: How is implied volatility different from a forecast? IV is a market price that includes a risk premium compensating sellers for bearing tail risk. Expecting IV to equal realized vol on average misunderstands what is embedded in the option price.

Sources

1. AQR Capital Management. "Understanding the Volatility Risk Premium." https://www.aqr.com/-/media/AQR/Documents/Whitepapers/Understanding-the-Volatility-Risk-Premium.pdf
2. CAIA Association. "What Is the Volatility Risk Premium?" https://caia.org/blog/2024/02/01/what-volatility-risk-premium
3. The Hedge Fund Journal. "Harvesting the S&P 500 Volatility Risk Premium." https://thehedgefundjournal.com/harvesting-the-s-p-500-volatility-risk-premium/
4. Macrosynergy. "Realistic Volatility Risk Premia." https://macrosynergy.com/research/realistic-volatility-risk-premia/

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.