Skip to main content
Implied Volatility

Implied Volatility Is a Forecast, Not a Fact

Pomegra Learn

Implied Volatility Is a Forecast, Not a Fact

Implied volatility is a prediction, not a promise. Every time a trader prices an option at a certain IV, they are saying "I believe the market will move at this annualized rate over the next month." But the market often surprises. The stock might swing far less than implied, or it might convulse more. A trader who understands that implied volatility prediction is often wrong—and acts on that belief—can turn volatility forecasting errors into systematic profits.

Lede​

Implied volatility is the market's consensus forecast of future price swings, embedded in option prices through mathematical models. But it is a forecast, subject to error. Realized volatility—the actual price movement that occurs—frequently diverges from what IV predicted. This gap is not a flaw; it is an opportunity. A trader who believes the market has overestimated future volatility can sell options expecting realized volatility to come in lower than implied. Conversely, one who expects bigger swings than IV suggests can buy options for edge. Understanding that implied volatility prediction is fallible is the first step toward systematic volatility trading.

Quick definition: Implied volatility is the market's estimate of future annualized price movement, calculated from option prices. Realized volatility is the actual price movement that occurs. The two often differ, and that difference is where volatility traders find edge.

Key Takeaways​

  • IV is an estimate, not a fact—the market's best guess about future swings, subject to revision and error
  • Realized volatility (what actually happens) differs from IV (what the market forecast) a substantial portion of the time
  • When realized volatility < IV, option sellers profit; when realized volatility > IV, option buyers profit
  • IV forecasts tend to be biased: overestimating before calm periods, underestimating before crisis periods
  • Systematic traders build models to forecast realized volatility and compare it to IV to identify mispricings

The Forecast Nature of Implied Volatility​

At its core, IV is a forecast. When traders discuss an option's IV, they are implicitly saying: "The market believes this underlying will experience annualized price movement of X percent over the next Y days, with 68% probability of staying within ±X percent (one standard deviation)."

This is a statement about the future, not the present. The future is unknown. Therefore, every IV forecast is subject to error.

Consider a stock trading at $100 with a 60-day call priced at an IV of 30%. The market is saying: "We expect this stock to move about 30% annualized over the next 60 days." That 30% prediction is the market's belief, aggregated from all traders bidding and offering options. But in 60 days, the realized volatility might be 18%, 32%, 50%, or any other value. The forecast was a best estimate, not a certainty.

Realized Volatility: The Actual Outcome​

Realized volatility is the annualized standard deviation of actual daily returns that unfold over the period the option is alive. It is calculated the same way as historical volatility—standard deviation of returns, annualized—but it looks at the specific time window during which the option was held.

Example:

  • Option expires in 30 days.
  • IV at purchase: 28%.
  • Over those 30 days, the stock's realized volatility (actual swings, annualized): 19%.
  • Result: Realized volatility < IV. Option sellers profit (they collected premium priced for 28% volatility but only 19% realized). Option buyers lose (they paid for 28% volatility but got 19%).

Realized volatility is objective and mechanical—you calculate it the same way every time, and the answer is unambiguous once the period ends. The gap between IV and realized volatility is a hard number that you can measure.

Why IV Forecasts Fail​

Market forecasts fail for predictable, recurring reasons:

1. Overnight information: An earnings surprise, geopolitical shock, regulatory decision, or macroeconomic data point can hit the market during hours when options aren't trading. Realized volatility spikes without warning. IV could not have forecast the exact timing or magnitude.

2. Volatility clustering: Volatility has memory. Low-volatility periods tend to persist; high-volatility periods tend to cluster. If IV is set based on recent calm conditions, the first shock can trigger a cascade of bigger shocks. Realized volatility exceeds IV because volatility clusters.

3. Tail events: IV is typically calibrated to normal market conditions, often using normal distributions in pricing models. Extreme tail events (crashes, crashes, black swans) occur more frequently than normal distributions predict. When a 5-sigma event hits, realized volatility far exceeds IV.

4. Structural changes: A stock's fundamental risk profile can change due to merger rumors, competitive disruption, or regulatory action. The realized volatility in the post-change period differs from the IV priced before the shift.

5. Consensus errors: All traders might be wrong in the same direction. During the 2008 financial crisis, every market participant underestimated realized volatility—IV was systematically too low. During the 2020 COVID crash, similar underestimation occurred. Herding and groupthink can bias forecasts.

The Volatility Smile and Skew​

A subtle point: IV is not a single number for all options on the same underlying and expiration. Deeper out-of-the-money options often have higher IV (skew). This is the market's way of saying "we expect bigger tail moves than normal distributions predict."

A stock with forward IV skew is telling you: "We think big downside moves are more likely than symmetric normal distributions imply." In other words, the market has already partially forecasted tail risk. But that forecast is still subject to error.

A Historical Example: Earnings Volatility​

Tech stock earnings are a classic case of IV forecast error:

Before earnings:

  • Stock at $150, IV at 45% (market priced in big move)
  • Market consensus: Stock could move ±7% on earnings (rough 68% probability band from 45% IV)

Earnings announcement:

  • Stock rises 3% ($150 → $154.50)
  • Much smaller than the ±7% move IV suggested

Realized volatility:

  • Post-earnings realized volatility: 18% (much lower than 45% IV)
  • Option sellers collected premium priced for 45% volatility but only 18% realized
  • Option buyers paid for 45% volatility but stock only moved 3%

This is not a failing of the Black-Scholes model or option pricing. It is the nature of forecasting. The market made a reasonable bet that earnings would be volatile, but the reality was calmer than expected.

Can You Predict When IV Forecasts Will Be Wrong?​

Not perfectly, but professional traders identify patterns:

Overpriced IV (IV > likely realized volatility):

  • Before major company events (earnings, earnings calls, analyst meetings) that turn out uneventful
  • When news flow has been heavy and is likely to slow
  • When IV rank is elevated (IV is high by historical standards) but the catalyst is past
  • In stable periods when traders are bidding up IV purely from fear, unwarranted by fundamentals

Underpriced IV (IV < likely realized volatility):

  • Before regulatory decisions with high uncertainty
  • Before Fed policy changes when dissent among officials is high
  • In trending markets where momentum can accelerate unexpectedly
  • In periods of market calm that historically precede regime shifts (volatility clusters)

Traders who systematize these patterns—building models to predict realized volatility and comparing to IV—can achieve positive expected value by consistently selling expensive IV and buying cheap IV.

The Role of Uncertainty​

A subtlety: IV is not just volatility; it is volatility given the current information set. If new information arrives that does not change price much but increases uncertainty, IV rises even if recent price swings were small.

Example:

  • Stock at $100, trading calmly with 15% HV.
  • FDA announces a clinical trial delay; outcome is now genuinely uncertain.
  • IV spikes to 35%, even though price barely moved.
  • The new IV of 35% is the market's forecast of future swings, conditional on this new uncertainty.

In this case, if the delay is eventually resolved without a shock, IV reverts to normal levels and realized volatility comes in much lower than 35% IV predicted. Sellers of options priced at 35% IV profit because the forecast (35% volatility) was too high relative to the reality (much lower realized volatility).

Real-World Examples​

Example 1: Pre-earnings calm

  • Biotech stock 5 days before earnings: 55% IV (high uncertainty).
  • Market expects a binary reaction: up 15% or down 10%.
  • Actual earnings: modest positive surprise, stock up 2%.
  • Realized volatility: 12%.
  • Result: IV was too high. Sellers of calls and puts (sold at 55% IV) collected premium and profit as IV compresses to 15% post-earnings. Buyers paid for 55% volatility but got 12%.

Example 2: Calm before the crash

  • Large-cap index historically shows 12% IV.
  • Market has been calm for 8 months: VIX at 10, IV at 8%.
  • Traders think volatility is dead; IV remains suppressed.
  • Black swan event hits: market crashes 8% in two days.
  • Realized volatility: 60% annualized.
  • Result: IV was far too low. Buyers of put options (bought at 8% IV) profit massively. Sellers who sold calls at 8% IV expecting calm face huge losses as volatility clusters.

Example 3: Fed decision day

  • Fed decision in 4 hours: IV at 28%.
  • Market pricing in 0.50% chance of aggressive taper surprise (stock market risk).
  • Fed announces moderate taper, within expectations.
  • Realized volatility: 8%.
  • Result: IV was too high. Sellers of straddles and strangles profit. Buyers overpaid for the expected move.

Using Realized Volatility for Backtesting​

Professional traders use realized volatility to evaluate their IV forecasts. If you sold calls at 30% IV and realized volatility turned out to be 22%, you made a good trade (sold expensive volatility). If you sold at 30% and realized was 40%, you made a bad trade (sold cheap volatility).

By tracking the distribution of your IV forecasts versus realized volatility, you can measure your forecast accuracy and calibrate future IV forecasts.

Common Mistakes​

1. Assuming IV is always overpriced: Some traders default to selling options because IV looks expensive. But IV is sometimes cheap, especially before crises. Sell selectively, based on analysis, not habit.

2. Ignoring that IV changes before realization: IV doesn't stay constant until expiration. If you sell options at 30% IV, IV might drop to 20% (your favor), then spike to 40% (against your favor) when news hits. The realization path matters, not just the final realized volatility.

3. Using the wrong realized volatility lookback: If you trade 30-day options, measure realized volatility over the 30-day hold period, not 60-day or 252-day HV. Match your realized volatility window to your option's life.

4. Forgetting gamma risk: If you sell options, realized volatility coming in lower than IV is great for you—unless price moves against you sharply. Gamma (price-movement risk) is separate from vega (volatility risk). A realized volatility lower than IV helps vega, but a sharp adverse price move hurts gamma. Manage both.

5. Failing to account for volatility of volatility: IV itself is volatile. IV can swing from 25% to 35% to 20% week-to-week, even if realized volatility is stable at 22%. This IV chop can wipe out gains from an accurate realized-volatility forecast if you hold too long.

FAQ​

Q: Is IV ever a perfect forecast? A: Rarely. IV is almost never exactly equal to realized volatility. If IV and realized volatility matched perfectly all the time, there would be no volatility trading profits—but professional traders consistently extract edge from IV forecasting errors, so the forecast is systematically off in exploitable ways.

Q: Can I measure my IV forecasting skill? A: Yes. Track every trade where you made a call on whether IV was too high or too low. Measure realized volatility at the end of the period. Calculate your hit rate and average profit/loss per forecast. Over time, this tells you whether your IV edge is real or luck.

Q: Does implied volatility predict direction? A: No. IV is about magnitude (how much), not direction (up or down). High IV means big swings are expected in either direction. A 60% IV means the stock might move ±15% either way, not specifically up or specifically down.

Q: What happens to IV after the forecast period ends? A: If an option expires, IV becomes irrelevant—the realized outcome is fixed. If you close the option before expiration, the remaining time value depends on the new IV (which may differ from when you entered).

Q: How do professionals forecast realized volatility? A: Using GARCH models, machine-learning regressions, options-market sentiment indicators (VIX term structure, put skew), historical volatility patterns, and fundamental analysis. There is no single method; professionals combine multiple signals for better edge.

Q: Is volatility mean-reverting? A: Generally, yes, but volatility clusters. After a calm period, volatility tends to rise (mean reversion); once volatility rises, it often stays elevated for a while (clustering). This is why IV forecasts are hard—volatility has both mean-reverting and trending properties.

Summary​

Implied volatility is the market's best estimate of future price movement, but estimates are fallible. Realized volatility—the actual swings that unfold—often diverges significantly from IV. When realized volatility is lower than IV, option sellers profit; when realized volatility exceeds IV, option buyers profit. This gap between forecast (IV) and reality (realized volatility) is where systematic traders make their living. By understanding that implied volatility prediction is prone to error, and by building methods to detect those errors before they materialize, you convert volatility forecasting into an edge.

Next​

→ Why High IV Means Expensive Options