Earnings and volatility crush

빠른 답변

Earnings reports are scheduled uncertainty. The market knows a stock is likely to jump on a known date, so options in expirations that span the report carry extra implied volatility — a priced-in earnings move sitting on top of ordinary day-to-day movement.

Once the report is out, that uncertainty resolves and the extra volatility drains from option prices quickly, usually within a session. The drop is known as volatility crush, and it happens whether the stock moves a lot or barely at all.

Upcoming reports, with the moves currently priced into their option chains, are listed on the earnings page.

자세히 보기

An expiration that spans a report has to price two things at once: normal drift on the ordinary days, plus one day expected to be anything but ordinary. That second part inflates the expected move and the implied volatility of near-dated options in the weeks before the date.

A worked example with round numbers:

• Stock price before the report: $80
• ATM straddle in the week's expiration: $6.00 — an expected move of ±7.5%
• Implied volatility before: 90%
• Stock after the report: $80.80, a move of +1%
• Implied volatility after: 45%
• The same straddle after: roughly $3.00

The stock moved 1% against a priced move of 7.5%, and the straddle lost about half its value in a single session. Little of that came from the calendar — it came from implied volatility halving once the event passed. This is the mechanical reason option prices around earnings behave so differently from ordinary weeks.

Options Band measures each earnings reaction from the close before the report to the close after — two sessions — because the data source does not publish whether a company reports before the open or after the close. The convention, and what it does and does not capture, is documented at the earnings-reaction methodology page.

선택 사항: 세부 내용 확인 문제에는 필요하지 않음

Formally, implied variance in an expiration spanning a report can be decomposed into ordinary diffusive variance plus an event variance attributable to the announcement. Short-dated expirations hold the least diffusive variance, so they show the event most clearly: annualized implied volatility in a two-day expiration spanning earnings can print far above the same stock's monthly figure without any inconsistency. That is a term structure effect, not a mispricing.

The main misconception is scoring the market on a single outcome. A post-report move smaller than the priced move does not mean the pricing was wrong — the expected move describes a distribution, and under the model's own assumptions a one-straddle range should contain the outcome only about 60% of the time. Options Band records each priced move before the report and scores it afterward; see the historical database methodology. Nor is the crush itself a defect: implied volatility falling after uncertainty resolves is the market repricing, not misbehaving.

Limitations: the two-session measurement window can include market-wide movement unrelated to the report, and delayed Cboe quotes mean the before-report straddle is a snapshot near the close rather than the final print.