Home / Implied Volatility / Event Volatility
Implied Volatility Intermediate The variance a scheduled event contributes Forward-looking

Event Volatility

Variance is additive, so a single date's contribution can be weighed on its own.

Quick answer: Event volatility is the portion of an option's implied volatility attributable to a single scheduled event — such as an RBI decision or the Union Budget — isolated by treating total variance as the sum of an ordinary diffusive component and a concentrated one-day event component.

In simple words

An option that spans a big scheduled event is really charging for two different things at once: the ordinary, everyday wobble of the market on all the normal days, plus one extraordinary day when the announcement lands. Event volatility is a way of separating those two charges. Because variance — volatility squared, times time — simply adds up across independent days, you can subtract the ordinary background wobble from the option's total and be left with the piece that belongs to the event alone. Do that and you can answer a genuinely useful question: how big a one-day move is the market actually pricing for budget day, or for the RBI decision?

This matters because the headline implied volatility on an option that contains an event is misleadingly high — it is blending a normal market with one special day. If you compare that inflated number to a normal level and conclude 'volatility is high', you have missed that almost all the excess is one dated event that will resolve and crush. Splitting it out tells you how much of the premium is the event, and therefore how much will disappear the moment the event passes.

Not to be confused with: Event volatility versus the headline implied volatility of the event-containing expiry. The headline number blends the event with all the ordinary days around it and is diluted by them; event volatility is the extracted one-day contribution, which is far higher than the headline and is the number that actually describes the announcement day.

The picture

One expiry carries the event; the others dilute it

At-the-money implied volatility across expiries bracketing a scheduled event date.

10%12%14%16%18%20%22%-25d-20d-15d-10d-5devent+5d+10d+15dscheduled eventpeak 18.8%crush to 11.8%Trading days relative to the eventAt-the-money implied volatility
The expiry that first contains the event prints a visibly higher implied volatility than its neighbours, and longer expiries print lower because the same fixed event variance is spread over more days. That the excess concentrates in one expiry is exactly what lets you back the event's own variance out of the term structure.

Professional explanation

Variance adds; volatility does not

The entire method rests on one property: variance is additive across independent time periods, while volatility — its square root — is not. If ordinary days each contribute their own variance and the event day contributes a much larger one, the total variance over the option's life is simply their sum. This is why you cannot just subtract a normal volatility from an event volatility and get anything meaningful — you have to work in variance, add and subtract there, and only take the square root at the end. Every mistake people make decomposing event volatility comes from doing the arithmetic in volatility units instead of variance units, and the correction is always the same: square first, subtract, then square-root.

The decomposition, stated properly

Write the total implied variance of the event-containing expiry as its diffusive part plus its event part: σ²_total × T = σ²_diffusive × (T − 1/365) + σ²_event × (1/365). The left side is the total variance the option's implied volatility implies over its life T. The first term on the right is the ordinary market variance accumulated over every day except the event day. The second term is the entire contribution of the single event day, which occupies 1/365 of a year. Rearranging gives σ²_event directly: σ²_event = [σ²_total × T − σ²_diffusive × (T − 1/365)] × 365. Every symbol here has to be defined and kept straight, because the whole result is only as trustworthy as the diffusive baseline you assume for the non-event days.

From event variance to an implied one-day move

The event variance on its own is abstract; the number a trader actually wants is the move the market is pricing for that one day. Convert the annualised event volatility σ_event to a one-day standard deviation by multiplying by √(1/365), then by the spot to get points: implied one-day event move ≈ S × σ_event × √(1/365), which is the same as S × √(σ²_event / 365). This is the market's own estimate of how far the underlying could travel on the announcement, extracted from option prices rather than guessed. It is the honest version of the question everyone asks before a budget — 'how much is priced in?' — and it is answerable precisely because variance is additive.

Why the first expiry after the event absorbs all of it

The event contributes a fixed amount of variance, and that fixed amount lands entirely in the first expiry whose life includes the event date. A shorter expiry that ends before the event carries none of it and prints an ordinary implied volatility. The first expiry after the event carries the whole thing, concentrated into few days, so it prints a sharply higher implied volatility. Longer expiries carry the identical fixed event variance but spread it over many more days, so the same contribution dilutes to a smaller bump — the event's fingerprint fades along the term structure. This is why the near-dated expiry over an event looks so expensive and why calendar structures that are short the near expiry and long a far one are the natural way to express a view on the event premium: the two expiries contain the same event, weighted completely differently.

Backing out the implied one-day move

The event-day variance extracted from the term structure, expressed as an implied one-day move on NIFTY.

10%12%14%16%18%20%22%1d12d30d60d90dthe event (e.g. an RBI policy date)expiries BEFOREthe event: cleanthe first expiry AFTER the event absorbs all of its variance,then dilutes as more ordinary days are addedDays to expiryImplied volatility
Once the event variance is isolated, converting it to a one-day standard deviation gives the move the market is pricing for the announcement itself. This is a number the raw implied volatility hides, and it is the practical output of the whole decomposition.

Formula

Additive-variance decomposition of event volatility

σ²_total × T = σ²_diffusive × (T − 1/365) + σ²_event × (1/365)

Total implied variance over the option's life equals the ordinary diffusive variance accumulated over the non-event days plus the entire variance contributed by the single event day, which occupies 1/365 of a year. Solve for σ²_event, then take the square root for the annualised event volatility. All arithmetic must be done in variance, never in volatility, because only variance is additive.

  • σ_totalTotal annualised implied volatility of the event-containing expiry, read off its at-the-money option (decimal).
  • TTime to expiry in years, calendar days ÷ 365.
  • σ_diffusiveThe ordinary annualised volatility the market realises on a normal, non-event day — estimated from an expiry that does not contain the event, or from a nearby non-event baseline.
  • σ_eventThe annualised volatility contributed by the single event day — the unknown being solved for.
  • 1/365The fraction of a year occupied by the single event day, in calendar-day convention.

Implied one-day event move

Move_event ≈ S × σ_event × √(1/365) = S × √(σ²_event / 365)

Converts the annualised event volatility into the one-standard-deviation move the market is pricing for the announcement day itself, in points of the underlying. This is the practical output — the market's own answer to 'how much is priced in for the event?'

How to back out the implied event move from the term structure

  1. Identify the event date and the first expiry whose life contains it — that expiry carries the whole event variance.
  2. Read the total at-the-money implied volatility σ_total of that expiry, and its time to expiry T in years.
  3. Estimate the diffusive baseline σ_diffusive from an expiry that ends before the event, or from a normal non-event level for the same underlying.
  4. Work in variance: compute σ²_total × T and subtract σ²_diffusive × (T − 1/365) to isolate the event's variance contribution.
  5. Multiply that residual by 365 to annualise it, then take the square root to get σ_event, the event's annualised volatility.
  6. Convert to the implied one-day move with S × σ_event × √(1/365), giving the points the market is pricing for the announcement.
  7. Sanity-check against the near-expiry straddle price, which should imply a similar move, and against how much the event-containing expiry's implied volatility exceeds its neighbours.

Practical example

NIFTY worked example

NIFTY is at 24,000, and an RBI policy decision falls inside a 5-day expiry. That expiry's at-the-money implied volatility reads 18%, while a normal, non-event diffusive baseline for NIFTY is about 13%. Work in variance. Total: σ²_total × T = 0.18² × (5/365) = 0.0324 × 0.013699 = 0.0004438. Diffusive over the four non-event days: σ²_diffusive × (T − 1/365) = 0.13² × (4/365) = 0.0169 × 0.010959 = 0.0001852. The event's variance contribution is the difference, 0.0002586, and annualising it by ×365 gives 0.0944, whose square root is σ_event ≈ 30.7%. So the single RBI day is being priced at a 30.7% annualised volatility — more than double the ordinary market. Converting to a one-day move: 24,000 × √(0.0944 / 365) = 24,000 × 0.0161 ≈ 386 points. Interpret it: the market is pricing roughly a 386-point one-standard-deviation move for NIFTY on RBI day alone, and the reason the headline 18% looked only modestly high is that this large event was diluted across four ordinary days.

BANKNIFTY worked example

BANKNIFTY around the Union Budget shows why the same headline number means different things on different underlyings. Suppose BANKNIFTY at 52,000 has a budget day inside a 5-day expiry printing 24% implied volatility, against a diffusive baseline of 16%. In variance: total 0.24² × 5/365 = 0.0576 × 0.013699 = 0.000789; diffusive 0.16² × 4/365 = 0.0256 × 0.010959 = 0.000281. Event variance 0.000508, annualised 0.1855, so σ_event ≈ 43.1%, and the implied one-day budget move is 52,000 × √(0.1855/365) = 52,000 × 0.02254 ≈ 1,172 points. That is a far larger absolute and percentage event move than NIFTY's, which fits: BANKNIFTY concentrates a single, budget-sensitive sector, so a fiscal announcement lands harder on it. The lesson is that you cannot compare the raw 18% and 24% headlines and conclude much — but you can compare the extracted event moves, and they tell you the market genuinely expects the budget to hit banks harder than the broad index.

Lot sizes used above (NIFTY 75, BANKNIFTY 30) are those in force at the time of writing; NSE revises them periodically. Figures exclude brokerage, STT, exchange charges, stamp duty and GST. Examples are teaching scenarios built on round numbers — they are not historical quotes, not backtests and not trade calls.

Advantages & limitations

What it is good for

  • It turns an inflated, blended implied volatility into a clean, interpretable number: the move the market is actually pricing for one specific day.
  • It works purely from observable option prices and the calendar, needing no forecast — the market's own event expectation is extracted, not assumed.
  • It explains the term structure's shape around an event, showing why the near expiry looks so expensive and longer expiries dilute the same fixed contribution.
  • It is the honest way to answer 'how much is priced in?' before a budget or policy decision, replacing a gut feeling with an arithmetic the reader can check.
  • It underpins calendar and event structures, because it quantifies exactly how differently two expiries weight the same event, which is what those structures trade.

Where it breaks down

  • The result is only as good as the diffusive baseline you assume; a wrong estimate of the ordinary non-event volatility flows straight into a wrong event volatility.
  • It assumes the event day's variance is independent of and additive to the ordinary days, which the underlying is not obliged to respect — volatility can cluster around the event too.
  • It treats the event as a single, clean one-day contribution, but some events resolve over multiple sessions or leak information beforehand, smearing the variance across days.
  • The extracted event move is a one-standard-deviation figure from the option market's pricing, not a prediction; the actual move is frequently larger or smaller.
  • It relies on the at-the-money implied volatility being a fair summary, yet the event may express itself more in the skew, which this symmetric decomposition ignores.
  • Near expiry the total implied volatility itself becomes unstable, so decomposing a very short-dated event expiry inherits all the fragility of an IV computed on a nearly-decayed premium.

Common mistakes

  • Doing the arithmetic in volatility instead of variance — subtracting 13% from 18% and calling the difference the event volatility. Only variance is additive; you must square, subtract, then take the root.
  • Comparing the raw event-expiry implied volatility across two underlyings and drawing conclusions. The headline blends the event with ordinary days differently for each; only the extracted event moves are comparable.
  • Using a stale or wrong diffusive baseline — for instance a calm-period volatility during a nervous market — which throws the whole decomposition off, usually overstating the event.
  • Treating the extracted one-day move as a prediction rather than a one-standard-deviation figure. The market prices a distribution; the actual outcome routinely lands outside it.
  • Forgetting that longer expiries dilute the same fixed event variance, and being surprised that a far-dated option barely reacts to an event the near-dated one is screaming about.
  • Assuming the event contributes symmetrically to calls and puts when the real event risk is directional and lives in the skew, which the at-the-money decomposition cannot see.
  • Applying the clean one-day model to an event that actually unfolds over several sessions, so the variance is smeared and the single-day extraction overstates the concentration.

Professional usage

Event-volatility desks run this decomposition continuously to price the event premium embedded in every expiry that brackets a known date, and it is the backbone of how they quote and hedge around RBI meetings, the budget, and results. Knowing the isolated event variance tells them precisely how much of the near expiry's implied volatility will crush when the date passes, and how to weight a calendar that is short the event-laden near month and long a diluted far one so the residual exposure is a clean bet on the event premium rather than on the general level of volatility. The extracted implied one-day move is also their sanity check against the straddle: the two should agree, and a divergence flags a mispriced expiry.

Risk and structuring desks use the event-versus-diffusive split to stress positions around scheduled dates, because a book that looks flat in headline vega can be dangerously exposed to a single event day once the variance is attributed correctly. On the sell side, comparing the market's extracted event move to the firm's own estimate of the likely move is how a desk decides whether an event's options are rich or cheap — and the discipline of stating that comparison in one-day-move points, rather than in an annualised volatility that hides the concentration, is what keeps the judgement honest. The uncomfortable part they will admit internally is that the diffusive baseline is a judgement call, so two desks can decompose the same expiry and disagree on how much of it is 'the event'.

Key takeaways

  • Event volatility is the portion of an option's implied volatility attributable to a single scheduled event, isolated by treating total variance as diffusive plus event variance.
  • The decomposition is σ²_total × T = σ²_diffusive × (T − 1/365) + σ²_event × (1/365); all arithmetic must be done in variance because only variance is additive.
  • Solving for σ_event and converting via S × √(σ²_event/365) gives the implied one-day move the market is pricing for the announcement — the number the headline IV hides.
  • The first expiry after an event absorbs the whole fixed event variance and prints a high IV; longer expiries dilute the same contribution over more days.
  • The result is only as reliable as the diffusive baseline assumed, and the extracted move is a one-standard-deviation figure, not a forecast.

Event volatility is the arithmetic that turns 'the RBI meeting is priced in' from a slogan into a number. Because variance adds and volatility does not, a single date's contribution can be lifted cleanly out of the term structure and expressed as the one-day move the market is actually charging for — 386 points on NIFTY for a policy decision, more on BANKNIFTY for a budget that hits banks harder. The honest caveat, the one worth keeping, is that the whole extraction hangs on the diffusive baseline you assume, so the precision of the output should never be mistaken for certainty about the input. It is the market's estimate, made legible — not a prediction, and not yours.

Frequently asked questions

What is event volatility?
Event volatility is the portion of an option's implied volatility that comes from a single scheduled event, such as an RBI decision or the budget. It is isolated by splitting total variance into an ordinary diffusive part and a concentrated one-day event part.
How do you separate event volatility from normal volatility?
You work in variance, not volatility, because only variance is additive. Total variance equals diffusive variance over the non-event days plus event variance for the one event day; subtract the diffusive part to isolate the event's contribution.
What is the formula for event volatility?
σ²_total × T = σ²_diffusive × (T − 1/365) + σ²_event × (1/365). Solve for σ²_event, then take the square root for the annualised event volatility. The event day occupies 1/365 of a year in calendar-day convention.
Why must the arithmetic be done in variance, not volatility?
Because variance is additive across independent days and volatility is not. Subtracting one volatility from another gives a meaningless number; you must square to variance, add or subtract there, and only take the square root at the end.
How do I find the implied one-day event move?
Convert the event volatility to a one-day standard deviation and multiply by spot: Move ≈ S × σ_event × √(1/365), which equals S × √(σ²_event/365). That gives the points the market is pricing for the announcement day.
Why does the near expiry over an event look so expensive?
Because it carries the entire fixed event variance concentrated into just a few days, so its implied volatility prints sharply higher. The event's whole contribution lands in the first expiry whose life includes the event date.
Why do longer expiries show a smaller event bump?
Because they carry the identical fixed event variance but spread it over many more days, so the same contribution dilutes to a smaller effect on the annualised implied volatility. The event's fingerprint fades along the term structure.
What diffusive baseline should I use?
The ordinary volatility the underlying realises on a normal, non-event day — estimated from an expiry that ends before the event, or from a calm non-event level for the same underlying. The whole result is only as good as this estimate.
Is the extracted event move a prediction?
No. It is a one-standard-deviation figure implied by option prices, not a forecast. The actual move on the day routinely lands larger or smaller; the number tells you what the market is charging for, not what will happen.
Can I compare event volatility across NIFTY and BANKNIFTY?
Yes, and it is more meaningful than comparing raw headline implied volatilities. The extracted one-day event moves are directly comparable, and they often show the budget or RBI hitting BANKNIFTY harder because it concentrates a single sector.
How does event volatility relate to IV crush?
The event volatility is precisely the part of the premium that will crush when the event resolves. Decomposing it tells you in advance how much of the near expiry's implied volatility is event premium and therefore how much the crush will remove.
Why does the event contribute 1/365 of variance?
Because the calendar-day convention treats one event day as 1/365 of a year, so its variance term is σ²_event × (1/365). The non-event days occupy the remaining (T − 1/365) of the option's life at the diffusive volatility.
What if the event unfolds over several days?
Then the single-day model overstates the concentration, because the variance is smeared across multiple sessions. Some events also leak information beforehand, so the clean one-day extraction becomes an approximation rather than an exact split.
Does event volatility live in the at-the-money option?
The symmetric decomposition uses the at-the-money implied volatility, but a directional event often expresses itself in the skew instead. The at-the-money split cannot see that, so it can understate a one-sided event risk.
How do calendar spreads use event volatility?
A calendar that is short the event-laden near expiry and long a diluted far expiry trades the difference in how the two weight the same event. Quantifying the event variance is exactly what tells you how differently they are exposed.
Can event volatility be higher than 100%?
The annualised event volatility can be very large — a single tense day annualises to a big number — but that is an artefact of annualising a one-day figure. What matters practically is the one-day move it implies, not the annualised headline.
How do I sanity-check an extracted event move?
Compare it to the near-expiry straddle price, which implies a similar move, and to how much the event expiry's implied volatility exceeds its neighbours. If they disagree badly, your diffusive baseline or the quote is likely off.
Does event volatility explain why headline IV looked only mildly high?
Yes. A large event diluted across several ordinary days lifts the blended headline only modestly, so an 18% event-expiry reading can hide a 30%-plus event day. The decomposition reveals the concentration the headline conceals.
Is event volatility the same as India VIX?
No. India VIX is a model-free 30-day summary of the whole near-dated NIFTY chain; event volatility is the extracted contribution of one specific date. India VIX can rise into an event, but it does not isolate the event the way this decomposition does.
What is the biggest source of error in the decomposition?
The diffusive baseline, which is a judgement call. Two analysts can assume different normal volatilities and extract materially different event volatilities from the same expiry, so the output's precision should not be mistaken for certainty.

Voice search & related questions

Natural-language questions people ask about event volatility.

How much is the RBI meeting priced into options?
You can extract it: split the event-expiry's implied volatility into a normal-days part and an event-day part using variance, and convert the event part to a one-day move. On NIFTY a policy day often prices in a few hundred points.
Why can't I just subtract normal volatility from event volatility?
Because volatility doesn't add up — variance does. You have to square both numbers to get variance, subtract there, then take the square root. Subtracting the percentages directly gives a number that means nothing.
What does it mean that one expiry carries the whole event?
The event adds a fixed lump of variance, and it all lands in the first expiry whose life includes the event date. Shorter expiries that end before it carry none; longer ones spread the same lump thinner, so the bump fades out along the curve.
How do I turn event volatility into a number of points?
Take the event volatility, multiply by the square root of one over 365 to get a one-day figure, then multiply by the spot price. That gives the one-standard-deviation move the market is pricing for the announcement day.
Is the priced-in event move usually right?
It is the market's one-standard-deviation estimate, not a forecast, so the real move is often bigger or smaller. It tells you what you are being charged for the event, which is a different thing from what the event will actually do.
Why does the budget hit BANKNIFTY harder in the numbers?
Because BANKNIFTY is concentrated in one sector that fiscal announcements move a lot, so its extracted event volatility and implied one-day move come out larger than the broad index's, even when the headline implied volatilities look similar.
Does event volatility depend on my assumption of normal volatility?
Heavily. The diffusive baseline you assume for ordinary days flows straight into the answer, so a wrong baseline gives a wrong event volatility. That judgement call is the biggest source of error in the whole decomposition.

Sources & references

Last reviewed 10 July 2026. Educational content only — not investment advice.

← Previous
IV Crush
Educational content only — not investment advice. Every diagram on this page is generated from the site's own model, using illustrative inputs rather than live quotes. Options and futures carry substantial risk, including loss exceeding your deposit on short-volatility positions. See our Risk Disclosure and SEBI Disclaimer.