Volatility & Options Intermediate The IV life cycle around any scheduled event Forward-looking

IV Around Events Event IV

The option cannot tell you what will happen — only what it costs to be surprised.

Quick answer: IV around events is the predictable arc an option's implied volatility traces as a scheduled announcement approaches: it rises convexly while the uncertain day sits inside the option's remaining life, peaks on the eve, and collapses the instant the outcome is public — because the option correctly stops charging for a risk that no longer exists.

In simple words

Some days are ordinary and some days are not. An RBI decision, the Union Budget, a company's results and an election count are days on which the market can move several times its usual range in a single session. An option that has to survive one of those days must charge more, because whoever sold it is underwriting an extraordinary risk. So as the known date approaches, the option's implied volatility climbs — a NIFTY at-the-money option might drift from a calm 12.2% up to 18.8% on the eve of the event. Then the outcome is announced, everybody learns what happened, and the extra uncertainty simply vanishes. The next morning that same option's implied volatility is back near 11.8%. That drop is called IV crush, and it happens whether or not the market actually moved.

The reason the rise is convex — slow at first, then steep near the date — is that the event has to fit inside the option's remaining life to matter to it. A month out, the event is one uncertain day diluted across thirty; the day before, it is one uncertain day out of one or two, so it dominates the price. Implied volatility is an average of the variance still to come, and as the calm days ahead of the event get used up, the single loud day counts for more and more of what is left.

Not to be confused with: A market shock, which looks similar on a chart but is the opposite mechanism. Around a scheduled event, implied volatility rises because a known date approaches and falls because the outcome resolves the uncertainty. In a shock, implied volatility jumps because something unknown just happened and the market has repriced how uncertain the world is — and it then decays slowly over weeks, because nothing has resolved. A scheduled event has a built-in end; a shock does not.

The implied volatility life cycle around a scheduled event

Convex rise, vertical crush

At-the-money NIFTY implied volatility through a scheduled event: base 12.2%, peak 18.8% on the eve, 11.8% the morning after.

The build-up is a curve and the collapse is a cliff. Implied volatility rises gently and then steeply as the event slides inside the option's remaining life, but it falls almost vertically the moment the outcome is known — because the uncertainty the option was pricing does not decay, it is deleted. The asymmetry between the two sides of the peak is the whole point: you accumulate the premium slowly and lose it all at once.

Professional explanation

Implied volatility is an average of variance, and variance is additive in time

The cleanest way to think about event implied volatility is to stop thinking about volatility and think about variance, which is volatility squared. Variance has a property volatility does not: it adds up across non-overlapping periods. The total variance an option prices between now and expiry is the sum of the variance of each day in between. Most days contribute an ordinary amount; the event day contributes an extraordinary amount. Implied volatility is just the square root of that total variance re-expressed per year, so it is a blended average — it dilutes the loud day across all the quiet ones. This single fact explains everything else on the page: why the rise is convex, why the near expiry spikes and the far expiry barely moves, and why you can algebraically pull the event day back out of the quoted number.

Why the rise is convex rather than linear

If an event adds a fixed lump of variance to an option, and implied volatility is that lump diluted across the remaining days, then as the calm days ahead of the event are used up the lump is divided by a smaller and smaller number. A month before the event the extra variance is spread across thirty days and barely lifts the quoted implied volatility. A week before, it is spread across seven. The night before, it is spread across one or two, and the quoted number leaps. The curve is steep at the end not because the market suddenly got more worried but because arithmetic concentrates a constant quantity into a shrinking denominator. A trader who buys a week early and expects the same daily IV gain they would get the day before has misread the geometry.

Scheduled events crush; shocks decay

The defining difference between the two things that lift implied volatility is what happens on the far side. A scheduled event resolves: at a known time, the number is read out, the result is public, and the uncertainty that the option was charging for ceases to exist. Implied volatility does not glide back down, it drops in a single session — often overnight — to slightly below where the build-up began. An unscheduled shock does the opposite. Nothing resolves; the market has simply learned that the world is more dangerous than it thought, and it reprices every option upward to reflect the new, higher baseline uncertainty. That elevation then decays over days and weeks as realised volatility comes in lower than the frightened price, not because any single fact was revealed. Same chart shape, opposite cause, opposite trade.

The first expiry after an event absorbs all its variance

An event's variance is paid for entirely by the first expiry that closes after it. That expiry must contain the event day, so its total variance includes the event lump; the next expiry out contains the same lump but spreads it across many more days, so its at-the-money implied volatility is lifted far less. The result is a term structure that spikes at the front expiry and decays outward — not a smooth hump you can interpolate away, but a genuine kink that reflects where the risk actually sits. Traders exploit this deliberately: a calendar spread that is short the near expiry and long the far one is, in part, a bet on that kink flattening after the event, when the near expiry's excess variance has been spent and the two implied volatilities converge.

The uncomfortable sentence

Buying an option for an event is a bet that the actual move will exceed a premium that has already been marked up for exactly that possibility. The market has seen the same calendar you have; the elevated implied volatility is not an oversight you are exploiting, it is the price of the ticket. Being right about the direction is not enough, and being right that the market will move is not enough — you have to be right that it will move more than the amount already charged for. Most of the time the move, however real, is smaller than the premium paid for it, which is why the systematic edge, such as it is, has historically sat with the seller who underwrites the day rather than the buyer who insures against it. That is a sentence a brokerage running an options-buying campaign would quietly delete.

Only one expiry pays for the event

The extra implied volatility an event adds, spread across expiries by their remaining life.

The event contributes a fixed amount of variance to whichever expiry contains it. On the near expiry that variance is divided by a handful of days, so the implied volatility bump is large; on a far expiry the same variance is divided by many days, so the bump is small. That is why the event term structure spikes at the front and decays outward — a shape you cannot smooth over by interpolating, because it is real, not a data glitch.

Formula

Splitting an event day out of total implied variance

σ²_total · T = σ²_diff · (T − t_e) + σ²_event · t_e, t_e = 1/365

Implied variance is additive across time, so the total variance an option prices (left) is the ordinary diffusive variance over the days that are not the event, plus one event day's worth of variance. Multiplying through by 365 and writing D for calendar days to expiry gives σ²_event = σ²_total · D − σ²_diff · (D − 1). The one-standard-deviation move the option market is pricing for the event day is then m = σ_event · √(1/365).

σ²_totalTotal implied variance the option is pricing — the square of the quoted at-the-money implied volatility on the eve (0.188² here).
σ²_diffDiffusive (ordinary-day) variance — the square of the calm base implied volatility away from the event (0.122² here).
σ²_eventThe variance contributed by the single event day, the quantity being solved for.
σ_eventThe event day's own annualised volatility, √(σ²_event). Here it works out to about 39.8%.
σ_totalQuoted eve-of-event at-the-money implied volatility, 18.8%.
σ_diffCalm base implied volatility, 12.2%.
TTime to expiry in years on the eve of the event, measured as calendar days ÷ 365.
t_eThe length of one event day as a fraction of a year, 1/365 — a single calendar day, because interest and time accrue on the calendar even though the event is one trading session.
DCalendar days to expiry on the eve of the event; D = 7 in the worked example.
mThe one-standard-deviation event-day move the option market is implying, as a fraction of spot: m = σ_event · √(1/365).

The move the option is charging for, in one line

m = √( σ²_total · D − σ²_diff · (D − 1) ) ⁄ √365

This is the number worth extracting before any event trade. With σ_total = 18.8%, σ_diff = 12.2% and D = 7, the event variance is 0.188²·7 − 0.122²·6 = 0.158104, so σ_event = 39.76% and m = 39.76% ⁄ √365 = 2.08%. On NIFTY at 24,000 that is a one-day move of about 499 points already inside the price. Your long option only wins if the market beats that.

How to read an event out of an option chain

Identify the calm base implied volatility — the at-the-money level on a nearby expiry with no event inside it, or the same underlying's implied volatility a week before the build-up began. Here it is 12.2%.
Read the eve-of-event at-the-money implied volatility on the first expiry that closes after the event. Here it is 18.8%.
Count the calendar days from the eve to that expiry. Here D = 7.
Compute the event variance: σ²_event = σ²_total · D − σ²_diff · (D − 1) = 0.188²·7 − 0.122²·6 = 0.158104.
Take the square root for the event-day annualised volatility: σ_event = 39.76%.
Convert to a one-day move: m = 39.76% ⁄ √365 = 2.08% of spot, about 499 NIFTY points at 24,000.
Interpret it: this is the move already paid for. A long straddle needs the market to beat it after costs to profit; a short position keeps the premium only if the market stays inside it. Neither is a free lunch.

Practical example

NIFTY worked example

NIFTY is at 24,000 with a scheduled event seven calendar days ahead of the near expiry. Away from the event the at-the-money implied volatility sits at 12.2%; on the eve it has climbed to 18.8%. Split the event day out. Total implied variance on the eve is 0.188² × (7/365) = 0.035344 × 0.019178 = 0.000678. Ordinary variance over the six non-event days is 0.122² × (6/365) = 0.014884 × 0.016438 = 0.000245. The difference, 0.000433, is one event day's variance; annualised that is 0.000433 × 365 = 0.158, so σ_event = √0.158 = 39.8%. As a one-day move, m = 39.8% × √(1/365) = 2.08% of spot — about 499 NIFTY points. Interpret it: the option market is not forecasting a 499-point move, it is charging for one. If you buy the straddle you need NIFTY to travel more than roughly 499 points on the day, net of the premium you paid and the crush that follows, merely to break even. The move can be real and violent and you can still lose.

BANKNIFTY worked example

BANKNIFTY teaches the term-structure half of the lesson. With BANKNIFTY at 52,000, suppose the near weekly expiry closes two days after an event and prints an at-the-money implied volatility of 19.0%, while the monthly expiry twenty-eight days out prints only 14.2% against a 13.0% base. Both expiries contain the same event, but the weekly divides its variance across a couple of days and the monthly divides it across four weeks, so the weekly's implied volatility towers over the monthly's. A trader who sees the weekly at 19.0% and the monthly at 14.2% and calls the weekly 'expensive' has misdiagnosed the shape: the weekly is not overpriced relative to the monthly, it is simply the expiry that has to carry the event. After the event resolves, the weekly's excess variance is spent and the two implied volatilities collapse toward each other — which is precisely the convergence a post-event calendar spread is built to harvest, and precisely the convergence that fails to arrive if the event lands violently.

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.

Risk note. The rule of thumb that you should 'sell options before an event to capture the crush' is held by many practitioners, and the crush is real and reliable. But the premium you collect is not a gift — it is payment for the gap that occasionally arrives. Most events resolve inside the priced move and the seller keeps the premium; a minority resolve far outside it, and on those the seller pays out a multiple of everything collected across the quiet ones. A short straddle into an event has capped reward and, on a genuine surprise, effectively uncapped loss. The crush is the compensation for standing in front of that tail, not evidence that it is not there.

Advantages & limitations

What it is good for

It is knowable in advance. Unlike a shock, a scheduled event sits on a public calendar, so the implied volatility build-up can be anticipated, measured against prior cycles of the same event, and priced rather than reacted to.
It lets you separate the event from the drift. The variance decomposition cleanly splits an option's implied volatility into an ordinary-day component and an event-day component, so you can see exactly how much of the premium is the event and how much is the ambient market.
It converts directly into a tradeable expectation. Backing σ_event out of the chain gives the one-day move the market is pricing, which is the single most useful number before deciding whether an event position is worth taking.
It exposes the term-structure kink. Because only the first expiry after the event carries the event's variance, comparing expiries reveals where the risk actually sits and makes calendar and diagonal structures legible rather than guesswork.
It is symmetric information. The buyer and the seller are both looking at the same elevated number, so nobody is being tricked — the disagreement is honestly about whether the realised move will exceed the priced one.

Where it breaks down

The clean two-component split assumes exactly one event of interest inside the expiry. When two events share a window — an exit poll and a counting day, or results clustering in a busy fortnight — the decomposition attributes their combined variance to a single day and overstates any one of them.
It treats the base implied volatility as stable, but the ambient level drifts on its own. If the whole market's volatility is rising for unrelated reasons during the build-up, part of the 'event' bump you measure is really a change in the diffusive baseline, and σ_event comes out too high.
It assumes the event's timing is precise. When an announcement window is vague — a policy that 'may' come this meeting, results with an unconfirmed date — the market smears the variance across several sessions and the neat one-day interpretation breaks.
The crush is only instant for events that actually resolve. A scheduled event that produces an ambiguous or partial outcome — a split result, a deferred decision — leaves residual uncertainty, so implied volatility falls part-way rather than all the way, and the crush you sold for does not fully arrive.
It says nothing about direction or about which strikes move most. The decomposition is a magnitude statement; the skew can steepen or flatten through an event independently, so an at-the-money read can miss where the real repricing is happening in the wings.

Common mistakes

Buying an option for an event and expecting a directional win to pay off. The premium is already marked up for the move, so being right about direction is not enough — the market has to travel further than the amount you were charged, after the crush, just to break even.
Reading the eve-of-event implied volatility as a forecast of the move. It is a price, not a prediction. An 18.8% eve reading tells you what the option costs, not that NIFTY will deliver the 2.08% one-day move it implies.
Selling the crush without respecting the tail. 'Sell before the event to capture the crush' works on the quiet majority and is catastrophic on the loud minority; the premium is payment for that minority, not evidence it will not show up.
Interpolating across the term-structure kink. The near expiry's implied volatility spike is real — it is the expiry that carries the event — so smoothing it into the neighbouring expiries erases exactly the information you were trying to read.
Confusing a shock with a scheduled event and expecting an overnight crush. A shock decays over weeks because nothing resolved; waiting for a one-session collapse that only scheduled events produce is a good way to hold a losing short-vega position too long.
Comparing the near expiry's high event implied volatility to a far expiry's low one and calling the near 'expensive'. The near is simply where the variance sits; the levels are consistent, not mispriced.
Forgetting that a long option can lose on a big move if the crush is bigger. Vega collapses as the event resolves, and if the move is inside the priced range, theta plus IV crush can overwhelm the intrinsic gained even on a day that moved in your favour.

Professional usage

Event desks build an explicit two-regime model of the underlying's variance: an ordinary diffusive rate that governs quiet sessions and a separate event-day variance for each dated catalyst on the calendar. Rather than quoting a single implied volatility, they fit the term structure so that the first expiry after each event carries that event's variance and the expiries beyond it dilute it correctly, which keeps the whole surface arbitrage-free through the build-up. The trade is then expressed relative to their own estimate of the event-day move — long the event if the chain is pricing less variance than their model expects, short it if the chain is pricing more — and the position is delta-hedged so the residual profit and loss depends on realised versus priced event variance rather than on which way the market broke.

On the sell side and in risk, event implied volatility is used as an early-warning gauge of positioning. A near-expiry event bump that is unusually large relative to prior cycles of the same event signals crowded protection buying, and a skew that steepens into the event tells the desk that the demand is concentrated in downside strikes. Margin and stress models lean on the event-day variance directly, sizing the potential one-session move from σ_event rather than from a trailing historical estimate that, by construction, has never seen the event that is about to happen.

Key takeaways

Implied volatility rises convexly into a scheduled event because the event's fixed variance is diluted across a shrinking number of remaining days, and it crushes vertically afterward because the uncertainty resolves rather than decays.
Scheduled events crush instantly; unscheduled shocks decay over weeks. Same chart shape, opposite mechanism — one resolves, the other merely reprices how uncertain the world is.
Only the first expiry after an event carries its variance, so the event term structure spikes at the front and decays outward — a real kink you must not interpolate away.
You can back the event out: σ²_event = σ²_total · D − σ²_diff · (D − 1) gives the one-day move already inside the price. Here that is 2.08% of spot, about 499 NIFTY points.
Buying an option for an event is a bet the move will beat a premium marked up for exactly that move. Direction alone does not pay.

Learn to read the event bump as a receipt, not a warning. When implied volatility climbs into a known date it is telling you the price of surviving that date, and the arithmetic lets you turn that price back into the move the market is charging for. Once you can see the 499-point move sitting inside the premium, the event trade stops being 'the market is about to move a lot, so buy options' and becomes the far harder question of whether it will move more than the amount already paid for — which is the only question that decides who wins, and the one the elevated number cannot answer for you.

Frequently asked questions

What is IV crush?

IV crush is the sudden fall in an option's implied volatility the moment a scheduled event's outcome becomes public. The option had been charging extra for the uncertainty of that day; once the result is known the uncertainty is gone, so the extra premium is deleted rather than decayed. A NIFTY option can drop from a 18.8% eve reading to 11.8% the next morning regardless of whether the market moved.

Why does implied volatility rise before an event?

Because the option still has to survive the event day, and whoever sold it must be compensated for a session that can move several times a normal day's range. As the known date slides inside the option's remaining life, the event's variance is diluted across fewer and fewer ordinary days, so the quoted implied volatility climbs — gently at first, then steeply on the eve.

Why is the rise convex and not a straight line?

Because the event adds a fixed lump of variance, and implied volatility is that lump divided by the remaining days. A month out it is divided by thirty and barely shows; the night before it is divided by one or two and leaps. The steepness at the end is arithmetic concentrating a constant quantity into a shrinking denominator, not the market suddenly getting more worried.

What is the difference between a scheduled event and a shock?

A scheduled event has a known date and resolves at a known time, so implied volatility rises into it and crushes the instant the outcome is public. A shock is unscheduled and resolves nothing — it just teaches the market the world is more dangerous, lifting all implied volatilities to a new baseline that then decays slowly over weeks. Same chart shape, opposite cause.

Which expiry carries an event's volatility?

The first expiry that closes after the event carries essentially all of it, because that expiry must contain the event day. Longer expiries contain the same event but spread its variance across many more days, so their implied volatility is lifted far less. This is why the event term structure spikes at the front and decays outward.

Can I interpolate the event bump away in the term structure?

No — that would erase the information. The near expiry's spike is not a data glitch; it is the expiry that actually carries the event's risk. Smoothing it into neighbouring expiries throws away exactly the signal that tells you where the risk sits, and it will misprice any calendar structure built on top of it.

How do I calculate the move the market is pricing for an event?

Split the event day out of total variance: σ²_event = σ²_total · D − σ²_diff · (D − 1), where D is calendar days to expiry on the eve. Take the square root for the annualised event volatility, then multiply by √(1/365) for the one-day move. With 18.8% peak, 12.2% base and D = 7 it comes to 2.08% of spot.

Does a high event implied volatility mean the market will move a lot?

It means the option market is charging as though it will. The eve reading is a price, not a forecast — it tells you the move already paid for, not that the move will arrive. On average the priced move slightly exceeds what realises, which is why the systematic edge has historically sat with the seller, not the buyer.

Can I lose money buying an option for an event even if the market moves?

Yes, and it is the most common event disappointment. The premium is marked up for the move, so if the market travels less than the priced amount you lose despite being right that it would move. And because implied volatility crushes as the event resolves, vega loss can overwhelm the intrinsic gained even on a day that moved in your favour.

Is selling options before an event a reliable edge?

It works on the quiet majority of events and fails badly on the loud minority, so it is a rule of thumb, not an edge. The crush you collect is payment for the occasional event that resolves far outside the priced move, on which a short straddle can pay out a multiple of everything gathered across the calm ones. The premium is compensation for a tail, not proof the tail is absent.

Why does implied volatility fall even if the market gapped hard on the event?

Because the crush is about resolved uncertainty, not about calm. Once the outcome is public the option no longer has to price an unknown day, so its event-day variance is deleted — even if the resolution was violent. The realised move rewards whoever owned gamma into the day; it does not keep implied volatility elevated afterward.

What is the diffusive base implied volatility?

It is the ordinary-day implied volatility away from any event — the level the option would trade at if nothing were scheduled inside its life. On NIFTY that base sits around 11% to 13%. It is the baseline you subtract from the eve reading to isolate the event's contribution to the premium.

Does the base implied volatility stay constant through the build-up?

Not necessarily, and that is a real limitation. If the whole market's volatility is drifting up for unrelated reasons while the event approaches, part of the bump you measure is a change in the baseline rather than the event, and your extracted event move comes out too high. The decomposition assumes a stable diffusive level it does not always get.

Why do two events in one window break the calculation?

Because the clean split assumes exactly one special day inside the expiry. When an exit poll and a counting day, or a cluster of results, share a window, the two-component decomposition attributes their combined variance to a single day and overstates whichever one you are trying to isolate. You need a multi-event model to separate them.

What happens to the skew around an event?

It often steepens as hedgers concentrate demand in downside strikes, then normalises after the outcome. The at-the-money decomposition on this page is a magnitude statement and does not capture that — an event can leave the at-the-money implied volatility looking ordinary while the wings reprice sharply, so a straddle read can miss where the real action is.

Is IV crush the same as theta decay?

No. Theta is the steady daily loss of an option's time value as expiry approaches, a smooth grind. IV crush is a one-off collapse in the implied volatility input itself when an event resolves, which hits every option's premium at once regardless of days to expiry. Both erode a long option, but they are different Greeks acting for different reasons.

How far ahead does the event build-up begin?

It depends on the event's size and how far it sits inside liquid expiries, but the convex shape means most of the visible build happens in the final week or two, with the sharpest move on the eve. A large event like a Budget can lift implied volatility over roughly a month; a routine one may only show for a few sessions.

Does India VIX show the event bump?

Partly. India VIX interpolates to a constant 30-day horizon across the near-dated chain, so a sharp single-expiry event spike is diluted in it. A near-expiry at-the-money implied volatility will show an event far more vividly than the headline VIX, which is smoothed across strikes and rolled to a fixed tenor.

Why does the crush sometimes only go part-way down?

Because the event only partly resolved. If the outcome is ambiguous, split, or defers a decision to a later date, residual uncertainty remains, so implied volatility falls part-way rather than all the way. An event that opens a new source of uncertainty can even leave implied volatility above where the build-up began.

Should a beginner trade options around events?

Only with a clear understanding that the elevated premium is the point, not an obstacle. Events are where the gap between 'the market will move' and 'the market will move more than the price' is widest, and where beginners most often confuse the two. Nothing here is a recommendation; event positions carry the same uncapped-loss risk as any short-volatility trade.

What is a calendar spread's role around an event?

A calendar spread that is short the near expiry and long the far one is partly a bet that the term-structure kink flattens after the event, when the near expiry's excess variance has been spent and the two implied volatilities converge. It profits from the crush being concentrated in the front expiry — and loses if the event lands violently enough to move the far expiry too.

Voice search & related questions

Natural-language questions people ask about iv around events.

What happens to option prices around an event?

They get more expensive going in and cheaper coming out, even if nothing changes in the underlying. The extra cost is the option pricing the uncertainty of the event day, and the drop afterward — IV crush — is that uncertainty being deleted once the outcome is known. A move on the day rewards gamma, not the elevated implied volatility.

Why did my option lose money even though I was right about the move?

Because the market probably moved less than the price already assumed, and because implied volatility crushed as the event resolved. The premium you paid was marked up for exactly the kind of move you predicted, so being right about direction only pays if the move beats the amount already charged for it.

Is it better to buy or sell options into an event?

Neither is generally better — they are opposite sides of the same honest disagreement about whether the move will beat the price. Buyers win when the market travels further than the priced move; sellers win when it stays inside, and collect the crush. The seller carries a tail risk with effectively no cap, which is what the crush is paying them for.

How do I know how big a move the market expects?

Split the event out of the implied volatility. Take the eve reading and the calm base, apply σ²_event = σ²_total · D − σ²_diff · (D − 1), and convert to a one-day move. For a typical NIFTY event that is around 2% of spot, roughly 499 points at 24,000 — the move already inside the premium.

Why does implied volatility spike on the front expiry but not the back?

Because the event's variance is a fixed lump, and the front expiry divides it across a few days while the back expiry divides it across many. Same event, very different implied volatility, because implied volatility is an average and the averaging windows differ. That is the term-structure kink, and it is real.

Does a shock behave like an event?

On a chart, briefly — both spike implied volatility. But a shock resolves nothing, so instead of crushing it decays slowly over weeks as the frightened price proves too high. An event has a built-in end date and crushes at it. If you are waiting for an overnight crush after a shock, you will wait a long time.

Can implied volatility end up higher after an event than before?

Yes, when the event opens a new source of uncertainty rather than closing one. If an outcome is ambiguous or reshapes the landscape, the crush is incomplete and the base level resets higher. That partial crush is the signature of an event that resolved less than the market hoped it would.

Sources & references

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