IV Rank
Fourteen percent is neither high nor low until you know where it sits in the last year. IV rank is that context in one number.
In short: IV rank measures where the current implied volatility sits between its lowest and highest values over the past 52 weeks: 0 means at the year’s low, 100 at the year’s high. It is computed on the at-the-money IV of a reference expiry, usually 30 days. IV percentile answers a different question, the share of days in the past year when IV was below today’s level, and the two can diverge after a single spike. Low rank is read as volatility being cheap relative to its own history and high rank as rich; neither says where volatility is going next.
The intuition
An option is expensive or cheap only relative to something. Comparing today’s IV with the realized volatility of the underlying is one comparison; comparing it with its own past is another, and that is what IV rank does. If NIFTY’s 30-day ATM IV has ranged between 9% and 24% in the past year and is 14% today, it is a third of the way up its range: not cheap, not rich. The same 14% in a year that ranged from 11% to 16% would be near the top. The number is a scale, not a forecast; its use is to stop you selling options because the premium looks large in rupees when it is actually small in the context of the year.
Formulas
IV rank = (IVtoday − IV52-week low) ÷ (IV52-week high − IV52-week low) × 100
IV percentile = (number of days in the past 52 weeks with IV below IVtoday) ÷ (number of days) × 100
Both are computed on a consistent series, normally the 30-day ATM IV or an index such as India VIX, not on the IV of whichever strike or expiry happens to be open.
Worked example (illustrative figures)
Suppose NIFTY is at 24,000 and the 30-day ATM IV is 14%. Over the past 52 weeks it ranged from 9% to 24%, but the 24% was a three-day spike around one event; on 230 of the 250 trading days it was below 14%.
| Measure | Calculation | Result |
|---|---|---|
| IV rank | (14 − 9) ÷ (24 − 9) × 100 | 33 |
| IV percentile | 230 ÷ 250 × 100 | 92 |
| Reading | Rank says volatility is in the lower third of its range; percentile says it is higher than on nine days in ten. The spike stretched the range and made rank understate how elevated 14% is relative to a typical day. | |
| ATM straddle at 14%, 30 days | 451 + 323 | ₹774 per unit, ₹50,310 per lot |
| Same straddle at the 9% low | Black-Scholes at 9% | about ₹503 per unit; the 14% reading is pricing about 54% more movement than the year’s low |
A trader who looked only at rank might call 14% cheap and favour buying the straddle; one who looked at percentile would call it rich and favour selling it. The disagreement is the information: the year contained one spike and a long quiet stretch, and the right question is whether the next 30 days look more like the spike or the stretch.
Why it matters
- It frames the choice between debit and credit structures. A common rule of thumb reads rank below about 25 as a zone where buying options and debit spreads cost less relative to history, and above about 75 as a zone where written premium is large relative to history. The rule is about relative cost, not about outcome.
- It normalises across underlyings. A 20% IV is unremarkable on MIDCPNIFTY and extreme on NIFTY; rank puts each on its own scale.
- It flags event pricing. Rank and percentile both jump before scheduled events as IV rises, and fall after, which is the IV crush that short-dated buyers suffer.
- It is only a context measure. High rank can go higher in a crisis and low rank can stay low for months; the number describes the past year, not the next month.
Common mistakes
- Using rank without percentile. After a spike, rank understates how elevated current IV is for most of the following year.
- Computing it on the wrong series. The IV of a weekly 2-day option or a far out-of-the-money strike is not comparable day to day; use the 30-day ATM IV or India VIX.
- Treating high rank as a reason to sell on its own. High IV is often high because realized volatility is high; the Vol Spread between implied and realized is the better test of whether premium is actually rich.
- Ignoring the lookback. A 52-week window that includes a crisis gives very different numbers from one that does not; know what the range contains.
Indian specifics
India VIX is the natural series for NIFTY IV rank because NSE publishes it daily on a consistent 30-day basis; for other indices and for stocks, the 30-day ATM IV must be computed from the chain. NIFTY’s weekly Tuesday expiries make the nearest-expiry IV unsuitable for rank, since it swings with days to expiry rather than with the market’s volatility view. Indian index IV ranges have historically been compressed outside crisis periods, so a rank of 50 on NIFTY often corresponds to an IV in the low teens, while the same rank on a single stock around results can be far higher in absolute terms.
FAQ
What is IV rank?
Where today's IV sits between its 52-week low (0) and high (100), computed on a consistent series such as 30-day ATM IV.
What is the difference between IV rank and IV percentile?
Rank = position in the year's range; percentile = share of days with lower IV. A single spike lowers rank but not percentile.
What is a good IV rank for selling options?
Above about 75 is commonly read as premium being rich relative to the year; it is a relative-cost reading, not a forecast.
Which IV should IV rank be calculated on?
A constant-maturity series such as 30-day ATM IV or India VIX, not a single strike or the nearest weekly expiry.
Can IV rank be above 100 or below 0?
No; a new high reads 100 and a new low reads 0.
IV Rank on the chain, read against the last 52 weeks
The TradePulse option chain’s analytics strip shows IV Rank next to ATM IV and the Vol Spread, flagging below 25 as cheap and above 75 as rich relative to the past year.
Related terms
- ATM IV, the series rank is computed on
- IV percentile, the share-of-days alternative
- India VIX, NSE’s 30-day series
- Implied vs realized, the other rich-or-cheap test
- IV crush, why rank falls after events
- Vega, what a change in IV is worth