IV Percentile calculator

Count how many of the last year's days closed with a lower implied volatility than today's. Every day counts once.

Quick answer: IV Percentile is the share of days in the lookback window whose implied volatility closed below today's. Unlike IV Rank it uses every observation, which makes it robust to the single spike that can distort a 52-week range for a year.

IV Percentile Calculator

Paste one year of daily IV readings, separated by commas or newlines. The calculator also shows the IV Rank of the same data for comparison.

Figures are per unit of the underlying and exclude brokerage, STT, exchange charges, stamp duty and GST.

How this calculator works

Every day counts once

IV Percentile asks a simple question: of the days in the window, what fraction closed below today? A single stress spike contributes exactly one day, the same as any quiet Tuesday. That is the whole difference from IV Rank, and it is why the percentile survives an outlier that destroys the rank's denominator.

Why it reads higher than IV Rank

Volatility's distribution is right-skewed: most days cluster in a narrow band, and a thin tail stretches far to the right. The 52-week HIGH therefore sits a long way above the typical day, while the 52-week LOW sits close to it. Today lands low in the RANGE and high among the DAYS. On the dataset shipped with this calculator, IV Rank is about 18 and IV Percentile about 62 — for the same day, from the same data.

It still cannot see an event

Neither metric knows whether a scheduled event sits inside the current option's remaining life. An implied volatility in the 80th percentile the week before the Union Budget is not 'high': it is correctly pricing a day that has not happened. Both numbers will tell you options are expensive, and both will be describing a fact about the calendar rather than about the option market.

Regime shifts poison the window

A percentile assumes the days in the window are drawn from the same distribution. For an underlying whose volatility regime changed mid-year — a stock that levered up, an index after a structural change in participation — the older half of the window describes a different asset. The percentile will be precise and meaningless.

IV Percentile

IVP = (count of days with IV_i below IV_today) ÷ n × 100

A rank statistic. It uses only the ORDER of the observations, not their magnitudes, which is exactly what makes it robust to outliers — and also what makes it blind to how far above the crowd today actually is.

IVPIV Percentile, on a scale of 0 to 100.
IV_todayToday's implied volatility.
IV_iThe implied volatility on the i-th day of the lookback window.
nNumber of days in the window. Conventionally 252 trading days, i.e. one year.

Using it, step by step

Collect one year of daily at-the-money implied volatility readings for a consistent tenor.
Paste them into the box, separated by commas, spaces or newlines. Order does not matter — the metric is a count.
Enter today's reading.
Read the IV Percentile, and read the IV Rank the calculator shows beside it.
If the two differ sharply, look at the histogram: you will see the single spike that stretched the range and left the rank low.

Worked example

NIFTY

A year of readings clusters between 9% and 15%, with a brief excursion to 30% during one stress episode. Today reads 12.4%. Counting days, about 62% of them closed below 12.4%, so IV Percentile is about 62 — a middling, slightly rich reading. IV Rank on the identical data is about 18, which reads as exceptionally cheap. The spike contributed one day to the percentile and the entire top of the range to the rank. If you were deciding anything on the strength of 'IV Rank is 18', you were deciding it on the basis of two days out of two hundred and fifty-two.

Assumptions and limitations

What this calculator assumes. Every number it produces is only as good as the assumptions below, and each of them is wrong to some degree in a real market. The output is a model's opinion, not a measurement.

All readings come from the same tenor and the same measurement convention. Splicing near-month implied volatilities across expiry rollovers introduces jumps that are contract artefacts, not volatility.
The days in the window are treated as drawn from one distribution. A mid-year regime shift makes the percentile precise and meaningless.
252 trading days is the convention. Some platforms use 365 calendar days, which changes the answer; the number should always be quoted with its window.
The metric uses only ordering, so it cannot tell you HOW far above the crowd today sits. An IVP of 99 is the same whether today is 1% or 40% above the 98th percentile day.
Like IV Rank, it is blind to scheduled events inside the option's life, which are frequently the entire reason the reading is where it is.

Common mistakes

Using IV Percentile and IV Rank interchangeably. They answer different questions and routinely differ by tens of points.
Reading a high percentile as evidence that options are overpriced. It is evidence they are expensive relative to this underlying's own year, which is usually because larger movement is coming.
Building the history from nearly-expired contracts, whose implied volatilities are artefacts of an almost-zero premium.
Applying a percentile computed on an index to a single stock, or comparing percentiles across underlyings as though they were the same scale.
Ignoring that an IVP of 99 conveys nothing about how much further volatility can go — the metric is bounded and volatility is not.

Frequently asked questions

What is IV Percentile?

IV Percentile is the percentage of days in the lookback window whose implied volatility closed below today's. An IVP of 62 means today is higher than about 62% of the last year's days.

How is IV Percentile calculated?

Count how many days in the window had an implied volatility lower than today's, divide by the number of days, and multiply by 100. Every day counts exactly once, regardless of how extreme it was.

Why is IV Percentile usually higher than IV Rank?

Because volatility's distribution has a long right tail. The 52-week high sits far above a typical day, stretching IV Rank's denominator, while most days cluster near the low. Today therefore lands low in the range and high among the days.

Which is better, IV Rank or IV Percentile?

IV Percentile is more robust because it is not defined by two extreme days. The best practice is to compute both: when they agree the reading is unambiguous, and when they diverge by more than twenty points a spike is distorting the rank.

What is a high IV Percentile?

Conventionally above 50, and often above 70 or 80. These are conventions. What matters more is whether a scheduled event sits inside the option's remaining life, which no percentile can see.

How many days of history do I need?

252 trading days — one year — is the convention, because it spans a full cycle of scheduled events. Fewer days makes the metric noisy; more days risks including a volatility regime that no longer applies.

Does IV Percentile predict future volatility?

No. It is a description of where today sits in a past distribution. Volatility does mean-revert, so an extreme percentile is more likely to fall than to rise — but mean reversion says nothing about when, and it does not prevent the reading from doubling first.

Can IV Percentile be 100?

Yes, whenever today's reading exceeds every day in the window. At that point the metric stops conveying information, because it cannot tell you how much further volatility might go — and that is precisely the situation in which you most want to know.

Voice search & related questions

What does IV Percentile mean?

It means the fraction of days over the past year whose implied volatility was lower than today's. An IV Percentile of 62 means today is more expensive than about 62% of the last year's days.

Why do IV Rank and IV Percentile disagree?

Because IV Rank divides by a range that a single spike can define, while IV Percentile counts every day equally. Volatility has a long right tail, so the rank usually reads lower.

How much IV history do I need for IV Percentile?

One year, or 252 trading days, is the standard window. It spans a full cycle of scheduled events like budgets, policy meetings and earnings seasons.

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