Home / Reference / IV Decision Tree

The IV decision tree

A structured way to interrogate a volatility number. Every branch ends in understanding; none of them ends in an instruction.

Quick answer: The IV decision tree asks five questions in order — which volatility is this, high compared with what, does a scheduled event sit inside the option's life, what does the term structure say, and what is realised volatility doing — and narrows what you are looking at without ever telling you what to do about it.

A structured way to interrogate a volatility reading. It narrows what you are looking at; it never tells you what to do, because that depends on facts about you — your capital, your other positions, your tolerance for a bad month — that no web page can know.

Interrogating a volatility reading

Read left to right. Every path ends in understanding, not in an instruction.

A volatility numberwhich one is it?impliedhistorical/realisedImplied volatilitycompared with what?Realised volatilityover which window?rank ≈ percentilethey disagreeThe reading isunambiguousA spike distorts the ranktrust the percentileIs a scheduled eventinside the option's life?yesIV is not "high" — it iscorrectly pricing a daythat has not happenedCompare like with like30-day IV vs 30-day RVnot 30-day IV vs 10-day HVIV > RV: seller is paidIV < RV: buyer is paidneither is an edge aloneThe tree narrows the question.It never answers it. Liquidity,spread, margin and position sizesit outside it, and decide most outcomes.
No branch of this tree ends at "buy" or "sell", because a recommendation requires knowing your capital, your existing positions and your tolerance for a bad month — none of which a web page can know.

Question 1 — which volatility is this?

Before anything else. A screener showing "IV 34%" for a mid-cap stock and a news article citing "India VIX at 13" are not comparable numbers, and neither is comparable to a 20-day historical volatility you computed in a spreadsheet. See the comparison matrix.

Question 2 — high or low compared with what?

Compute IV Rank and IV Percentile on the same year of data. If they agree, the reading is unambiguous. If they disagree by more than about twenty points, one extreme day is distorting the rank, and the percentile is the more honest number.

Question 3 — is there an event inside the option's life?

If a scheduled event — RBI policy, the Union Budget, election counting, an earnings release — falls before expiry, the implied volatility is not "high". It is correctly pricing a day that has not happened. Selling it is selling insurance against a specific known risk, and the crush you are hoping for is the market's payment for having borne it.

Question 4 — what does the term structure say?

Contango and backwardation are answering a different question from IV Rank. Rank tells you where the level sits against its own history; the slope tells you whether the market thinks today's level is temporary. A high IV Rank in steep backwardation is a frightened market. A high IV Rank in contango is usually an event.

Question 5 — what is realised volatility doing?

The only comparison that speaks to whether an option is expensive is implied against realised over a comparable window. If implied is 18% and the underlying has been realising 22%, the option is cheap despite a high IV Rank.

Where the tree ends. Every branch above ends at "now you understand what you are looking at." None of them ends at "therefore buy" or "therefore sell." Liquidity, bid-ask spread, margin, position size, your existing book and your ability to survive the worst two weeks all sit outside this tree, and they decide most outcomes.

Frequently asked questions

How do I know whether implied volatility is high?
Compute IV Rank and IV Percentile on the same year of data. If they agree the reading is unambiguous; if they diverge by more than about twenty points, one extreme day is distorting the rank and the percentile is more honest.
What if a scheduled event falls before expiry?
Then the implied volatility is not high — it is correctly pricing a day that has not happened. Selling it is selling insurance against a specific known risk, and the crush you are hoping for is the market's payment for having borne it.
Does a high IV Rank mean options are overpriced?
No. It means they are expensive relative to this underlying's own past year. Whether they are overpriced depends on what the underlying subsequently realises, which nobody knows yet.
What does the term structure add that IV Rank does not?
IV Rank tells you where the level sits against its own history. The slope tells you whether the market believes today's level is temporary. A high IV Rank in steep backwardation is a frightened market; a high IV Rank in contango is usually an event.
Why doesn't the decision tree end in buy or sell?
Because liquidity, bid-ask spread, margin, position size, your existing book and your ability to survive the worst two weeks all sit outside the tree, and they decide most outcomes. A recommendation would require knowing things a web page cannot know.
What is the single most useful comparison?
Implied volatility against realised volatility over a window of the same length. It is the only comparison that speaks to whether an option is expensive relative to what the underlying is actually doing.

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

Educational content only — not investment advice. See our Risk Disclosure and Methodology.