What is open interest in trading?

Everything about open interest follows from one idea borrowed from accounting: the difference between a stock and a flow. A flow is measured over a period and then resets, the way a day's rainfall is counted and the gauge emptied for tomorrow. A stock is measured at an instant and carries over, the way the water level in a reservoir is what it is right now, regardless of how much rain fell to fill it. Volume is a flow: it counts every contract that trades in a session and returns to zero at the close. Open interest is a stock: it counts the contracts still open at a moment and carries that level forward, overnight and across weeks, until the positions themselves are closed or expire.

That is why the two numbers can tell completely different stories about the same day, and why confusing them is the root of almost every open-interest mistake. A market can churn furiously, printing huge volume, while its stock of live contracts barely changes, because the same contracts are being passed from hand to hand rather than newly created. Equally, a quiet session with modest volume can add a large, durable stock of new positions if most of that trading was opening rather than closing. The flow tells you how busy the day was. The level tells you how much is still at stake when the bell rings.

Open interest is a level, not a flow. It can tell you how much conviction is committed, and it can never tell you which way that conviction will resolve.

The reason for that last, blunt limit is structural rather than a matter of interpretation. Every derivative contract is an agreement between two parties: for each open long there is exactly one open short, and open interest is simply the count of those pairs. The longs and the shorts are always equal, because they are two ends of the same contracts. So the number is directionless by construction. It can grow to record and shrink, but it can never lean bullish or bearish, because it is counting balanced pairs, not net bets. Hold on to that and the rest of the guide is a set of consequences.

A derivative contract does not exist until a trade brings it into being. When a buy order and a sell order match, the clearing corporation registers one long position against one short position, and that single pair is one unit of open interest, whether the contract is a future or an option. What the next trade does to the count depends entirely on what the two sides were each trying to do, and there are only four possibilities, built from whether each side is opening a new position or closing one it already holds. The grid below is the whole mechanism.

Two consequences matter for the rest of the guide. First, because open interest is a stock, it is genuinely comparable from one day to the next: today's level and yesterday's level are the same kind of measurement, so the change between them is meaningful in a way a change in daily volume is not. Second, a contract can leave the count with no closing trade at all. When a series is exercised or reaches expiry, every position still open is settled by the clearing corporation, and open interest for that expiry collapses to zero without anyone placing a sell order. Watching the near-month count drain in expiry week while the next month's count swells is watching settlement and rollover, not a change of heart. Futures and options carry open interest in exactly this way; where they differ is in payoff, not in how the count works, a distinction laid out in futures versus options in India.

Walk it from an empty series to see the transfer in action. A buys a lot from B, both opening, so open interest goes from zero to one. C buys a lot from D, both opening, and it rises to two. A then sells out to a newcomer, E, who is opening: A closes while E opens, so the contract merely changes owner and open interest holds at two, even though volume has climbed to three. Only when two existing holders finally close against each other does the count fall back. Four trades, four units of volume, and an open-interest line that went one, two, two, one. Volume recorded all the activity; open interest recorded only what survived it.

This is the confusion the whole guide turns on, so it deserves a picture. Volume answers the question, how much changed hands today? Every matched contract adds to it, whatever the two sides intended, and the counter is wiped clean at the close. Open interest answers a different question, how much is still at stake? It moves only when the population of live contracts changes. The two diverge in exactly one way that matters: a position opened and closed within the same session adds to volume twice and to end-of-day open interest not at all. Hold volume constant and open interest can do anything, and the figure below shows the two extremes of that freedom.

Scale the survivors and the distinction stops being academic. Take an illustrative stock futures contract with a lot of 1,100 shares on a stock near ₹500: one contract carries a notional of 1,100 multiplied by ₹500, or about ₹5.5 lakh. If the series ends the day with 40,000 contracts of open interest, roughly ₹2,200 crore of live, margined exposure is outstanding and must eventually be closed, rolled or settled. A volume print of 40,000 tells you nothing of that kind: it might be 40,000 fresh contracts, or ten contracts churned four thousand times between intraday desks. The exposure that survives the close is the number that matters for risk, which is exactly why the regulator writes its position limits against open interest rather than volume. Volume has its own honest uses, as the flow side of this pair, and those are the subject of what is volume in trading; the point here is only that it answers a different question.

One practical wrinkle follows from open interest being a settled, cleared figure rather than a tape print. The authoritative number is struck at the end of the day, once the clearing corporation has reconciled every position, and it is that end-of-day figure the exchanges publish in the bhavcopy and the ban-list files. The change in open interest shown on a live option chain through the session is a useful provisional estimate, not the final word, and it can be revised once the day is settled. That is one more reason to read the level as a considered daily measurement rather than a tick-by-tick indicator: it is at its most reliable exactly when the session is over and there is nothing left to churn.

Because open interest is a comparable level, the useful daily read is not its absolute size but its change, set against the change in price. Rising open interest means the move is being built on fresh commitment: new contracts are being created and margined at these prices, so someone is willing to add exposure here. Falling open interest means the move is being powered by exits, and exits are finite fuel: a rally driven by shorts buying back their positions ends when the shorts are done, with no new owner standing behind the price. That one distinction, new money against leaving money, generates the four combinations in the grid below.

Three limits keep the grid honest, and they are the reason each cell is a hint rather than a signal. First, open interest is anonymous: it counts contracts, not intentions, so it cannot say who opened a position, at what average price, or with what conviction. Second, it aggregates: one open-interest number folds directional bets, hedges against cash holdings, arbitrage pairs and the individual legs of multi-leg option structures into a single count, so rising open interest on a falling price is sometimes patient cash-and-futures arbitrage rather than aggressive new shorts. Third, every label is a classification of the session that just ended, not a forecast of the next. The grid tells you what the crowd just did. It is silent on what the crowd will do, and the table restates each cell with exactly that honesty.

A quieter discipline matters just as much: a change in open interest only means something against a yardstick. A rise of a few thousand contracts is noise in a series that routinely carries lakhs of them and news in a thinly held one, so the same absolute change reads differently depending on the base it sits on and the typical daily swing for that contract. The honest way to use the grid is therefore relative. Compare today's change in open interest to the contract's own recent range rather than to a fixed number, and treat a build-up or an unwind as notable only when it is genuinely large for that particular series. A number without its yardstick is the most common way an open-interest read goes wrong.

The most repeated claim about open interest is that a large reading at an option strike is a wall: resistance at the strike with the biggest call open interest, support at the strike with the biggest put open interest. It is a half-truth, and it is worth seeing exactly which half is true. The true half is that a heavy strike is a level the market is genuinely watching, and that large written positions there can behave like a barrier, because the writers who sold them have both an incentive to defend the level and hedging flows that dampen movement around it. The missing half is decisive: open interest cannot show whether that strike is dominated by those writers or by buyers who want it to break. The same number points in opposite directions depending on a fact it does not contain.

In practice the change tells you far more than the level. A wall that grows as price approaches it is being actively contested, with new writers adding to defend it; a wall that unwinds as price approaches has already been abandoned, and the level tends to give way without a fight. So the day's change-in-open-interest column is the argument currently being had, while the absolute figure is only history. Reading that whole distribution across strikes and expiries, and pairing it with the other columns of the chain, is a craft in itself, and it is the subject of how to read an option chain in India. The single lesson to carry from this figure is narrower and firmer: because you cannot see long against short, no open-interest level is ever a guarantee, and the loudest ones can break the hardest.

Held that way, a wall becomes a hypothesis with an invalidation rather than a barrier with a promise. If price approaches the largest call strike and that strike's open interest keeps growing, the writers are still defending and the level has committed money behind it; if the same strike's open interest starts falling as price arrives, the defenders are stepping aside and a break becomes more likely, not less. Neither pattern is a trade on its own. Both are simply the level paired with the direction of its change, which is far more than the headline figure most commentary stops at, and it is the difference between watching a number and reading it.

The same open-interest data feeds a family of popular derived numbers, and each is honest as a description and dangerous as a forecast. The most cited is max pain: the expiry price at which the total payout owed to all option buyers across the chain would be smallest, which is equivalently the point most favourable to option writers in aggregate. The convention holds that price drifts toward it as expiry approaches. Because it is arithmetic, it can be computed exactly from the open interest at each strike, and the figure below does precisely that, so you can see what the number is and what it is not.

The put-call ratio, or PCR, is the other favourite: the total put open interest divided by the total call open interest, read as a gauge of how bearish or bullish the positioning is, and often as a contrarian one at extremes. It carries every caveat the wall figure just made, only aggregated across the whole chain: it is a ratio of committed contracts that folds hedges, spreads and directional bets into a single figure, it has no universal threshold that means anything, and a high or low reading describes a crowd's position rather than predicting its fate. Rollover percentage is a third derived read, the share of an expiring series' open interest that carries into the next series, and it measures how much positioning is being retained rather than which way it leans; because it is a topic in its own right, the mechanics and the honest way to read it are set out in what is rollover in futures.

The practical upshot is a habit of translation. Whenever one of these derived numbers is quoted at you as a forecast, translate it back into the open interest it came from and ask what that raw data can support. Max pain becomes where the least is owed to buyers today, not where price must go. A put-call ratio becomes a snapshot of how many put contracts are open against call contracts right now, not a verdict on the next move. A rollover percentage becomes how much positioning chose to carry forward, not which way it leans. Translated back, each read is informative and none is a signal, which is exactly the discipline the raw count has been teaching from the first figure.

In India, open interest is not only something you watch; it is something the regulator acts on, which is the part most general explanations omit. Every stock in the derivatives segment carries a market-wide position limit, or MWPL, a ceiling on the combined open interest across all futures and all option strikes on that stock taken together. When combined open interest crosses 95 percent of the limit, the exchange moves the stock into a ban period in which no participant may add fresh exposure, and the stock leaves the ban only when open interest unwinds back below 80 percent. The exchanges publish the ban list before every session, so it is a public, mechanical trigger rather than a judgement call, one of the few moments in markets where a rule, not an opinion, decides what may happen next.

The framework was rewritten in 2025, which is why older articles describe a regime that no longer applies. As of 17 July 2026, under SEBI's revised measures, the open interest measured against the limit is a delta-adjusted, futures-equivalent figure rather than a raw contract count, so a deep out-of-the-money option no longer weighs the same as a futures contract; the flat freeze on new positions has become a risk-reduction rule, under which trades during a ban are allowed only if a participant's futures-equivalent exposure in that stock is lower by the end of the day than at the start; and utilisation began to be monitored through the session rather than only at the close. These are the operative principles, but the precise thresholds and effective dates are the kind of number that moves, so confirm the current figures against SEBI's own circular and the exchange ban-period rules before you rely on them.

One last point completes the picture, and it is a genuinely useful read rather than a caveat. A ban is itself information. Because only risk-reducing trades are allowed once a stock is in the ban period, its combined open interest can only hold or grind lower until the 80 percent release, never build, so participants who follow single-stock derivatives watch utilisation climbing through the eighties precisely because the 95 percent line is a known, mechanical wall. It is the one place in this guide where an open-interest level does force an outcome, and even there what it forces is a restriction on trading, not a direction for the price.

There is a second reason to read Indian open-interest history carefully, and it has nothing to do with sentiment. The regulatory response to retail losses in derivatives physically moved where open interest sits. The backdrop is stark: in its study of September 2024, SEBI found that about 93 percent of individual traders in equity derivatives made net losses over FY22 to FY24, with aggregate net losses exceeding ₹1.8 lakh crore. The measures that followed reshaped the index-derivatives landscape, and two of them changed the open-interest numbers directly rather than merely the behaviour behind them.

The first was consolidation of weekly expiries. Each exchange now runs weekly expiries on a single benchmark index rather than spreading them across several, so the weekly option open interest that once distributed across many contracts concentrated into far fewer, and expiry-day positioning now stacks where it used to spread. The second was a resizing of index contracts to carry a larger notional per contract, in a band of roughly ₹15 lakh to ₹20 lakh of underlying value. Both are regulatory facts as of 17 July 2026 and worth confirming at the exchange source, because dated thresholds move; the point for a reader of open interest is what the resizing did to the count.

Set against the whole picture, open interest is the derivatives market's census: it records how many commitments exist and where they sit, and it is honest and precise about exactly that. Used within its boundary it does a small number of real jobs well. Against price, it classifies a move as fresh positioning or as exit fuel. Across an option chain, it maps where obligations cluster and which levels the market is currently arguing over. Against the market-wide position limit, it warns when a single stock is approaching a mechanical trading restriction. Each of those is context that disciplines a read. None of them is a signal that replaces one, and the ledger below draws the line between the two.

The failure mode is always the same act: promoting a count into a forecast. High open interest does not make a level hold, a long build-up does not oblige a rise to continue, and max pain does not summon the price. Everything open interest offers is on the left of that ledger, and every popular myth about it lives on the right. Deciding which levels deserve attention in the first place, and stating in advance what evidence would invalidate a read, is upstream judgement rather than anything the count provides, and that upstream discipline is exactly what the method we teach is built to train. The count is public and free. The judgement about how much weight it can bear is the part worth learning, and it begins with the one sentence this guide keeps returning to: open interest tells you how much conviction is committed, never which way it will resolve.

• Indian retail derivatives outcomes. The Securities and Exchange Board of India, in its study of September 2024, reported that about 93 percent of individual traders in the equity derivatives segment made net losses over FY22 to FY24, with aggregate net losses exceeding ₹1.8 lakh crore, the backdrop to the framework changes described here. sebi.gov.in
• The revised position-monitoring framework. SEBI's measures for the equity derivatives segment moved position monitoring to a delta-adjusted, futures-equivalent open-interest basis and revised the market-wide position limit mechanics; the operative thresholds and dates should be confirmed against the current SEBI circular. sebi.gov.in
• The F&O ban-period mechanics. The exchanges publish the daily list of securities in the ban period, the 95 percent market-wide-position-limit trigger for entry and the 80 percent release on unwinding; these are published before each session on the NSE and BSE websites.
• Open interest data itself. Per-strike and per-contract open interest and the day's change in open interest are exchange data, published through the session and in the end-of-day bhavcopy files by NSE and BSE, and mirrored by trading platforms from the same feed.
• On the reads treated with caution. Max pain, the put-call ratio and rollover percentage are conventions computed from open interest; this guide follows the standard definitions and, where the literature and market practice disagree on their predictive value, states the limit rather than the claim.