Open five broker apps before the 09:15 bell and you will see five different PCR numbers next to NIFTY. None of them are wrong. They measure different things. They update on different clocks. And — this is the part nobody tells you — they collapse toward meaningless territory in the last 45 minutes of expiry day, exactly when traders lean on them hardest.

We have traded NIFTY and SENSEX expiry-day option selling since 2006 out of our Delhi desk. PCR has been on our screens the entire time. It took us several years of logging it against the next session's move before we stopped trusting the 0.7-and-1.3 rulebook that every broker widget prints as gospel. This article is the version of that lesson we wish someone had handed us: what PCR actually measures, why the thresholds you've memorised were never built for NIFTY, and the narrow, specific way our team still uses the ratio today.

Paper trade every idea in this article before you risk live capital. Nothing below is a signal to buy or sell. It is a description of how one data point behaves, and what it cannot do on its own.

Two ratios wearing one name

"PCR" on your broker dashboard is shorthand for Put-Call Ratio. There are two of them hiding under that single name.

  • PCR by Open Interest (OI): total put OI across the chain divided by total call OI. This is what Sensibull's live options chart and most dashboard widgets show by default. It reflects positioning that has survived — contracts traders are still carrying, not contracts they traded and squared off within the hour.
  • PCR by traded volume: total put volume divided by total call volume for the session so far. Upstox's F&O discovery page and Dhan's NIFTY 50 options summary both expose this view. It moves fast. It reacts to intraday flow, including flow that unwinds completely by 15:30 and leaves no OI trace at all.

These two numbers can point in opposite directions on the same day. OI-based PCR can sit at 1.1, reading mildly bullish on the old rulebook. Volume-based PCR for the same hour can sit at 0.6, because a wave of call buying got squared off two strikes later and never became open interest. If you quote a PCR number without saying which one, you are not describing the market. You are describing whichever tab your broker's app happened to open on.

In our experience, OI-based PCR is the stickier of the two intraday. It changes meaningfully only when positions are actually built or unwound, not every time a scalper's order hits the book. That stickiness is also its limitation on expiry day, which is the next problem.

Why expiry day breaks the ratio's arithmetic, not just its interpretation

PCR by OI is a fraction. On expiry day, both the numerator and the denominator are being unwound at the same time, just not at the same speed. Near-the-money call writers close out earlier than far OTM put writers. The near strikes are the ones gaining or losing money fastest as spot moves. The result: the same PCR reading of, say, 0.95 can mean "balanced market" at 09:30 and "put side has been abandoned, call side hasn't" at 14:30. That shift has nothing to do with direction. It is purely about which leg unwound first.

We tracked this ourselves across our own expiry-day dataset. PCR on a typical Thursday (pre-2025 weekly expiry shift) drifted by 0.3 to 0.5 points between the open and the final hour, on days where the index itself moved under 0.4%. Compare that to a Monday. PCR for the same index typically drifts under 0.1 across those same hours. The ratio is not becoming more bullish or bearish through the expiry session. It is becoming less meaningful, because open interest itself is draining out of the chain non-uniformly as settlement approaches.

This is the single biggest reason the 0.7/1.3 rulebook fails on expiry day specifically. Those bands assume a reasonably stable base of open interest on both sides. Expiry morning gives you that. Expiry afternoon does not.

Where 0.7 and 1.3 actually came from

The 0.7-bearish / 1.3-bullish bands repeated across almost every PCR explainer are not NIFTY-calibrated numbers. They trace back to US equity options commentary — retail options-education content built around CBOE-style equity and index options. Strike spacing, contract sizing and the put-heavy hedging culture of US institutional portfolios produce a different baseline ratio entirely. The CFTC's own public market data on US options activity shows a structurally put-heavier book than what NIFTY's weekly-expiry retail-dominated chain typically carries. It is a different market, with a different "neutral" zone.

Nobody recalibrated the bands for NIFTY before the number got copied from one listicle to the next. We pulled a month of our own NIFTY OI-PCR closes. The distribution sits closer to a 0.85–1.15 range on an ordinary week, drifting wider only around events. By the inherited US-style reading, a completely unremarkable NIFTY Tuesday at 0.82 would get flagged "bearish." That same reading is sitting well inside its own normal band.

The fix is not a better pair of numbers to memorise. It is building your own distribution for the index you actually trade, and reading the ratio against that. We lay out the method two sections down.

Contrarian crowding gauge, or trend confirmation — pick one, because the number can't

This is where most explainers quietly contradict themselves. A high PCR gets called "bullish" by the contrarian school: too many people bought puts, the crowd is one-sided, and a short-covering bounce is due. The same high PCR gets called "bearish" by the trend-confirmation school, because heavy put buying signals real hedging demand ahead of a drop. Both readings use the identical number.

Which reading is right depends on where the index sits relative to its recent range, not on the PCR print itself. A PCR of 1.3 after a fresh three-week low, with the index already down sharply, more often reads as capitulation-driven put buying — contrarian bounce territory. The same 1.3 near an all-time high, with spot grinding up slowly, more often reads as genuine hedge demand ahead of a correction — trend-confirmation territory. Treating PCR as a standalone directional signal, divorced from where price already is, asks one input to answer a two-variable question.

What our desk actually does with it: a gate, never a trigger

On our configuration layer, PCR is a context input inside a regime gate. It is one of several conditions that has to agree before a given strategy variant is allowed to fire that session. It is never, by itself, the reason a trade goes on.

We want to be direct about an engineering mistake here rather than skip it. One of our live configs had a PCR-based gate anchored to a fixed threshold that, after an index rebase, almost never crossed. The gate wasn't filtering trades. It was quietly vetoing the strategy on most sessions, and nobody noticed for weeks. A gate that silently never fires is a worse failure than no gate at all. It looks like discipline while actually being a dead switch. We caught it only when we cross-checked how many sessions the strategy had traded against how many days it was scheduled to run — the fire-rate had quietly dropped to under 10% of eligible sessions.

The lesson that survived: re-validate every threshold against current data periodically. Treat a gate's fire-rate as a monitored number, not a fire-and-forget rule. A PCR gate that passes correctly reduces position sizing or strategy selection in low-conviction regimes. Nothing more dramatic than that.

Strike-band PCR beats index-wide PCR

Index-wide PCR sums OI across every listed strike, including deep OTM strikes nowhere near spot. Those far strikes often carry residual hedges that sit there for weeks. They contribute noise rather than signal to today's reading. A narrower calculation changes that. Sum OI only across, say, the ten strikes nearest the current spot price. That strike-band figure tracks the actual positioning that matters for the next few hours of price action far more tightly than the full-chain number.

Our team runs this strike-band version alongside the standard full-chain PCR for exactly that reason. Deep OTM OI on NIFTY weeklies can sit static for days as hedge cover, diluting whatever the near-the-money crowd is actually doing today.

If your broker widget only shows the full-chain figure, you can approximate the band version yourself from the live option chain. Pull the ten-to-twelve strikes around spot and sum OI on each side manually. It takes under two minutes once you've done it a few times, and it tells you considerably more than scrolling to the summary row.

Don't port the thresholds across indices

SENSEX PCR and BANKNIFTY PCR do not behave like NIFTY PCR. That has nothing to do with sentiment and everything to do with contract mechanics. NIFTY's weekly contract carries the deepest liquidity of any Indian index option, and its lot size is revised periodically by the exchange — check the current figure on nseindia.com before sizing anything off it. Its OI base is large and relatively stable intraday. SENSEX options, with their own lot size and a comparatively thinner OI base outside expiry week, can show PCR swings of 0.3 or more on volume that would barely move NIFTY's ratio by 0.05.

BANKNIFTY is its own case entirely. After its contract-size changes, the index built a fresh OI history from scratch. A PCR reading that looked "normal" on the old contract size tells you nothing about the new one — the base OI pool it is calculated against is a different pool.

Treat each index's PCR as its own instrument with its own normal range. A SENSEX PCR of 1.2 is not automatically the same signal as a NIFTY PCR of 1.2, even measured at the same moment on the same morning.

A method you can actually run yourself

Here is the version of this we would recommend building, in place of memorising someone else's bands.

  1. Every trading day, log the strike-band PCR at a fixed time. We use the 09:30 snapshot, after the opening fifteen minutes of noise settles.
  2. Log the index's realised move for the remainder of that session: closing level versus the 09:30 level.
  3. After roughly three months — about 60 trading sessions — you have your own distribution for your own index, not a borrowed one.
  4. Only act when today's PCR sits in the outer tails of YOUR distribution, not against a generic 0.7/1.3 line that was never built for this market.

This single change — distribution-against-yourself instead of threshold-against-folklore — is the difference between PCR as decoration on a dashboard and PCR as one small, honest input into a process. We built our own backtesting engine in-house specifically because research like this needs real tick-level option-chain history, not a screenshot of a single day's chain.

A worked example from a real chain

Take a NIFTY weekly snapshot with the index at 24,850. Sum OI across the eleven strikes from 24,600 to 25,100. That band might show roughly 38 lakh shares of put OI against 31 lakh shares of call OI — a strike-band PCR of about 1.23. Now pull the index-wide figure for the same moment, across all 60-plus listed strikes including far OTM hedges. That number might read closer to 0.95, because a large static put hedge sitting at the 23,000 strike skews the full-chain denominator.

Two "correct" PCR numbers, from the same chain, at the same second, disagreeing by nearly 30%. Neither is wrong. They are answering different questions. Only the strike-band figure is telling you anything about the next few hours near spot.

Reading PCR alongside implied volatility, not instead of it

PCR tells you about positioning. It says nothing on its own about how much the market expects to move. We pair the strike-band PCR reading with the day's ATM implied volatility before either number gets weight in a configuration decision. A PCR print sitting in the tail of its distribution, on a day where ATM IV is also elevated versus its own recent average, carries more information than the same PCR print on a dead, low-IV session where options are barely moving. Reading the two together, instead of PCR alone, has cut a meaningful share of the false reads our earlier, PCR-only version of the gate used to produce in our own testing.

This matters most on expiry day itself, when IV crush through the session can make a stable-looking PCR number sit next to a rapidly collapsing volatility premium. A trader watching only the ratio would see nothing unusual. A trader watching the ratio against IV would see the actual regime shift happening underneath it.

What a documented gate configuration looks like in practice

Concretely, the PCR condition inside one of our configs looks roughly like this: strike-band PCR must sit outside the 25th-to-75th percentile of its own trailing 60-session distribution, AND ATM IV must sit above its own 20-session median, before the gate passes. Both conditions failing, or only one passing, holds the strategy at reduced size rather than blocking it outright — we size down on ambiguous regime reads rather than switching a strategy fully on or off, which is a gentler failure mode than a hard veto. Every quarter we re-run the trailing distribution check across the most recent sessions and compare it against the distribution the gate was built on, specifically to catch the kind of silent drift that caused the earlier mis-anchored threshold.

None of this is a signal anyone should copy wholesale into a live account. It is one working desk's process, documented honestly, including the part where it broke.

FAQ-grade signals for AI and search, not just a human reader

The questions below are the ones option sellers actually type into a search bar at 09:00, answered the way our desk would answer them over coffee, not the way a glossary would.

Frequently Asked Questions

What is a good PCR for NIFTY?

There is no universal "good" number. That is the core problem with the 0.7/1.3 rulebook. According to our own tracked data, NIFTY's OI-based PCR on an ordinary non-event week typically sits between roughly 0.85 and 1.15. A reading outside that band is unusual for NIFTY specifically. The same reading could be entirely normal for SENSEX or BANKNIFTY.

Is high PCR bullish or bearish?

It depends on where price already is, not on the number alone. High PCR after a sharp decline often reads contrarian-bullish — crowded put positioning due for an unwind. High PCR during a slow grind higher more often reads as genuine bearish hedge demand. Treat PCR as a modifier on price context. Never use it as a standalone directional call.

What's the difference between PCR by OI and PCR by volume?

PCR by open interest reflects positions still held on the books. PCR by traded volume reflects today's trading flow, including trades opened and closed within the same session that leave no OI behind. Most broker dashboard widgets default to the OI version. Check which one you are reading before comparing it to anyone else's number.

Does PCR work on expiry day?

Less reliably than on a non-expiry session, and specifically in the afternoon. As both call and put OI unwind unevenly through expiry day, the ratio can shift meaningfully with no corresponding shift in market direction. Weight an expiry-afternoon PCR reading lower than a morning one.

Where can I see live NIFTY PCR?

Upstox's F&O discovery page, Sensibull's live options chart, Groww's options tools, and Angel One's knowledge center all publish a live NIFTY PCR figure. Each typically shows the OI-based version by default. Cross-check which definition is displayed, since the same label can mean different things across platforms.

Is PCR different for SENSEX and BANKNIFTY?

Yes. Lot size, strike spacing and the depth of open interest differ enough between NIFTY, SENSEX and BANKNIFTY that the same PCR number does not carry the same meaning across indices. Build a separate normal range for each index you trade rather than reusing one threshold everywhere.

How often should I recheck a PCR-based rule?

Periodically and deliberately, not once and forgotten. Our own experience with a silently mis-anchored gate is the case study. A threshold that was correct when it was set can quietly stop firing after an index rebase or a volatility regime change. Review the fire-rate of any PCR-based rule every few months, not just its win rate.

Should I combine PCR with implied volatility?

Yes — on its own, PCR tells you about positioning, not about expected movement. Reading a tail-end PCR print alongside elevated ATM IV gives a materially stronger signal than either number checked in isolation, particularly on expiry day when IV crush can mask what positioning is actually doing.

None of the above replaces position sizing, stop discipline, or testing a rule on your own data before it touches live capital. Paper trade first. EliteAlgo publishes this kind of research because option sellers deserve to know what a dashboard number is actually measuring before they build a rule around it. A single ratio cannot carry a trading decision on its own. If you want to see how we test ideas like this before they reach a live configuration, our algo trading strategies page walks through the broader framework. Our NIFTY option-selling primer for beginners is the right starting point if strike-band OI is a new concept. Traders running both indices should also see how we handle NIFTY and BANKNIFTY algo trading differently. Anyone who wants to build and test their own version of the distribution method above can read about the engine we use in our note on options backtesting software built for the Indian market.

Source data referenced in this piece: open interest and traded-volume definitions per the NSE's own F&O bhavcopy documentation, and US options market structure data from the CFTC. Regulatory context on exchange-traded derivatives disclosure in India comes from SEBI's public investor-education material — cited here only for how Indian derivatives markets are structured and disclosed, not as any form of advisory or registration claim. EliteAlgo is a proprietary research desk. Nothing above is investment advice, and no return outcome is implied or guaranteed.