Participant-Wise Open Interest: Reading FII, Pro and Client Positions Without Guessing Direction

Every trading evening NSE puts out a small CSV that splits the whole F&O market's open positions into four buckets: Client, DII, FII and Pro. Within a few hours, screenshots of it are all over Telegram with captions like "FIIs heavily short, crash coming." Most of those captions are wrong, and not because the data is bad. The file is accurate. It just doesn't answer the question people ask it.

This guide explains what the participant-wise open interest file actually contains, how to turn its sixteen-odd columns into a handful of net numbers you can track, and the two traps that catch almost everyone: reading a net-short FII options position as a bearish call, and reading one day's change as a trend. At the end we show how our team uses it at EliteAlgo, which is far more modest than the Telegram version. We treat it as context for how wide to keep strangle wings in a lopsided week. It is never an entry signal.

What the participant-wise OI file is and when NSE publishes it

Participant-wise open interest is a daily end-of-day snapshot of outstanding F&O contracts on NSE, broken down by who holds them. NSE publishes it in its derivatives reports section alongside a sister file, participant-wise trading volume. The OI file is the one that matters for positioning, because volume tells you who traded today while open interest tells you who is still holding at the close.

The file is a plain CSV with a date in the name (the long-running pattern is fao_participant_oi_DDMMYYYY.csv). It usually appears in the evening after the 15:30 IST close, once the clearing data for the day is settled. NSE has reorganised its reports pages more than once since 2020, so check the current "All Reports" page under Derivatives on the NSE website rather than trusting an old bookmark. If you automate the download, build in a retry loop: on busy days the file can land later than you expect.

Each row is a participant type and each column is a contract category, measured in number of contracts, not rupees and not units. The layout looks like this:

Column group	Columns	What it counts
Index futures	Future Index Long, Future Index Short	Open NIFTY, BANKNIFTY and other NSE index futures contracts held long or short
Stock futures	Future Stock Long, Future Stock Short	Open single-stock futures, all underlyings combined
Index options	Option Index Call Long, Call Short, Put Long, Put Short	Open index option contracts, all strikes and all expiries lumped together
Stock options	Option Stock Call Long, Call Short, Put Long, Put Short	Same split for single-stock options
Totals	Total Long Contracts, Total Short Contracts	Sum across every category for that participant

Two facts about the file shape everything that follows. First, the bottom TOTAL row always balances: total longs equal total shorts in every column, because every open contract has one buyer and one seller. Second, the index option columns don't tell you the strike, the expiry or even the underlying. A deep out-of-the-money NIFTY put sold for ₹2 and an at-the-money BANKNIFTY call sold for ₹400 count as one contract each. Keep that in mind every time someone quotes a big number from this file.

One more scope note. The NSE file covers NSE contracts only. SENSEX and BANKEX options trade on BSE and do not appear in it, so a SENSEX expiry-day seller gets, at best, an indirect read from NIFTY positioning.

The four participant types and what each really is

The labels look self-explanatory. They aren't quite.

Client is the biggest and most mixed bucket. It is everyone who trades through a broker and isn't one of the other three: retail traders, HNIs, family offices, corporates, and proprietary trading firms that clear through someone else's membership. When people say "Client = retail," they're only partly right. A large non-broker prop shop sits in this row too.

DII means domestic institutional investors: mutual funds, insurance companies, banks and pension money. In F&O they are mostly hedgers. Their derivative footprint is small next to the other three, and a mutual fund buying index puts against a large equity book is not making a market call. Expect this row to look dull most days.

FII is the old name that stuck. The category today is foreign portfolio investors, regulated under the SEBI FPI framework. It covers global funds, hedge funds and foreign banks' trading arms. Many of them hold large Indian equity portfolios in the cash market, which matters: a short index future held by an FPI is very often a hedge against shares it owns, not a naked bet on a fall.

Pro is proprietary trading by trading members themselves, that is, brokers and clearing members trading their own capital through their own membership. This includes the large market makers and arbitrage desks. A lot of option selling and a lot of arbitrage sits here, which is why the Pro row often shows big numbers on both sides that net down to something small.

A useful way to hold this in your head: the four rows are not four opinions. They are four very different businesses, each with its own reasons to be long or short that often have nothing to do with tomorrow's direction.

Turning longs and shorts into net positions, step by step

The raw file has too many numbers to read at a glance. Our methodology is to reduce each participant to five net figures and then look at those over time. Here is the exact sequence.

Step 1: Net index futures

Net index futures = Future Index Long minus Future Index Short. A positive number means the group is net long index futures. Some people prefer the long share, Long ÷ (Long + Short), because it is easier to compare across months when total OI grows or shrinks. A long share of 25% means three short contracts for every long one.

Step 2: Net index calls and net index puts

Net calls = Call Long minus Call Short. Net puts = Put Long minus Put Short. Keep these separate. Folding them into one "options net" number throws away the only directional hint the options columns carry.

Step 3: A rough directional tilt

Long calls and short puts both gain if the market rises. Long puts and short calls both gain if it falls. So a crude options tilt is (net calls) minus (net puts). Treat this as a very blunt instrument. It weights a ₹2 lottery-ticket put the same as a ₹300 at-the-money put, and it has no idea about expiry.

Step 4: Check the zero-sum

Add the four participants' net figures for any column. They should sum to zero, because the TOTAL row balances. If your spreadsheet shows anything else, you have a parsing bug, usually a header row read as data or a comma inside a number. We built this check into our own loader after a column shift once turned a clean file into nonsense.

Step 5: Track the change, but over several days

Store each day's net figures in a table and look at the 5-day and 20-day change, not just today versus yesterday. The next sections explain why.

Here is a worked example with illustrative numbers only, not a real day's file, to show the arithmetic:

Participant	Index fut long	Index fut short	Net index fut	Long share
Client	2,10,000	1,40,000	+70,000	60%
DII	20,000	60,000	-40,000	25%
FII	90,000	1,80,000	-90,000	33%
Pro	80,000	20,000	+60,000	80%
Total	4,00,000	4,00,000	0

Notice how the net column sums to zero: +70,000 − 40,000 − 90,000 + 60,000 = 0. That is the zero-sum check doing its job. It also shows something people forget. If FIIs are net short 90,000 index futures, someone else is net long exactly 90,000. Saying "FIIs are bearish" is the same statement, mathematically, as saying "Clients and Pros are bullish." Neither sentence tells you who is right.

And the same illustrative day's index options for the FII row:

FII index options	Long	Short	Net
Calls	3,00,000	4,20,000	-1,20,000
Puts	4,50,000	3,50,000	+1,00,000
Crude tilt (net calls − net puts)			-2,20,000

On a screenshot, that −2,20,000 would be captioned "FIIs massively bearish." The next section is about why it doesn't mean that.

Index options: why 'FII net short calls' is not a forecast

This is the trap that does the most damage, so it's worth slowing down.

A net short call position can come from several completely different trades. A fund that owns ₹5,000 crore of Indian large caps can sell out-of-the-money NIFTY calls against that book to earn premium, a covered-call overlay. It is short calls, and it is very much long the market overall. A desk running call credit spreads sells one strike and buys a higher one; both legs land in the file, and the net shows short calls even though the risk is capped and small. A volatility seller running short strangles is short calls and short puts at the same time, betting on a quiet market, not a direction. None of those three is a forecast that NIFTY will fall.

Now add what the file hides. It lumps together every strike, every expiry and every NSE index underlying. Net short 1,20,000 calls could be mostly far out-of-the-money weekly NIFTY calls worth a few rupees each, or it could be near-the-money monthly BANKNIFTY calls with real delta. The contract count is identical. The market exposure isn't remotely the same. Without strike-level data, which this file doesn't have, you can't convert the count into delta, and delta is what actually moves with price.

There is also the counterparty point from earlier. In our illustrative day, FIIs were net short 1,20,000 calls, which means the other three groups were net long 1,20,000 calls between them. If you read that as bullish for the other side, you've made the same mistake in reverse.

Where does the "FIIs are smart money" idea come from? Partly from the US, where the CFTC's Commitments of Traders report has split futures positions by trader category since the 1960s and has a long research literature behind it. According to the CFTC's own explanatory notes, those categories describe what kind of business a trader runs, not whether its positions are right. Even there, the categories that look most "informed" are often commercial hedgers whose positions run against price for structural reasons. Indian participant data is younger and cruder, and it mixes hedges, arbitrage and speculation inside one row.

So what can you say about index options? A few modest things hold up:

• A large, sustained rise in FII long puts, together with a fall in their long index futures over several sessions, suggests they are adding protection. That says something about their caution, not about where NIFTY will be on Thursday.
• Pro short options rising sharply across both calls and puts usually means premium sellers and market makers are absorbing demand. That is often a sign that implied volatility is being sold, which matters for how much premium is on the table.
• Client net long options at extreme levels tends to show up around event weeks when retail buys cheap out-of-the-money options. It describes who is paying premium, not where price is headed.

Using a multi-day trend as context for strike width

One day's change in this file says very little. A single session's swing can come from an index futures rollover near monthly expiry, an arbitrage desk unwinding, or a big fund hedging ahead of a holiday. All three show up as large one-day moves and none carry a view. That's why we only look at participant data as a rolling picture: the 5-day and 20-day change in each group's net index futures and the long share over the same windows.

Here is the honest version of how our team uses it. EliteAlgo's desk trades NIFTY and SENSEX expiry-day and weekly option selling through its own backtest engine. The entry, exit and stop rules come from backtests, not from this file. Participant data feeds in at one point only: as one of several context inputs when deciding how wide to keep the wings of a short strangle in a week that already looks lopsided.

In practice it works like this. When FII net index futures have moved steadily in one direction for 8 to 10 sessions, and their long share has gone to an extreme for the quarter, we treat the week as one where a sharp move is more plausible than usual. In our experience, those stretches tend to come with more one-sided intraday moves. So instead of trading tighter, we keep the strangle a little wider on the side the crowd is leaning against, or trade fewer lots. We don't flip to a directional trade, and we don't skip a week because of this number alone. The point is risk sizing, not prediction.

Compare that to the common way the data gets used:

	Common use	How we use it
Time frame	Today vs yesterday	5-day and 20-day change
What it decides	Buy or sell tomorrow	Wing width and size, inside rules set by backtests
Options columns	"FII net short calls = bearish"	Read as hedging and premium-selling context only
Weight	Primary signal	One context input among several, such as India VIX and the option chain

India VIX does a lot of the heavy lifting here. If you want the premium side of this, our piece on using India VIX to budget short strangle premium covers how much premium a given VIX level actually pays for. Participant data adds a second lens: who is leaning which way, and how hard.

Paper trade this first. Any rule you build from participant data, whether it's a long-share threshold, a number of sessions, or a wing adjustment, should run on paper for at least 2 to 3 months of live sessions before real capital touches it. This data is easy to overfit, because it is published daily and looks like it should mean something every day.

If you are building this into an automated process, keep the participant file as a separate input that changes a parameter, such as wing distance or lot count, rather than a trigger that fires orders. That separation makes it much easier to test whether the input is adding anything. On our own NIFTY and SENSEX algo trading setup, every context input has to earn its place in a backtest before it touches sizing.

How it relates to FII/DII cash flows and the option chain

Two other daily data sets get confused with participant-wise OI, and it helps to keep them apart.

The FII/DII cash market figures, the "provisional" net buy/sell numbers that come out every evening, measure money flowing into or out of shares, in crore. Participant-wise OI measures open derivative positions, in contracts. An FPI can sell ₹2,000 crore of shares in the cash market and buy index futures on the same day, for example to stay invested through a portfolio rebalance. The two numbers can point in opposite directions without either being wrong. We go into how those cash flows behave intraday in our guide to FPI and DII flows and what they mean for NIFTY intraday, which is the natural companion to this page.

The option chain is the opposite of the participant file in one way that matters: it shows open interest strike by strike and expiry by expiry, but it doesn't tell you who holds it. The participant file tells you who, but not where. Used together they cover more ground than either alone. Our walkthrough of NSE option chain analysis for sellers covers the strike-level side, and if you want the myths about open interest in general cleared up first, start with NIFTY OI data without the folklore.

Where this data stops being useful

Every data set has a range where it helps and a point where it starts to mislead. For participant-wise OI, the limits are fairly clear.

It is end-of-day only. The file reflects positions at the close. On expiry day, when most of the action is between 09:15 and 15:30 IST and a large share of weekly positions expire that afternoon, yesterday's snapshot says almost nothing about the afternoon you're trading. For expiry-day sellers, intraday option chain changes and price action matter far more.

It counts contracts, not exposure. As covered above, a contract is a contract regardless of strike, premium or delta. Lot sizes for NSE index contracts have also been revised several times; a change in lot size changes how much exposure one contract represents, which breaks naive comparisons across those dates. Check the current lot size on the relevant NSE circular rather than assuming one.

It mixes underlyings and expiries. NIFTY, BANKNIFTY, FINNIFTY, MIDCPNIFTY and NIFTY NEXT 50 index options all land in the same columns. Per the SEBI circular of October 2024, each exchange now keeps only one weekly index expiry, so the mix has shifted toward NIFTY weeklies on NSE, but the file still doesn't separate them.

Rollover distorts the futures columns. In the last week of the monthly series, index futures OI moves sharply as positions roll from one month to the next. A big one-day drop in FII short futures around rollover is often just rolling, not covering. Compare net figures from just after the new series starts with the same point in earlier months instead.

Categories are broad. The Client row contains retail scalpers and multi-crore prop firms. The FII row contains passive index funds hedging flows and aggressive macro hedge funds. Averaging opposite behaviours into one number hides the signal you hope is there.

It is widely watched. When everyone reads the same CSV at 19:00 and posts the same conclusion, whatever simple edge existed in it has had a lot of time to be traded away. That doesn't make the data useless. It means the plain "follow the FIIs" reading is the least likely version to work.

If you want to see what the data looks like over longer stretches, aggregators such as Trendlyne's monthly participant OI view lay it out by month, and StockEdge's primer on open interest covers the basics if OI itself is new to you. The Metropolitan Stock Exchange also publishes a participant-wise OI file for its own derivatives in the same shape, which is handy for seeing the format but covers a tiny market. For the definitions of options terms themselves, the Options Industry Council has neutral, exchange-agnostic explanations.

A short routine for reading the file each evening

Put together, this is a routine that takes about 10 minutes once the loader is built:

1. Download the day's participant-wise OI CSV after it is posted, and run the zero-sum check on every column.
2. Compute, for each group, net index futures, long share of index futures, net index calls and net index puts. Append them to a running table.
3. Look at the 5-day and 20-day change for FII and Pro net index futures and long share. Ignore one-day moves unless they persist.
4. Flag the week as "lopsided" only if the multi-day move is large and the long share is near the edge of its range for the past 3 months.
5. If lopsided, check India VIX and the option chain before changing anything. Then adjust size or wing width inside your tested rules, never the entry logic.
6. Write down what you changed and why. After 20 to 30 flagged weeks you'll have a small dataset of your own to judge whether the adjustment helped.

That last step matters more than it sounds. Our data from flagged weeks is what tells us whether this input is worth keeping, and it is the only honest way to answer that question for your own trading style. EliteAlgo treats this as analytical work for its own proprietary book. It is research and software, not investment advice.

Frequently Asked Questions