Why option-chain open interest is mostly noise (and what isn't)

Open a Nifty option chain during market hours and scroll through any retail trading community on Telegram or Twitter. You will see the same screenshots over and over — strikes highlighted in yellow, big red arrows next to the PCR number, and confident captions about "max OI", "support", and "resistance". The implication is always that the chain is telling a story. Often it isn't. Often it is telling many stories simultaneously, most of them contradictory, and most of them noise.

This note is the internal framework we use for reading the chain — the three columns we actually pay attention to, the five patterns we deliberately ignore, and the reasoning for each. It is educational. Nothing below is a recommendation or a trading call.

The OI addiction

Open interest is seductive for a specific psychological reason: it looks like insider information. The argument sounds sensible — OI represents "smart money" positioning, therefore if the 22,500 PE has the highest OI, 22,500 must be "support". This is a story with almost no statistical support, and it survives because it is satisfying, not because it is true.

Three things are wrong with the naive reading:

1. OI aggregates buyers and sellers. A high PE OI can come from writers (bullish) or buyers (bearish). The chain does not tell you which.
2. OI includes hedging positions from institutional books that are indifferent to direction.
3. Static OI is a snapshot. The same 1 crore OI at 22,500 PE means different things if it built over a week at lower prices versus one session at the current price.

Before you can read OI well, you have to stop reading it as a verdict and start reading it as a question.

What OI actually tells you

OI is the count of outstanding contracts at a strike — every long matched with every short. So the useful information is not the absolute level but the change in OI, joined with the change in price and volume. These three together are a position-building signal. In isolation they are each half a story.

Here is the 2×2 we teach internally:

                  PRICE UP          PRICE DOWN
OI UP         Long build-up      Short build-up
OI DOWN       Short covering     Long unwinding

This matrix is roughly 40 years old. It still works because it describes a tautology about participant behaviour — if contracts are being added while price rises, someone is aggressively opening longs; if contracts are being closed while price falls, longs are exiting. None of this is predictive on its own. All of it is descriptive of what is happening right now.

What OI does NOT tell you

Three things, all of which are frequently asserted and all of which are structurally unknowable from the chain:

• Direction of the next move. The same OI build-up can resolve up, down, or sideways. The matrix describes who is in the boat, not where the boat is going.
• Who is on which side. Retail vs institutional vs proprietary vs arbitrage — the chain does not label participants.
• Hedging intent. A large part of index-option OI is offsetting delta on cash-market positions or other derivatives. That OI is directionally neutral; it just looks big on the chain.

If you catch yourself saying "smart money is building shorts at 22,700", stop. You don't know who is selling and you don't know why.

The three columns we use

On our internal chain view, we hide every column except these three. They are genuinely informative when watched across the day.

1. Change in OI (not absolute OI)

Absolute OI accumulates from the start of the series. A high absolute number at one strike often just means "this is a round-figure strike people have been trading for two weeks". Change in OI since the previous close (or since the previous scan) isolates what is happening today.

Rule of thumb: we only look at strikes where change in OI is in the top five (positive or negative) on either side of the chain. Everything else is background.

2. Volume-to-OI ratio

If a strike has 10 lakh OI and 20 lakh volume intraday, the position is rotating fast — short-term traders, no conviction holding. If a strike has 10 lakh OI and 50,000 volume, the position is sticky — it was opened, it is sitting, it is expected to be held.

Sticky OI matters more than rotating OI, because it represents committed capital that has to be unwound before the strike loses its magnetic effect on price.

3. IV at the strike

Implied volatility at the strike tells you the price being paid for the position. Rising IV on rising OI is a different signal from rising OI on falling IV. The first is panic; the second is comfortable, confident position-building.

Five patterns to ignore

All of these are seen every day in retail commentary. None of them survive honest back-testing in our internal research, which is why we stopped using them.

1. "Maximum OI at strike X = strong support/resistance." Max OI is a description of past activity. It is correlated with round-figure psychology, not with future price pivots.
2. "PCR above 1.3 = bullish, below 0.7 = bearish." PCR in Indian indices is dominated by institutional put-writing for income, not by directional conviction. Retail PCR rules of thumb do not reliably translate.
3. "OI buildup at higher CE strikes = ceiling." Higher-strike CE writing is frequently the short leg of a covered call or a bull-call spread. It is not a ceiling; it is a hedge.
4. "Max pain = target for expiry." Max pain is a theoretical optimisation, not a gravitational law. The empirical tendency of spot to close near max pain on expiry is weak and inconsistent.
5. "Sudden OI spike = someone knows something." Block rollovers, delta-hedging adjustments, and institutional rebalancing all create OI spikes that have zero directional content.

A reading framework

When we look at the chain — which we do three times a day, not continuously — we run the same short sequence:

1. What are the top three strikes by change in OI on each side today?
2. Is the change in OI confirming the price action, or fighting it?
3. Is IV at those strikes expanding or compressing?
4. Is this the first time this level of activity has shown up this week, or is it a continuation?

The chain is a diagnostic tool, not a signal generator. It tells you what kind of market you are in, not what to do next.

Most of the time the answer to the sequence above is "nothing unusual, continue with your plan". Once every week or two, the chain will flag something genuinely interesting — usually a sharp disconfirmation between price and positioning. That is when it earns its place in the process.

This framework is not exotic. It is deliberately boring. The edge in reading the chain is not a secret pattern — it is the discipline of ignoring most of what is on screen. We will publish a follow-up study note on the specific options-chain heuristics we tested against 2023–2025 index data, with the ones that held up and the ones that didn't.