The Expiry Scorecard — NIFTY

The question we are keeping score on

There is a widely repeated idea that the market drifts to the strike carrying the most open interest, so the most crowded strike tends to expire worthless. It goes by names like max pain or pinning. It is also widely dismissed as folklore. We are not going to settle that argument with an opinion, so we are keeping a public tally instead.

Each expiry we take the 5 strikes that gained the most open interest during the session, and check what they were worth at the close. Then — and this is the part that matters — we compare them against strikes that were equally far out of the money but not crowded. Without that comparison the number means nothing, because most out-of-the-money options expire worthless anyway.

Every expiry, one by one

The 5 strikes that gained the most open interest each session, and what each was worth at the close. “Added” is the change in open interest across the session.

6 October 2026 — settled 22,776

Across the whole chain that day, 116 of 204 contracts expired worth less than ₹1 (57%). The matched control group: 2 of 5.

22 September 2026 — settled 23,329

Across the whole chain that day, 96 of 170 contracts expired worth less than ₹1 (56%). The matched control group: 3 of 5.

15 September 2026 — settled 23,119

Across the whole chain that day, 92 of 157 contracts expired worth less than ₹1 (59%). The matched control group: 4 of 5.

8 September 2026 — settled 23,635

Across the whole chain that day, 88 of 155 contracts expired worth less than ₹1 (57%). The matched control group: 3 of 5.

1 September 2026 — settled 24,056

Across the whole chain that day, 105 of 173 contracts expired worth less than ₹1 (61%). The matched control group: 3 of 5.

18 August 2026 — settled 24,155

Across the whole chain that day, 105 of 172 contracts expired worth less than ₹1 (61%). The matched control group: 3 of 5.

11 August 2026 — settled 24,467

Across the whole chain that day, 109 of 180 contracts expired worth less than ₹1 (61%). The matched control group: 1 of 5.

28 July 2026 — settled 23,988

Across the whole chain that day, 105 of 203 contracts expired worth less than ₹1 (52%). The matched control group: 2 of 5.

21 July 2026 — settled 24,187

Across the whole chain that day, 93 of 163 contracts expired worth less than ₹1 (57%). The matched control group: 2 of 5.

14 July 2026 — settled 24,053

Across the whole chain that day, 99 of 196 contracts expired worth less than ₹1 (51%). The matched control group: 2 of 5.

7 July 2026 — settled 24,398

Across the whole chain that day, 100 of 174 contracts expired worth less than ₹1 (57%). The matched control group: 3 of 5.

How this is worked out

Where the settlement level comes from

Put-call parity on the option premiums themselves — S = K + C − P — not from a spot feed. The check is that several strikes independently imply the same level. Only strikes within 150 points of the money are used: at expiry, far out-of-the-money premiums round to five paise and the calculation falls apart. On one expiry the unbanded figure spread across 182 points while the banded one agreed to within 2.5.

What counts as worthless

A closing premium under ₹1. On NIFTY that is effectively zero.

The control group

For each crowded strike we take another strike of the same type (call against call, put against put) that was equally far out of the money and was not crowded. Matching on distance alone would pair an out-of-the-money call with the in-the-money put at the same distance, which would rig the comparison in our favour. We made that mistake first and corrected it.

What gets excluded

Any expiry where the strikes disagree about the settlement level by more than 12 points. If they disagree, the number is not a price and we will not quote it. Excluded so far: 4 August 2026 (89 points), 25 August 2026 (31 points), 29 September 2026 (30 points).

What this is not

It is not a trading signal, a recommendation, or a prediction. It is a record of what already happened. We do not know which way any expiry will go, and a page built on 11 observations is far too small to tell you either.

We also cannot see who traded — only what traded. A crowded strike can be crowded because people are buying it or because people are writing it, and those are opposite bets. Where our own data cannot separate the two, we say so rather than guess.