How often NIFTY has held at a strike

Sample

Index

NIFTY 50

Sessions

75

Period

2026-05-13 → 2026-09-22

Approach events

657

Spot: 1-minute NIFTY spot from Zerodha Kite. Strike grid: 50 points.

Hold and pin rates by horizon

Roughly four approaches in five did not push more than 15 points through the strike within a quarter of an hour; about two in three did not within half an hour. Fewer than half were still sitting at the strike after 15 minutes, which is the difference between “the level was not run through” and “spot stayed pinned”.

Hold rate at 15 minutes by days to expiry

Pooling the three expiry-week buckets: 83.9% with 0–4 days to expiry (n=349) against 74.7% at 5+ days to expiry (n=308). That gap of about nine percentage points is the one contrast in this sample that is wider than its sampling error. It is consistent with the mechanics of dealer hedging, which are strongest when gamma is concentrated in a near expiry, but consistency with a mechanism is not proof of it.

Chain-max open-interest strike vs other strikes

The max-OI strike pinned and held a little more often than the rest, but with 110 events on one side the 95% interval on the pin rate is about ±9.3 points wide, so the difference is within sampling error. We report it because it is what the sample shows, not because it is established.

Methodology

1. Approach. Spot moves to within ±15 points of a 50-point strike after having been more than 25 points away from it.
2. Hold. No 1-minute close more than 15 points beyond the strike (on the far side) within the horizon.
3. Pin. Spot is still within ±15 points of the strike at the end of the horizon.
4. Timing. Each event is measured strictly after its observation window; no information from inside the horizon is used to define the event. An event is stamped at the first 1-minute close inside the ±15 band; the horizon starts at the next bar.
5. Expiry bucket. Calendar days from the session date to the nearest listed NIFTY expiry, so weekly and monthly expiries are both included as 0-day events on their day.
6. Open-interest rank. Read from the option chain snapshot nearest to the approach; “chain-max” means the single strike with the largest call-plus-put open interest across listed expiries.
7. Confidence intervals. Normal approximation, 1.96 × √(p(1−p)/n), shown in percentage points. Events overlap in time, so the true intervals are somewhat wider than shown.

Caveats

• Differences between strike categories (chain-max open-interest strike versus other strikes) are within sampling error at these sample sizes; the expiry-week versus 5+ day difference is the only contrast large enough to stand out.
• One sample, one index, one summer. Rates measured in a different volatility regime may differ.
• Approach events overlap in time (the same strike can be approached repeatedly in a session), so the effective sample is smaller than the raw count of events.
• This is a description of how often spot stayed near a strike in our sample. It is not a performance record, a track record, or a claim about any strategy.