All strategies · Review page

Nifty 50 weekly expiries: the move from the open of the session before expiry to the close on expiry day

Based on the Nifty end-of-day data held on the VPS (NSE official daily OHLC; contract data from NSE's F&O files) · 405 weekly expiries, 14 Feb 2019 to 19 Oct 2026 · prepared 29 Oct 2026

1. Definitions used

• Weekly expiry E: each Nifty weekly (and monthly) expiry date from 14 Feb 2019 (first weekly expiry) to 19 Oct 2026, taken from the contracts in NSE's F&O files. Where an expiry was shifted for a holiday, the last session the contract actually traded is used. Expiry day was Thursday until August 2025 and mostly Tuesday since September 2025.
• E−1: the previous trading session (not the calendar day before). Open = NSE's official opening index value on E−1; Close = the index close on E (the settlement value of index options is the closing index value).
• Net move = Close(E) ÷ Open(E−1) − 1, in points and %. Breadth of the swing ("breath" in your request) = highest high minus lowest low over the two sessions, as % of the E−1 open; both are reported, because the net move is what settles and the swing is what you are exposed to on the way.
• OTM+k = k listed strikes (50 points each) beyond the at-the-money strike nearest to the E−1 open; the call is above, the put below. "Settled beyond" = the expiry close is past the strike; "touched" = the E−1 or E high/low reached the strike.

2. Answers to the five questions

1 and 2 and 3. Close on expiry, open on E−1, and the move between them

The complete table for all 405 expiries (open of E−1, close of E, points, %, swing) is in the box at the end of this section; the CSV is linked on the review page. Summary: net move mean -0.04%, median -0.05%, standard deviation 1.48%; absolute move mean 0.95%, median 0.65%, 90th percentile 2.10%, 95th 2.63%, 99th 4.87%, largest 11.72% (26 Mar 2020). The swing (high to low over the two sessions): median 1.48%, 90th percentile 2.75%, largest 14.25%.

4. How many times more than ±2% and ±5%

On the full swing rather than the net move, 99 of 405 expiries had a high-to-low range above 2% and 10 above 5%. For context, the same two-session measure on every Thursday from 2013 to Feb 2019 (before weekly options existed; a proxy, not expiries): 22 of 299 above ±2% (7.4%) and 1 above ±5%.

By year

3. Where your strikes sit against these moves (OTM+2, +3, +4)

OTM+2, +3 and +4 are only 100, 150 and 200 points from the open-based at-the-money strike, which is a much smaller distance than the ±2% (about 450 points) and ±5% moves above. So the exceedances above are not the relevant test for these strikes; the relevant test is how often the index finished or traded beyond 100, 150 and 200 points:

4. Why this matters for the sold options

A sold option keeps the premium if it expires worthless and loses the amount it finishes in the money. The breach rates above show that a 100 to 200-point cushion is exceeded often over a two-session hold, and the distribution of the net move has a long tail (the largest moves are shown in the chart above). The tail matters more than the average: a few expiries beyond ±2% produce losses of several hundred points per lot, which have to be paid for by many small premiums.

5. Indicative result of selling both strikes (before costs)

Read with care. This is not a back-test of the strategy. The true opening premium on E−1 is not in end-of-day data, so the premium is proxied by the exchange settlement price of each strike on the session before the opening (E−2), and the loss by the intrinsic value at expiry. No brokerage, STT, slippage, margin or stops are included, and the opening gap changes the actual premium collected.

Reading the table: premium is the cash collected; the payout is what is paid back at settlement on average; their difference is the indicative edge before costs, and the 5th-percentile and worst figures show the tail. Compare the size of the average edge with trading costs (several points per lot for four orders plus STT and slippage) before drawing conclusions. An earlier, more careful study of weekly option selling on this data (the OPTSELL study) found no variant reaching the 30% target with drawdown under 15%, and showed March 2020 turning plain short strangles into 30 to 45% drawdowns.

6. Questions and recommendations before relying on this

• Opening premium. The decisive number is the premium you can actually collect at the E−1 open. End-of-day data cannot show it. Our one-minute Nifty option data (December 2020 to October 2026) covers up to OTM+2 (ATM±2 strikes) and could be extended to ±10 strikes from Dhan; it would let us test the exact entry time and the real opening prices for OTM+2, +3 and +4.
• Costs and slippage. A four-order strangle pays brokerage on every leg plus STT, exchange fees and slippage (about 1% of premium over a round trip in our other studies). With strikes this far out the premium is small, so costs take a large share.
• Tail protection. The worst expiry cost roughly 620 to 660 points per lot, against average gains of about 30 to 36 points. Position size, a stop-loss, or buying a far wing change the picture; please tell us which of these you plan to use.
• Entry conditions. The distribution differs by volatility regime, expiry weekday (Thursday versus Tuesday since September 2025) and event days. A pre-registered filter test (VIX, previous-day move, results season) is a natural refinement.
• Sensex. If you plan the same on Sensex, the expiry day and strike spacing differ; Dhan has Sensex 1-minute option data from mid-2023 only.

7. Limits

• Daily data only: the open of E−1 and the close of E are exact, but the path between them is known only through the daily high and low. A strike can be touched intraday and recover.
• The ATM strike is rounded from the E−1 open; the real opening ATM depends on the first prices of the day.
• Premium proxy (E−2 settlement) is not the opening premium. This is the main weakness of section 5.
• Weekly options on Nifty only exist from 2019; 405 expiries is a limited sample for tail events (a few observations drive the ±5% counts). The 2020 crash and the 2024 election day are in the sample, but a larger event has not happened here.
• Expiry weekday changed in September 2025; the study treats all expiries alike.
• Source files: NSE official end-of-day index files (2013 onward), NSE F&O bhavcopy archives (contracts, strikes, settlement prices). Education and research only; not investment advice.

Research and education only. Hypothetical and back-tested results have inherent limitations and do not guarantee future results. Not investment advice; the publisher is not presented as a SEBI-registered investment adviser or research analyst.