On expiry Tuesdays, my bot's paper trades lasted 2.7 minutes instead of 4, and their results were 1.6 times as spread out as on an ordinary Wednesday to Friday. That's the mechanics of expiry showing up in real data.
This post explains how weekly and monthly expiry work in 2026, why option prices behave differently on the day, and what 24 expiry Tuesdays in my records look like.
How expiry works in 2026
Exchanges set fixed expiry days for derivatives. Since September 2025, the days are:
| Contract | Exchange | Expires |
|---|---|---|
| Nifty weekly options | NSE | Tuesday of each week |
| Nifty monthly, quarterly and half-yearly contracts | NSE | Last Tuesday of the expiry month |
| Stock futures and options | NSE | Last Tuesday of the expiry month |
| Sensex weekly options | BSE | Thursday of each week |
| Sensex monthly contracts and BSE stock derivatives | BSE | Last Thursday of the month |
Before September 2025, NSE's Nifty contracts expired on Thursdays, and the change moved the weekly action to Tuesday. If an expiry day is a market holiday, the exchange normally moves it to the previous trading day, so check the exchange calendar for the month you care about.
One definition helps: monthly expiry is the last expiry of the month. On that day the weekly contract, the monthly contract and every stock contract all expire together. My sample includes six: 28 April, 26 May, 30 June, 28 July, 25 August and 29 September.
Why prices behave differently
An option's price has two parts: what it would be worth if it expired now (intrinsic value), and a premium for the time left (time value). Time value shrinks as expiry approaches, and it shrinks fastest in the last hours.
On the day itself, an option that expires in a few hours has almost no time value left. Its price depends on where the index is relative to its strike. A move of a few dozen points can turn a cheap option worthless or multiply it several times within minutes.
That's the mechanic behind the pictures of options going "from ₹5 to ₹80" on expiry day. It works in both directions, and the losing side is quieter.
What my records show
My bot's paper trades on index options include 24 expiry Tuesdays, from 21 April to 29 September 2026. I compared them with the day before expiry (Monday) and with the rest of the week (Wednesday to Friday).
All of the figures are from paper trades on index options. The percentages are changes in the option's price, not the index.
| Expiry day (Tue) | Day before (Mon) | Wed to Fri | |
|---|---|---|---|
| Days in sample | 24 | 23 | 65 |
| Trade records per day | 91 | 71 | 85 |
| Median time in trade | 2.7 min | 5.0 min | 4.0 min |
| Average best price reached, in trade | +11.1% | +12.5% | +8.8% |
| Average worst price reached, in trade | −11.6% | −8.0% | −9.5% |
| Spread of results per trade (standard deviation) | 15.6% | 11.1% | 9.9% |
| Win rate | 49.1% | 48.0% | 44.4% |
| Payoff ratio (average win ÷ average loss) | 1.04 | 1.12 | 1.04 |
| Days that ended net positive | 10 of 24 | 11 of 23 | 27 of 65 |
What I read from it:
- Trades were shorter. The typical trade on expiry day was over in 2.7 minutes, compared with 4 to 5 on other days. Prices reach their stops and targets faster.
- The range was wider. The spread of results per trade was about 1.6 times that of an ordinary Wednesday to Friday (15.6% against 9.9%). Winners and losers were both larger than on Wednesday to Friday.
- The averages didn't stand out. The win rate on expiry days (49%) was close to the day before (48%) and a little above the other days (44%), and the payoff ratio matched Wednesday to Friday at 1.04. The share of days that ended net positive was similar too: 42% on expiry days, 48% the day before and 42% on Wednesday to Friday.
What this doesn't show
I'd be overstating it if I stopped there. These are the limits:
- It's 24 days. Records within a day are not independent. Many strategies fire on the same market move, so 2,186 expiry-day records behave more like 24 observations than 2,186. A difference of a few percentage points in win rate can easily be luck at this size.
- It's paper trading. Fills on expiry day, when spreads can be wide and prices move fast, are the place where paper is least like live. The paper vs live post covers that. My live sample on expiry days is too small to compare.
- Monthly expiry is only six days. On those six, results were a little more spread out than on the 18 weekly-only expiries (16.4% against 15.3%) and trades were longer (a median of 3.9 minutes against 2.4). Six days is far too few to call that a difference.
- It's my strategies, not the whole market. Other approaches will see expiry days differently.
The one finding I'd stand behind is the simplest: on expiry days, option prices move faster and further, and trades are shorter. That's what the mechanics predict, and the records agree.
What I'm leaving out
There's a lot of content on "how to trade expiry day." This isn't that. I don't have a view I'd publish on whether expiry days are better or worse to trade, and my data doesn't support one. If you're trading them yourself, the cost guide is worth reading first: short trades pay the same fixed costs as long ones.
Sources
- HDFC Sky: revised expiry days for NSE and BSE derivatives, effective September 2025
- Kotak Neo: SEBI reshuffles expiry days, NSE to Tuesday and BSE to Thursday from September 2025
- Trade data: my own trade log, paper records on index options from 21 April to 29 September 2026. Figures are my calculations.
This post explains exchange rules and describes my own trading records, for information only. It is not investment advice or a recommendation to trade on expiry days or any other day. I am not registered with SEBI as an investment adviser or research analyst.