Is rollover early in the month worth reading? Two expiries say not yet
On 30 October 2026, 60 stock futures were rolling behind their usual pace and 16 ahead. In the last two expiries, a stock's rollover position about 25 days before expiry said almost nothing about its position 4 days before expiry.
What the screen showed
On Friday 30 October, 24 days before the November futures expiry, 60 stock futures were behind their usual rollover pace and 16 were ahead. NIFTY stood at 7.8 % rolled against 11.6 % usual. The expiry is Monday 23 November, because Tuesday 24 November is an exchange holiday.
A list like that looks like a signal. So we checked whether an early reading has held up.
How we checked
We used the exchanges' day-end F&O files. Rollover is open interest in later months divided by open interest in all months still open. For the September and October expiries, we took each stock's rollover about 25 days before expiry and again 4 days before. Each reading was compared with the same stock at the same point of the previous cycle. Index futures were left out.
What we found
| Expiry (stocks) | Early, late | Rank match |
|---|---|---|
| 29 Sep (207) | 4 Sep, 25 Sep | 0.03 |
| 27 Oct (210) | 1 Oct, 23 Oct | −0.01 |
A rank match of 1 means the early order and the late order are the same. Near 0 means there is no link.
| Early position | Stayed there | All stocks |
|---|---|---|
| Sep, ahead 2 pts+ | 30 of 54 (56 %) | 58 % ahead |
| Sep, behind 2 pts+ | 22 of 52 (42 %) | 42 % behind |
| Oct, ahead 2 pts+ | 13 of 24 (54 %) | 40 % ahead |
| Oct, behind 2 pts+ | 41 of 67 (61 %) | 60 % behind |
Three of the four groups matched the base rate. October's small ahead group did better, 13 of 24, and two cycles cannot settle whether that means anything.
Single stocks show why. INFY was level with its usual pace on 1 October and ended 24.8 points behind. MANAPPURAM was also level, and ended 44.5 points ahead.
Why early readings are thin
Very little has rolled that early.
| Days to expiry | Sep cycle, median stock | Oct cycle, median stock |
|---|---|---|
| 25–26 | 4.4 % | 2.8 % |
| 8 | 9.7 % | 9.8 % |
| 6 | 17.9 % | 14.5 % |
| 5 | 38.3 % | 36.4 % |
| 4 | 62.6 % | 60.8 % |
| 1 | 82.4 % | 82.9 % |
The median stock moved from under 10 % to over 60 % in the last 8 days. A gap of a few points at 3 % rolled is a small base, not a finished decision. In the November cycle, the median stock stood at 2.3 % on 30 October.
What this cannot tell you
- This is two cycles, and our files start in July 2026.
- Corporate actions and new F&O entries are not removed.
- Here a stock is compared with its own previous cycle. The live page uses the last two cycles, so counts differ.
- Rollover pace describes where positions sit. It does not say where price goes next.
How to cite this note
Source: TBTflow (https://tbtflow.com/learn/early-rollover-study), data as of 30 October 2026. Data readings, not recommendations.
More on citing our data: press and data sources. How our data is built: the method page.
This note describes market activity. It is not investment advice and not a recommendation to buy or sell any security. TBTflow is not registered with SEBI.
Read every panel behind these numbers, chapter by chapter.
Trading by the Numbers walks through the quantitative reads behind a modern F&O dashboard — order-flow imbalance, footprints, VWAP bands, OI flow, gamma structure and the expiry endgame — in plain trader-to-trader language.