"GIFT Nifty indicates a gap-up start" is probably the most repeated line in Indian market news. I wanted to know how often it's right, so I checked 125 mornings. The short version: it calls the direction of the opening gap about 7 times in 8, gets the size within about 40 points on a typical day, and tells you nothing about what happens after 9:15.
This guide explains what GIFT Nifty is and when it trades, then walks through those numbers.
What GIFT Nifty is
GIFT Nifty is a futures contract on the Nifty 50 index. It is priced in US dollars and traded on NSE International Exchange (NSE IX) in GIFT City, Gujarat. It took over from SGX Nifty in 2023, when trading of offshore Nifty contracts moved from Singapore to India.
The reason people watch it is the timing. The Indian equity market trades from 9:15 AM to 3:30 PM. GIFT Nifty trades for about 19½ hours a day, in two sessions (in IST):
| Session | Hours |
|---|---|
| 1 | 6:30 AM to 3:55 PM |
| 2 | 4:35 PM to 2:45 AM the next day |
So when news breaks in the US or Asia overnight, GIFT Nifty prices it in hours before the Indian market can. By 7 AM it holds the market's best guess at where Nifty will start.
Two things worth knowing:
- It is a futures price, not the index. The price includes the cost of carrying the index to expiry, and it is quoted in dollars. So the level of GIFT Nifty sits away from the Nifty spot level, and the difference (the basis) moves around. That's why I compare moves and not levels.
- It is built for foreign and institutional participants, not for typical Indian retail traders, who trade Nifty derivatives on NSE. Its usefulness to most of us is as an indicator.
How I measured it
Every trading morning, my daily pre-market report notes the GIFT Nifty move. I went back through the data for 125 sessions between 24 March and 28 September 2026 and compared two things:
- The signal: how far GIFT Nifty had moved by 7 AM from where it stood when the Indian market closed the previous day.
- The actual gap: where Nifty opened at 9:15 compared with its previous close.
Then I asked three questions. Did it point the right way? How close was the size? And did it say anything about what happened after the open?
Did it point the right way?
Yes, mostly.
| Measure | Result |
|---|---|
| Sessions | 125 |
| Correlation between the 7 AM signal and the actual gap | 0.90 |
| Days where the signal and the gap had the same direction | 87% |
| Same direction, on days when the actual gap was bigger than 30 points | 92% |
On days with a big gap, GIFT Nifty was wrong on only about 1 day in 12. The days it misses direction are mostly days where the gap was small, and a small gap is a coin flip in either direction.
How close was the size?
Less precise. Taking the miss as the signal minus the actual gap:
| Measure | 7 AM signal |
|---|---|
| Average miss | 51 points |
| Middle (median) miss | 41 points |
| Within 25 points | 33% of days |
| Within 50 points | 57% of days |
| Within 100 points | 90% of days |
For comparison, the average gap itself was about 106 points. So knowing the 7 AM signal cut the uncertainty about the opening gap by roughly half. That is useful but far from exact. I use 7 AM rather than a later reading on purpose: the Nifty's opening price is set in the pre-open auction that ends a few minutes after 9, so a GIFT Nifty reading after that isn't a forecast any more.
Here are the last ten sessions in the sample, in points:
| Date | GIFT signal at 7 AM | Actual gap at open | Miss | Open to close |
|---|---|---|---|---|
| Tue 15 Sep | +104 | +178 | −74 | −458 |
| Wed 16 Sep | +12 | +83 | −71 | +16 |
| Thu 17 Sep | −31 | −22 | −9 | +75 |
| Fri 18 Sep | +30 | +64 | −34 | +12 |
| Mon 21 Sep | −28 | −16 | −11 | +84 |
| Tue 22 Sep | +85 | +40 | +45 | −125 |
| Wed 23 Sep | −42 | +23 | −66 | +95 |
| Thu 24 Sep | −172 | −225 | +54 | −159 |
| Fri 25 Sep | +34 | −28 | +62 | +106 |
| Mon 28 Sep | −38 | −76 | +38 | −285 |
The directions agree on eight of the ten days. Each miss is worked out from unrounded prices, so a row can differ by a point from the rounded columns. The last column is the part to look at next.
Did it say anything about the rest of the day?
No.
| Measure | Result |
|---|---|
| Correlation between the 7 AM signal and the open-to-close move | 0.03 |
| Days where the signal direction matched the open-to-close direction | 55% |
| Days with a gap up that then closed above the open | 33 of 63 (52%) |
| Days with a gap down that then closed below the open | 33 of 62 (53%) |
That is what you'd expect from a coin. GIFT Nifty tells you about the overnight part of the move, which has already happened by the time the market opens. Once the market is open, the gap is history, and the signal has no memory of what comes next. On 15 September, the signal and gap both pointed up, and Nifty then fell 458 points from open to close. On 25 September the signal was small and positive, the market gapped down slightly, and it then closed 106 points above the open. The gap and the day are separate events.
Crude and other overnight cues tell the same story; the full test is in does the pre-market read predict anything?
Outliers
The largest miss came on 26 August: the 7 AM signal was about +211 points and the market opened only about 7 points higher. It was the morning after a monthly expiry. At 3:30 PM on 25 August, GIFT Nifty was quoted almost exactly at the Nifty close. By 7 AM it sat about 210 points above it, and it stayed about 200 points above the index the next day too. That pattern fits a contract roll: the price series moved from the expiring contract to the next month's, which trades at a premium. So most of that morning's "signal" was the roll, not news.
Leaving that day out barely changes the results (correlation 0.91, same 87% direction match), but it's a useful habit: around expiry, check the gap between GIFT Nifty and the Nifty close before reading anything into the move.
What I take from it
- Use it for what it is: a good read on the direction and rough size of the opening gap.
- Don't use it for what it isn't: it says nothing reliable about how the day will go.
- Compare moves, not levels, because of the futures basis, and watch for contract rolls after expiry.
- Expect 30 to 60 points of noise in the size of the gap, and occasionally much more.
My daily pre-market reports show the number each morning. The checks in this post are why I describe it as "where the open is likely to start" and never as "where the market is going".
Sources
- Tata Mutual Fund: What is GIFT Nifty? Definition, benefits and timing
- Session data: my own compilation from the GIFT Nifty and Nifty 50 price history for 24 March to 28 September 2026, obtained through the Kite Connect API.
This guide describes a market indicator and my own data, for information only. It is not investment advice or a recommendation to trade. Past patterns say nothing certain about the future. I am not registered with SEBI as an investment adviser or research analyst.