Nifty overnight vs intraday returns: 30-year study insights

Why traders are talking about overnight returns

India’s trading community is debating a claim that sounds backwards at first glance. Multiple Reddit threads and finance feeds argue that the Nifty’s long-run wealth creation happened mostly when the cash market was closed. These posts point to charts where the open-to-close move contributes little over decades. Several summaries go further and say intraday returns were near-zero to slightly negative in the long run. In the same material, overnight returns are described as having a steady positive drift. The framing is blunt in some posts: the window when many participants “cannot trade” did the heavy lifting. This has triggered fresh arguments about whether frequent trading is structurally disadvantaged. It has also created confusion because “returns” can mean different windows inside the same day.

The two return windows: close-to-open vs open-to-close

The discussion uses two simple definitions that are repeated across posts. Overnight return is defined as yesterday’s close to today’s open. Intraday return is defined as today’s open (9:15 AM) to today’s close (3:30 PM). In the overnight window, the cash market is shut and many participants cannot trade the index directly. Posts argue that news, global cues, and movements in futures get reflected in the opening gap. Intraday is the window most traders watch in real time, including the candle-by-candle moves discussed on TV. The controversial point is not that overnight gaps exist, but that they appear to dominate long-term compounding in these charts. Some threads describe intraday as noisy and mean-reverting, while describing overnight as having momentum and drift. The debate is less about any single day, and more about what happens when you compound thousands of sessions.

What the widely shared Capitalmind summary says (2000-2023)

A specific dataset that is quoted widely is attributed to an analysis by Capitalmind Mutual Fund covering January 2000 to July 2023. In that 23.5-year period, the Nifty is stated to have risen from 1,592 to 25,057. The same summary claims the bulk of the gain, presented as 39,084 points, came from overnight changes. It also claims regular market hours (open to close) saw a loss of 15,620 points. As presented on social media, this implies an intraday-only approach would have eroded capital, with a cited loss of 84% for an intraday-only strategy. The same summary breaks return components into annual figures and states overnight returns delivered a median annual return of 5.7%. It says intraday activity contributed 2.4% annually and dividends added 1.4% annually. These numbers are being used online to argue that “activity” during market hours drags on total returns.

Another viral chart: 1995 onward, price index only

Alongside the 2000-2023 dataset, a separate viral post claims a longer history “since 1995.” That chart reports three CAGRs: buy-and-hold Nifty at ~10.95%, only overnight returns at ~20.02%, and only intraday returns at ~-7.56%. The same post claims ₹100 invested only in overnight moves would have grown to ₹26,842, while ₹100 exposed only to intraday moves would have fallen to ₹9. A key caveat is also highlighted in those threads: the analysis is based on the Nifty price index, not the total return index (TRI). As a result, the figures reflect price movement and exclude dividends. This matters because dividends are explicitly referenced as a return component in the Capitalmind-attributed summary. Traders are sharing the longer-history chart to support a bigger argument that the “market moves when you are not watching.” At the same time, the mixture of periods (1995 onward, 1999 onward, 2000 onward) is part of why the discussion is messy.

Consistency claims by decade and why they matter

Some posts focus less on cumulative points and more on how often each window is positive. One breakdown claims overnight sessions were positive about 59% of the time in the 2000s, versus 54% for intraday. It claims the gap widened in the 2010s, with overnight positive about 65% versus 46% for intraday. For the current decade, it says overnight stayed positive around two-thirds of the time, while intraday remained below half. Another widely shared explanation says average overnight returns rose from just under two basis points per day in the early 2000s to almost 12 in the 2020s. In the same telling, intraday returns fluctuated around zero or dipped into negative territory, creating a drag. Posts also describe the distribution difference: overnight gains are “compact” and cluster around a small positive mean, while intraday is more dispersed with negative skew. The key point being made is about persistence across cycles like the 2008 crash and COVID-19, which are explicitly referenced in social summaries. However, these consistency claims are still being debated because they depend on the exact sample, index series, and methodology used.

Why buy-and-hold can lag overnight-only in these charts

A recurring line in these threads is that “overnight-only can exceed buy-and-hold.” The logic presented is mechanical: buy-and-hold captures both overnight and intraday, so a negative intraday component can pull down the combined result. That is why some posts claim overnight returns exceed buy-and-hold returns in their charts. One widely shared example says that if you started with ₹100 in the Nifty on January 1, 1999 and captured only intraday returns, you ended with around ₹62. The same cluster of posts contrasts that with an overnight-only figure around ₹709 and a buy-and-hold figure around ₹440, presented as a striking gap. The takeaway in those posts is not just that overnight is positive, but that intraday “drags” total returns. This becomes a broader critique of intraday-style behavior, where frequent decisions are made in the noisiest window. Some threads even argue that to simply not lose money intraday, you must be right more than half the time, while to beat overnight you need to be right 53 out of every 100 sessions. Those hit-rate statements are also part of the viral narrative and are being repeated as rules of thumb rather than as audited claims.

Practical limits: what you can and cannot trade overnight

The viral framing leans on a real constraint: the cash market is shut overnight. Many retail participants cannot trade the index directly in that window, and the opening gap arrives before they can respond. Posts argue that futures markets keep moving, news breaks, and global markets shift, so the open price becomes the adjustment point. This is often used to suggest that the most valuable move is also the hardest to capture through short-term trading. At the same time, the posts are comparing stylised windows rather than a fully implementable strategy for every participant. For example, “buy at open and sell at close every day” is a clean way to isolate intraday returns, but it is also an extreme behavior. The social summaries also do not position transaction costs, slippage, or taxes as part of the core charts. Because the debate is being driven by charts and simplified examples, practical frictions are mostly discussed as secondary. The result is a conversation that is more about market structure and return decomposition than about a ready-made trading system.

What to take away if you trade or invest

The strongest common point across posts is that separating returns into overnight and intraday can change how people think about “when” markets reward exposure. The viral charts claim that long-run compounding is dominated by the close-to-open window, while open-to-close adds little or subtracts. The Capitalmind-attributed summary goes further by giving point contributions and a median annual return split that many users are quoting. Another set of posts extends the argument to “30 years” and reports extreme CAGRs, while also acknowledging it is price index data and not TRI. If you are using these charts to reflect on your own approach, the first step is to note exactly which period and index series is being discussed in the post you are reading. The second step is to separate the descriptive claim (return decomposition) from the prescriptive leap (what you should trade). Social media discussion often blends them, especially when it argues that “less interaction” is always better. The debate is useful because it forces clarity about what buy-and-hold actually contains: both overnight gaps and intraday moves. It also explains why many traders feel busy during market hours but still struggle to match long-term index outcomes that include the overnight component.