Nifty 50 Returns by Samvat Year (Diwali to Diwali): How Often Did It Make Money, 2007–2025? Diwali marks the start of a new Samvat year in the Hindu calendar, and many Indian investors treat it as the start of a new…
Nifty 50 Returns by Samvat Year (Diwali to Diwali): How Often Did It Make Money, 2007–2025?
Diwali marks the start of a new Samvat year in the Hindu calendar, and many Indian investors treat it as the start of a new investing year too. Samvat 2083 began with Diwali in early November 2026. Before the next round of predictions, here is what the numbers say about Nifty returns by Samvat year.
I measured every complete Diwali-to-Diwali year from Samvat 2064 (Diwali 2007) to Samvat 2081 (Diwali 2024 to Diwali 2025), 18 years on the Nifty 50 and 25 on the Sensex since 2000. Data ends on 9 October 2026, so the just-finished Samvat 2082 is not counted.
Quick answer
- The Nifty 50 rose in 14 of 18 Samvat years (78%), with an average of +12.8% and a median of +9.7%.
- The best was Samvat 2065 (+104%); the worst was Samvat 2064 (−56%), the 2008 crash.
- A Diwali start was not special: the share of positive years (78%) matched what you get from any random one-year start (78%).
- Even good years had drops. In 13 of 18 Samvat years the Nifty fell 10% or more from its high at some point.
- Samvat years are a calendar tradition, not a market signal. Use them for a yearly review, not for timing.
In this article
What did the Nifty do in each Samvat year?
Here is every Samvat year, with the date range, the return, and the deepest fall inside the year:
| Samvat year | Diwali to Diwali | Return | Deepest fall inside the year |
|---|---|---|---|
| 2064 | 8 Nov 2007 to 27 Oct 2008 | −55.7% | -60% |
| 2065 | 27 Oct 2008 to 16 Oct 2009 | +103.7% | -19% |
| 2066 | 16 Oct 2009 to 4 Nov 2010 | +22.2% | -11% |
| 2067 | 4 Nov 2010 to 25 Oct 2011 | −17.4% | -25% |
| 2068 | 25 Oct 2011 to 12 Nov 2012 | +9.5% | -15% |
| 2069 | 12 Nov 2012 to 1 Nov 2013 | +11.0% | -15% |
| 2070 | 1 Nov 2013 to 22 Oct 2014 | +26.8% | -6% |
| 2071 | 22 Oct 2014 to 10 Nov 2015 | −2.7% | -16% |
| 2072 | 10 Nov 2015 to 28 Oct 2016 | +11.0% | -12% |
| 2073 | 28 Oct 2016 to 19 Oct 2017 | +17.5% | -8% |
| 2074 | 19 Oct 2017 to 7 Nov 2018 | +4.5% | -15% |
| 2075 | 7 Nov 2018 to 25 Oct 2019 | +9.3% | -11% |
| 2076 | 25 Oct 2019 to 13 Nov 2020 | +9.8% | -38% |
| 2077 | 13 Nov 2020 to 4 Nov 2021 | +40.9% | -7% |
| 2078 | 4 Nov 2021 to 24 Oct 2022 | −1.0% | -16% |
| 2079 | 24 Oct 2022 to 10 Nov 2023 | +9.6% | -10% |
| 2080 | 10 Nov 2023 to 1 Nov 2024 | +25.1% | -8% |
| 2081 | 1 Nov 2024 to 21 Oct 2025 | +6.4% | -11% |
Nifty 50 return from Diwali to the next Diwali, Samvat 2064 to 2081. Price index; dividends excluded. Source: Yahoo Finance ^NSEI daily closes.
The year of the 2008 crash dominates both ends: Samvat 2064 lost 56%, and Samvat 2065, which started at the low, gained 104%. Leave those two out and the other 16 years look more ordinary, with 4 of the 18 years in total ending in a loss.
Is a Diwali start better than any other start?
Now the important comparison. Is starting at Diwali better than starting on any other day? I compared three rules over the same period:
| Rule | Years or windows | Positive | Average return |
|---|---|---|---|
| Diwali to Diwali (Samvat years) | 18 | 14 of 18 (78%) | +12.8% |
| Calendar years (Jan to Dec) | 18 | 15 of 18 (83%) | +11.6% |
| Every possible 1-year start day | 4,393 days | 78% | +11.6% |
Average one-year Nifty 50 return by start rule, 2007 to 2025. Price index; dividends excluded.
Diwali-to-Diwali averaged +12.8%, calendar years +11.6%, and all possible one-year start days +11.6%. The differences are small and come from a handful of years, especially 2008 and 2009. With 18 data points, you cannot say Diwali beats any other starting date.
What does the longer Sensex record show?
The Nifty only goes back to 2007 on daily data, so I repeated the test on the Sensex, which has daily history from 1997:
| Index | Samvat years | Positive | Average | Median | Best | Worst |
|---|---|---|---|---|---|---|
| Sensex, from Diwali 2000 | 25 | 19 of 25 | +17.3% | +11.2% | +104% (Samvat 2065) | −55% (Samvat 2064) |
| Nifty 50, from Diwali 2007 | 18 | 14 of 18 | +12.8% | +9.7% | +104% (Samvat 2065) | −56% (Samvat 2064) |
The longer record tells the same story. The Sensex rose in 19 of 25 Samvat years, close to the 77% you would get from any one-year start. A Diwali start did not make the odds better.
What about the dips inside a year?
Because a Samvat year is a full year, it includes the dips as well:
Deepest fall from the highest close within each Nifty 50 Samvat year. Price index; dividends excluded.
In 13 of 18 years the index fell at least 10% from its high at some point. Samvat 2076 (Diwali 2019 to Diwali 2020) ended +10% but fell 38% along the way, because of the March 2020 crash. A year that ends well can still feel awful in the middle.
How should you use this?
- Use Diwali as a review date. It is a good habit to check your goals, SIP amounts and asset mix once a year.
- Don't wait for it to invest. About 78% of one-year windows were positive, whenever they started. A SIP removes the need to pick a day. See our ₹5,000 SIP backtest.
- Don't read much into Muhurat trading. Our test of the one-hour session found no reliable edge: Muhurat trading: 20 years of data.
- Expect dips. Even a profitable year can fall 10–20% from its peak. See Nifty 50 returns by calendar year for the odds over 1, 3, 5 and 10 years.
- Keep money you need soon out of equity, so a fall doesn't force you to sell.
Caveats
Things to keep in mind:
- Diwali (Lakshmi Puja) dates move every year. I used the last closing price on or before the date; if you use a different convention, results shift slightly.
- The Nifty has 18 complete Samvat years on daily data, which is a small sample; the Sensex has 25. Averages are pulled by 2008 and 2009.
- Price indices exclude dividends, which would add roughly 1% a year to every figure.
- This is history, not a forecast for Samvat 2083.
You can test your own entry dates and rules on Indian index data with Welth Lab, WelthWest's no-code backtesting tool.
Frequently asked questions
What is a Samvat year in the stock market?
A Samvat year is one Vikram Samvat year, which starts around Diwali. Indian markets hold a special one-hour Muhurat trading session on Diwali evening to mark the new Samvat.
How often does the Nifty rise in a Samvat year?
In 14 of 18 complete Samvat years from Diwali 2007 to Diwali 2025, or 78% of the time, with an average gain of +12.8% and a median of +9.7%.
What was the best Samvat year for the Nifty?
Samvat 2065 (October 2008 to October 2009), when the Nifty rose about 104% after the global financial crisis low.
Is Diwali a good time to invest?
Not in a way the data can prove. About 78% of one-year windows starting on any day were positive, nearly the same as Diwali-to-Diwali.
Does Muhurat trading predict the year?
No reliable pattern. See our test in Muhurat trading: 20 years of Nifty data.
What happened in Samvat 2082?
It is not included because it ended in November 2026, after this analysis's data cutoff of 9 October 2026 (kept more than 30 days before publication). Check a current data source for its result.
Related reading: Muhurat trading: 20 years of Nifty data · Nifty 50 returns by calendar year · Gold vs Nifty since 2010 · ₹5,000 SIP for 10, 15 and 20 years
Sources and method: Daily closing values of the Nifty 50 (Yahoo Finance ^NSEI, from 17 September 2007) and the BSE Sensex (^BSESN, from 1997) to 9 October 2026, script samvat.py. A Samvat year runs from the last close on or before one Diwali (Lakshmi Puja) date to the last close on or before the next. Diwali dates were taken from NSE and BSE Muhurat trading circulars and holiday lists and should be re-checked before publication. Only complete Samvat years are counted. Price indices, no dividends. See NSE India and BSE India.
Disclaimer: WelthWest is an analytics and education platform, not a SEBI-registered investment adviser, research analyst or broker. Past performance does not indicate future returns. Research and drafting were assisted by AI; the data, calculations and conclusions were checked by the author. This article is for educational purposes only and is not investment advice.