Volatility Clusters – 18 Years of NSE Data Proves It

We mapped every trading day since 2008. What we found challenges a core assumption most investors make.

Most investors treat market volatility like lightning: unpredictable, random, impossible to anticipate. You’re either unlucky enough to be holding when it strikes, or you’re not.

The data tells a very different story.

We took 18 years of Nifty 50 and India VIX values: 4,458 trading days from March 2008 to March 2026 and mapped every single one. The result is the chart below. Take a moment to look at it before reading further.

Each cell is one trading day. The colour represents the India VIX reading that day, which is the NSE’s fear gauge, measuring expected volatility in the market. Pale cells are calm days. Deep red cells are days of high fear.

If volatility were random, you’d expect the red to be scattered – a speckle pattern distributed evenly across the calendar.

That is not what you see.

Fear Travels in Waves

The red arrives in dense blobs and then disappears for months, sometimes years. The calm periods are equally persistent: long stretches of pale colour where almost nothing happened.

This is not a visual trick or a consequence of how we’ve scaled the chart. It is one of the most robust and well-documented phenomena in financial markets, with a name: volatility clustering.

The observation dates back to Benoit Mandelbrot in 1963: large changes tend to be followed by large changes, and small changes by small changes. It was later formalised into the GARCH family of models by Robert Engle, for which he won the Nobel Prize in Economics in 2003. The phenomenon holds across virtually every liquid market ever studied.

The Indian market is no exception.

Reading the Chart: The Major Clusters

2008–2009: The Global Financial Crisis: India VIX peaked at 85.13 on November 17, 2008 — a level that has never been approached since. The cluster lasted the better part of 14 months, with only brief windows of calm. The 2009 general election result (May 18, 2009) produced the single largest one-day Nifty move in our dataset: +17.7%. Even that came within a period of elevated volatility.

2015–2016: The China Selloff and Demonetisation: Two distinct but partially overlapping clusters. The August 2015 China currency devaluation sent shockwaves through emerging markets. Just as things settled, the surprise demonetisation announcement of November 2016 reignited volatility. Notice on the chart how these show up as two closely spaced red patches — a pattern that superficially looks like one long event but is actually two separate shocks with a brief calm window between them.

March 2020: COVID: The fastest and most violent cluster in the dataset outside of 2008. India VIX went from under 15 to over 70 in a matter of weeks. What is equally striking on the chart is how quickly the calm returned — by late 2020 the palette had reverted to pale green, even as the pandemic itself continued. Markets had repriced and moved on.

2022: The Rate Hike Cycle: Less dramatic in peak VIX terms than the previous crises, but notable for its duration. The combination of global rate hikes, the Russia-Ukraine war, and FII outflows produced a sustained period of elevated volatility lasting most of the year. This is a good example of a slow burn cluster — not a spike but a prolonged regime of moderate fear.


What This Means for How You Invest

The persistence of volatility has a direct and practical implication that most investors overlook.

When you buy a drawdown on day three of a VIX spike, you are making a very different bet than buying on day sixty. In the early days of a cluster, the data suggests you are more likely to be buying into continued turbulence than into a turning point. The market has not yet found its footing. Mean reversion is real, but it operates on a longer clock than most people’s patience.

Conversely, deep into a prolonged calm when VIX has been below 15 for months and the heatmap is uniformly pale – that is precisely when complacency is most dangerous. Calm periods end. The chart shows they always have.

This is not a call to time the market. It is a call to be honest about what regime you are operating in, and to size your risk accordingly.

At East Green, our entry and exit rules are explicitly designed around this reality. We do not treat all market environments as equivalent. A signal generated in a low-volatility regime carries different weight than the same signal generated mid-cluster. The data justifies that distinction.


Explore the Data Yourself

The chart above is fully interactive. Hover over any cell to see the exact India VIX reading, Nifty closing price, and daily return for that day. Toggle between VIX-based colouring and absolute daily return to see the same clustering effect from two different angles.

The data is sourced from Investing.com and covers every trading day from March 2008 to March 2026.

If a single picture is worth a thousand words, eighteen years of daily data laid out in a grid is worth considerably more. The pattern speaks for itself.

Share this article: