Do volume spikes predict a stock's next move? We tested ~2,000 NSE stocks
Almost every scanner reads a volume surge as bullish. When we measured it across the whole NSE, the signal was strong — and it pointed the wrong way.
A big volume bar feels like conviction. When a stock trades two or three times its normal volume, most screeners light it up as a bullish signal — the crowd is here, something is happening, get in. It is one of the most widely trusted reads in retail technical analysis.
We were in a position to actually check it. Every evening our scanner logs which stocks fired which signals, along with the price at that moment, and later fills in what each stock did over the following days. That log lets us ask a plain question and answer it with numbers instead of intuition: after a volume surge, did those stocks beat the rest of the market, or lag it?
The answer was the single strongest result in the entire study — and it pointed the opposite way to the conventional wisdom.
What we tested
We defined a volume surge as a day where a stock trades at roughly twice its recent average volume or more (a relative volume of 2.0+) without matching price follow-through. For every day a stock fired that signal, we measured its return over the next three and ten trading days, then compared it against the return of the broad scanned universe on the same days. The number that matters is the excess — how the surge stocks did relative to everything else, which strips out whatever the market as a whole was doing.
| Signal | Horizon | Excess vs universe | t-stat | Days up |
|---|---|---|---|---|
| Volume surge | 3 days | −0.48% | −5.8 | 4 of 34 |
| Volume surge | 10 days | −0.78% | −4.3 | 3 of 27 |
How we measured this
Returns are aggregated to a daily mean first, then the statistics are computed over the series of days — not pooled across the ~1,900 stocks in a day, which would wildly overstate significance because those stocks all share the same market move. Overlapping forward windows are corrected with Newey–West (HAC) standard errors, and the significance bar is set from simulated random data rather than the textbook value, because the textbook value flags noise too often at these sample sizes.
Why a surge points down
A stock trading on double volume but not making real price progress is not a crowd rushing in — it is a crowd being sold into. Heavy volume without follow-through is the fingerprint of distribution: larger holders unloading stock into a burst of demand, right before the move gives back. The volume bar looks like excitement; over the next three days it behaved like exhaustion. Positive on just four of the thirty-four days we measured, this was not a marginal effect — it was the most consistent single signal we found, and it was negative.
Not all high volume is bad — only this kind
This finding is specifically about a spike without price follow-through. Sustained, elevated volume in a stock that is steadily trending is a different animal, and over longer horizons it behaves very differently. We pull that apart in a companion piece on the term structure of volume.
What we do with it
Because the evidence is this clear, StockLearn treats a volume surge as a warning label, not a confidence signal. It does not add to a stock's internal signal count, and it never nudges a verdict toward "bullish." A surge tells you to look more carefully at why the volume is there — not to buy the excitement.
The honest limits
This is measured over 34 trading days of component-level data, in one market period. The three-day result is robust; the ten-day version rests on fewer days and should be read as weaker. It is a measured tendency across thousands of stock-days, not a law of the market, and certainly not a trading instruction. What it does show is that a signal almost everyone reads as bullish has, in our data, leaned the other way — which is exactly the kind of thing worth measuring rather than assuming.
Key takeaways
- A volume surge without price follow-through underperformed the market by ~0.48% over three days.
- The result was strong and consistent (t ≈ −5.8), positive on only 4 of 34 days — the clearest single signal in the study.
- A spike on heavy volume with no progress reads as distribution (selling into demand), not accumulation.
- Sustained trend volume is different from a one-day spike — horizon matters.
- StockLearn treats a surge as a caution flag, not a buy signal.
See the scanner these numbers come from
StockLearn runs this read on ~2,000 NSE stocks every evening. Nifty 50 is free, no login.
Browse today's scan →This article reports measured technical results from StockLearn's own scan history over a specific, limited period. It is educational research, not investment advice or a recommendation to buy or sell any security. Past statistical tendencies do not predict future returns.