How much of an “uptrend edge” is just the market rising?
Stocks in an uptrend beat those that aren't — over a rising market. We measured how much of that is stock selection versus simply being long during a recovery.
"The trend is your friend" is probably the most repeated line in technical analysis, and it is easy to demonstrate: stocks in an uptrend do, on average, beat stocks that are not. Our own data shows it too. But there is a trap hiding inside that observation, and it is worth measuring rather than assuming.
When a stock's short-term average sits above its medium-term average — the 20-day above the 50-day, our "EMA-aligned" condition — it is, by definition, simply in an uptrend. The question is whether being in an uptrend is a genuine stock-selection edge, or just a way of being long the market while the market happens to be rising. Over a recovering market those are very hard to tell apart.
| Signal | Horizon | Excess vs universe | t-stat | Days up |
|---|---|---|---|---|
| EMA-aligned (in an uptrend) | 3 days | +0.18% | +1.5 | 62% |
| EMA-aligned (in an uptrend) | 10 days | +0.50% | +3.2 (thin) | 74% |
Positive — but is it skill or beta?
The spread is positive, and at ten days it looks strong on paper. Here is the catch: our measurement window is a post-crash recovery, a stretch where the market as a whole was climbing. During a rising market, "being in an uptrend" and "riding the market's rebound" are almost the same thing. Most of that +0.50% is very plausibly the momentum factor's return during one favourable regime — market beta wearing the costume of a stock-picking edge.
To separate the two, you need the signal to keep working when the market is not rising — in flat and falling regimes, where being "in an uptrend" no longer automatically means being long a rally. We do not yet have enough data across enough regimes to make that call, which is precisely why we will not claim this as an edge.
How we measured this
Daily excess of "in an uptrend" stocks over the scanned universe, over the day series with Newey–West standard errors. The "thin" flag fires when the number of days is too small for the significance test to be reliable at that horizon — and at ten days, with under thirty days of data, it is.
Why we say this out loud
It would be easy to publish "+0.50%, statistically significant" and let readers assume the scanner has found an edge. That would be dishonest. The professional position is that "EMA-aligned" is a measured candidate, not a proven signal — interesting, positive so far, and entirely capable of being nothing more than one regime's beta. We register it, keep measuring it, and let the data decide over time rather than declaring victory on a rising market.
The takeaway for your own analysis
When you see a stock that "went up in an up market," do not confuse that with "picked well." The honest question is whether it beat the market it rose with — and over a recovery, most uptrend signals do not clearly clear that bar.
Key takeaways
- Stocks in an uptrend showed a positive spread (+0.18% at 3 days, +0.50% at 10).
- But the window is a post-crash recovery, where "in an uptrend" and "long the rally" are nearly identical.
- Most of the spread is likely market beta, not stock-selection skill.
- Separating the two needs data from flat and falling regimes, which we do not yet have.
- We treat "in an uptrend" as a measured candidate, not a proven edge — and say so.
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.