When Intuition Fails: A Data-Backed Look at 200DMA Extremes
We tested a common belief: avoid stocks far above their 200DMA. The data showed the opposite—momentum persisted. More importantly, stocks 20%+ below their 200DMA delivered 2–4× higher monthly returns in extreme regimes. That non-linear insight reshaped our model.
A Simple Question That Deserved a Hard Test
During a portfolio review, we observed several small-cap stocks trading 30–40% above their 200-day moving average (200DMA).
From a traditional technical perspective, these appeared “extended.”
The intuitive question was straightforward:
Should stocks trading far above their 200DMA be avoided to reduce the risk of chasing momentum?
Rather than embedding this intuition directly into our system, we treated it as a testable hypothesis.
The Hypothesis
The prevailing assumption in technical analysis is well known:
Stocks that trade far above their 200DMA are overbought and should deliver weaker forward returns.
If this belief were correct, it should be clearly observable across time, thresholds, and market regimes.
Data & Scope
We evaluated this hypothesis using:
- Universe: NIFTY SmallCap 250
- Period: 2022–2025
- Observations: ~9,500 stock-months
- Rebalance Frequency: Monthly
- Return Horizon: Forward 30 trading days
Each observation was tagged with:
- Distance from 200DMA
- Forward return
- Market regime (risk-on, neutral, risk-off)
What the Data Actually Showed
We tested multiple exclusion thresholds—filtering out stocks above progressively higher 200DMA distances.
The result was consistent across thresholds:
- Stocks filtered out for being “too far above” the 200DMA outperformed those retained
- Return drag from these filters ranged from ~60 to 130 basis points per month, depending on threshold
- The effect was statistically significant at conventional confidence levels
In other words, any hard ceiling on 200DMA distance would have reduced returns.
Momentum was not a risk to be controlled—it was a source of performance.
The Non-Linear Structure We Did Not Expect
When we examined returns across the entire distribution of 200DMA distances, a clear U-shaped relationship emerged.
Approximate average 30-day returns by bucket:
| Distance from 200DMA | Avg 30D Return |
|---|---|
| Below -30% | ~10% |
| -30% to -20% | ~6% |
| -20% to -10% | ~2% |
| -10% to +10% | ~1–2% |
| +10% to +20% | ~3% |
| Above +30% | ~4%+ |
Two observations stood out:
- Stocks near the 200DMA delivered the weakest returns
- Returns peaked at the extremes, not the middle
This structure would not be captured by linear rules or binary filters.
Deep Value Was the Stronger Extreme
While both ends of the distribution performed well, the deep value side was dominant.
Stocks trading 20–30% below their 200DMA delivered 2–4× higher average monthly returns than stocks clustered near the moving average.
However, this effect was not universal.
Regime Dependency
We segmented results by market regime:
- Risk-On: Deep value stocks averaged double-digit monthly returns
- Risk-Off: Deep value remained strongly positive, acting defensively
- Neutral: Returns were statistically indistinguishable from zero
Deep value was therefore a conditional factor, effective primarily in extreme market states rather than sideways environments.
Why Momentum Needed No Adjustment
The U-shape showed that momentum (> +30% above 200DMA) also worked, with ~4%+ average monthly returns.
However, momentum stocks already score well in our system:
- Strong price trends
- Improving fundamentals
- High profitability and balance-sheet quality
These stocks naturally rise to the top of the rankings.
Deep value stocks do not.
The Structural Asymmetry
Price collapses tend to suppress backward-looking fundamentals:
- Earnings momentum weakens
- Margins compress
- Composite scores fall
Even when selling pressure has exhausted itself, these stocks remain penalized in purely fundamental scoring systems.
Without adjustment, mean-reversion opportunities were being systematically underweighted.
The Design Response
Rather than introducing a new factor or imposing hard rules, we implemented a measured correction:
- A moderate positive adjustment to conviction scores
- Applied only when prices were ≥20% below the 200DMA
- Activated only in risk-on and risk-off regimes
- No penalties applied to momentum stocks
The objective was calibration—not aggression.
Portfolio-Level Impact
At the full-system level, the adjustment resulted in:
- ~50–70 bps improvement in annualized returns
- Higher risk-adjusted performance (Sharpe improvement)
- No increase in maximum drawdown
- Slightly lower turnover, indicating improved selection stability
While portfolio-level statistical significance was modest (as expected for a controlled adjustment), stock-level effects were robust and consistent.
Why Conservative Won
Larger adjustments produced marginally higher headline returns, but also increased model sensitivity.
We chose the smallest intervention that:
- Corrected the structural bias
- Preserved portfolio behavior
- Avoided overfitting
This reflects a preference for robust improvements over optimized backtests.
What This Research Reinforced
- Intuition must be validated, not trusted
- Market effects are often non-linear
- Regime context determines signal existence
- Economic logic must accompany statistics
- Conservative changes compound better over time
Closing Thought
This research began with a common discomfort about “expensive-looking” stocks.
It ended with a clearer understanding that, in small-cap markets, returns concentrate at the extremes—not around the average.
The Deep Value adjustment was not an idea we set out to add.
It was an insight the data compelled us to implement—carefully.