Evidence by cohort
Signal performance laboratory
Fixed-horizon outcomes grouped without removing failures. Medians reduce outlier distortion; MAE and drawdown expose risk that a headline win rate can hide.
Signal set
Sample
140
Win rate
20.71%
Median return
-4.51%
Median excess
-0.94%
Median MFE
+4.31%
Median MAE
-7.65%
Cohort drawdown
-34.37%
Evidence
30+ sample
8 Sept 2026 to 21 Sept 2026Regime benchmark: Nifty 500
| Sector | Sample | Win rate | Median return | Median excess | MFE | MAE | Drawdown | Evidence |
|---|---|---|---|---|---|---|---|---|
| Pharma | 22 | 4.55% | -6.89% | -4.01% | +2.02% | -8.04% | -51.37% | Early |
| Energy | 21 | 9.52% | -5.62% | -1.17% | +4.50% | -7.34% | -41.70% | Early |
| It | 17 | 64.71% | +4.72% | +6.92% | +6.51% | -5.82% | -14.22% | Early |
| Financial Services | 15 | 0.00% | -11.82% | -7.70% | +3.35% | -16.85% | -72.35% | Early |
| Industrials | 12 | 41.67% | -1.46% | +1.80% | +5.09% | -8.77% | -20.35% | Early |
| Services | 12 | 8.33% | -5.50% | -1.45% | +3.67% | -8.45% | -38.62% | Early |
| Auto | 11 | 18.18% | -4.29% | +0.27% | +3.69% | -6.69% | -29.27% | Early |
| Power | 9 | 0.00% | -4.53% | -2.15% | +0.55% | -8.14% | -27.21% | Under 10 |
| Chemicals | 7 | 14.29% | -2.81% | +1.62% | +6.28% | -7.17% | -16.52% | Under 10 |
| Consumer | 6 | 50.00% | -0.37% | +2.99% | +6.93% | -4.92% | -1.86% | Under 10 |
| Materials | 3 | 0.00% | -2.07% | +0.02% | +1.27% | -9.12% | -9.28% | Under 10 |
| Fmcg | 2 | 100.00% | +0.26% | +2.21% | +9.24% | -5.12% | 0.00% | Under 10 |
| Media | 1 | 100.00% | +6.33% | +10.85% | +11.65% | -5.80% | 0.00% | Under 10 |
| Metals | 1 | 0.00% | -11.85% | -6.01% | +1.68% | -12.06% | -11.85% | Under 10 |
| Real Estate | 1 | 0.00% | -4.09% | +0.03% | +0.43% | -10.84% | -4.09% | Under 10 |
Returns: Median is used for return, excess return, MFE, and MAE to reduce outlier distortion. Return greater than 0% at the selected fixed horizon.
Drawdown: Signals from the same publication date are equal-weighted as one cohort, then cohort returns are compounded chronologically.
Market regime: Bullish, bearish, or sideways uses the benchmark close versus its trailing 50-session average and trailing 20-session return, using only data available by publication day.
Sample discipline: Groups below 10 outcomes are insufficient; 10-29 are early evidence; 30 or more are established samples, not proof of future performance.