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
72
Win rate
29.17%
Median return
-1.31%
Median excess
-0.86%
Median MFE
+1.33%
Median MAE
-2.28%
Cohort drawdown
-10.12%
Evidence
30+ sample
8 Sept 2026 to 18 Sept 2026Regime benchmark: Nifty 500
| Sector | Sample | Win rate | Median return | Median excess | MFE | MAE | Drawdown | Evidence |
|---|---|---|---|---|---|---|---|---|
| Financial Services | 15 | 26.67% | -1.51% | -1.79% | +1.17% | -2.65% | -13.78% | Early |
| Pharma | 11 | 27.27% | -1.36% | -0.94% | +1.42% | -1.71% | -11.29% | Early |
| Industrials | 10 | 40.00% | -1.36% | -0.93% | +1.54% | -3.07% | -10.77% | Early |
| Energy | 9 | 22.22% | -1.67% | -0.86% | +1.02% | -2.82% | -9.80% | Under 10 |
| Chemicals | 7 | 28.57% | -0.68% | +0.05% | +1.39% | -2.28% | -3.02% | Under 10 |
| Consumer | 4 | 50.00% | +0.25% | +0.24% | +2.05% | -2.09% | -2.16% | Under 10 |
| Materials | 3 | 33.33% | -1.25% | -0.83% | +1.27% | -2.12% | -1.85% | Under 10 |
| Auto | 2 | 0.00% | -2.07% | -2.04% | +0.26% | -2.53% | -2.07% | Under 10 |
| Fmcg | 2 | 50.00% | -0.35% | +0.37% | +2.19% | -1.51% | -2.89% | Under 10 |
| It | 2 | 50.00% | +0.35% | +2.05% | +3.78% | -1.53% | 0.00% | Under 10 |
| Power | 2 | 0.00% | -0.89% | +0.28% | +0.87% | -1.87% | -1.77% | Under 10 |
| Services | 2 | 50.00% | +0.17% | +0.39% | +1.34% | -2.09% | -0.10% | Under 10 |
| Media | 1 | 0.00% | -5.04% | -4.83% | +0.06% | -5.67% | -5.04% | Under 10 |
| Metals | 1 | 0.00% | -1.58% | -0.94% | +1.68% | -1.96% | -1.58% | Under 10 |
| Real Estate | 1 | 0.00% | -1.94% | -1.52% | +0.43% | -2.38% | -1.94% | 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.