Research · live validation
Compare strategies
Every strategy and variant gets identical metric treatment — same math, same sample-size badges. Small samples are labelled as noise, not hidden.
Candidates · not production
Not a recommendation. No real orders are placed. Exact entry, stop and target levels are never published. Educational market research only.
| Strategy | Variant | Status | n | Win % | PF | Expectancy | Sharpe | Net R | Max DD | Sample |
|---|---|---|---|---|---|---|---|---|---|---|
| Fear-Regime Demand Accumulation | 4R | Burn-in · Candidate | 34 | 8.8% | 0.34 | -0.656R | -0.46 | -22.30R | −22.30R | Provisional sample (n = 34) |
| Fear-Regime Demand Accumulation | 5R | Burn-in · Candidate | 34 | 5.9% | 0.28 | -0.743R | -0.52 | -25.27R | −25.27R | Provisional sample (n = 34) |
| RSI-2 Mean Reversion (Large Cap) | cash | Candidate · Negative seed result / Rework required | 564 | 57.1% | 0.79 | -0.042R | -0.08 | -23.47R | −33.68R | Statistically meaningful (n = 564) |
| RSI-2 Mean Reversion (Large Cap) | mtf_2_5x | Candidate · Negative seed result / Rework required | 564 | 55% | 0.72 | -0.060R | -0.12 | -33.61R | −42.00R | Statistically meaningful (n = 564) |
| D/S Graded Swing Long | trail_after_3pct | Candidate · Historical replay · Candidate live-forward tracking | 423 | 31% | 2.50 | +1.162R | 0.19 | +491.61R | −27.88R | Statistically meaningful (n = 423) |
| D/S Graded Swing Long | target_3pct_trail | Candidate · Historical replay · Candidate live-forward tracking | 423 | 47.8% | 1.65 | +0.402R | 0.20 | +169.89R | −10.49R | Statistically meaningful (n = 423) |
| Weekly Doji Low-IV Swing | low_iv | Candidate · Historical replay · Candidate live-forward tracking | 556 | 45.1% | 1.69 | +0.152R | 0.19 | +84.58R | −21.30R | Statistically meaningful (n = 6363) |
| Weekly Doji Low-IV Swing | uptrend | Candidate · Historical replay · Candidate live-forward tracking | 6363 | 45.3% | 1.61 | +0.160R | 0.18 | +1015.08R | −58.72R | Statistically meaningful (n = 6363) |
No equity-curve overlays are shown until every strategy has at least 30 closed trades — below that the comparison would be noise dressed as insight.
All metrics come from one shared engine. An R is one unit of planned risk. Sharpe/expectancy values on insufficient samples are shown as 'Too early' rather than as misleading numbers.