|Signal Test|REJECTED
Signal layered-threshold-variance: Rejected
Instead of one binary threshold, use per-source thresholds (fotmob 3.0, matchXg 2.5, v3 2.0). When 2+ sources independently flag regression, confidence is higher. Level 3 (all flag) should have highest per-bet ROI.. not-significant
Marginal ROI
+0.3%
entry-adjusted
N
109,015
CLV
+5.7%
Gates
8/10
Hypothesis
Instead of one binary threshold, use per-source thresholds (fotmob 3.0, matchXg 2.5, v3 2.0). When 2+ sources independently flag regression, confidence is higher. Level 3 (all flag) should have highest per-bet ROI.
Result
8/10 gates passed. Marginal entry-adj ROI +0.3pp (below +0.5pp practical significance). Bootstrap p=0.39 (need <0.10). Walk-forward 13/13 folds positive but effect too small and noisy.
Gate Results
| Gate | Name | Result |
|---|---|---|
| 1 | Pre-registered | PASS |
| 2 | True standalone | PASS |
| 3 | Minimum N | PASS |
| 4 | Marginal ROI | PASS |
| 5 | Bootstrap marginal significance | FAIL |
| 6 | Matchday interleave OOS | PASS |
| 7 | Regime stratification | PASS |
| 8 | Suspicious N | PASS |
| 9 | Practical significance | FAIL |
| 10 | Walk-forward | PASS |
Rejection Reason
not-significant
Failed gates: Bootstrap marginal significance, Practical significance
REJECTEDSignal: layered-threshold-variance|2026-04-11