event_driven_index_add on aggressive KILL
Did not survive the pre-registered gate. This is a first-class result, not an error: the honest thing to do with a hypothesis that fails is publish the kill.
The mechanical S&P 500 index-add trade (buy on announcement, sell a few sessions after the effective date, long-only, capacity-limited fixed- fraction tranches) wrapped as a synthetic instrument (EVTIDX) and added to the plain aggressive universe. Tests whether this capacity-limited, filing-driven flow effect survives the full deflation/walk-forward/PBO gate at micro-cap-realistic transaction costs, net of the harness's own calibration and no-lookahead verification.
The pre-registered gate
| Deflated incremental edge | +0.042 PASS (threshold +0.000) |
| Walk-forward out-of-sample uplift | +0.062 PASS (threshold +0.000) |
| Probability of backtest overfitting (PBO) | 0.00 PASS (threshold 0.50) |
No-lookahead probe: PASSED (divergence date 2024-03-27)
The underlying comparison
In-sample figures are the best-of-menu figure on one price history and are upward-biased by selection. Where they disagree with the deflated or walk-forward out-of-sample figures above, those win.
| In-sample Sharpe, candidate vs baseline | +1.080 vs +1.038 (uplift +0.042) |
| Walk-forward out-of-sample Sharpe, candidate vs baseline | +1.062 vs +1.000 |
| PBO combinations evaluated | 924 |
| Correlation to the plain baseline book | +0.998 (n=3851) |
Check the source
Every number above is read straight from these committed files, nothing recomputed for this page:
data/research/event_driven_index_add-aggressive.json # the ORIGINAL pre-registered verdict, machine readable (never rewritten) data/research/event_driven_index_add-aggressive.md # the same original verdict, written out data/backtest/validation/tightened-gate-reverification.json # the CURRENT authoritative verdict, re-run under the tightened gate data/research/event_driven_index_add-aggressive.verification.json # trust ladder rung for the original verdict