xgb_core_timing_q30_pr18| Family | ml |
| Evidence tier | T0_EXPLORATORY |
| Public status | NOT_PUBLIC |
| Lab status | HISTORICAL_ONLY |
| Effective status (this eval) | HISTORICAL_ONLY |
| Lab inception (forward start, no backfill) | 2026-07-09T03:55:54Z |
| source_report | reports/research/ml_10k_feature_discovery/reality_check_summary.json |
| source_pr | 18 |
| historical_net_pnl | — |
| historical_max_dd | — |
| historical_sharpe | — |
| historical_trade_count | — |
| validation_pnl | +5.99 |
| locked_test_pnl | +299.70 |
| search_adjusted_p | +0.04 |
| reality_check_status | PASSED_BUT_VALIDATION_TEST_INCONSISTENT |
No forward evaluation: no reproducible scorer/frozen params: nominal_passer_with_validation_test_inconsistency_no_scorer
No risk overlay on this candidate.
Not ranked (no forward trades yet).
Exploratory log entry: leakage-safe idea with weak/inconclusive evidence (search-adjusted p=0.035, reality check: PASSED_BUT_VALIDATION_TEST_INCONSISTENT). Monitoring is allowed; public claims are not.
| Policy requirement | Status | Detail |
|---|---|---|
| valid scoring bundle / reproducible definition | NOT MET | no reproducible scorer/frozen params: nominal_passer_with_validation_test_inconsistency_no_scorer |
| no safety violation | PASS | paper_only, real_orders_enabled=false, no backfill (validator-enforced) |
Not eligible for private-watchlist review yet — the unmet lines above are why.
SECRET LAB — INTERNAL ONLY. Paper simulation bookkeeping. No real orders, no live trading, no execution candidate, no promotion from this page. Forward evidence starts at lab inception and is never backfilled. Not on the public dashboard; do not link from public_site.