Research note
Validation report: native labels, unknown frozen results
IC, buckets, annual walk-forward and OOS checks keep their native labels. No frozen report or headline trading verdict is inventoried.
Backtest verdict: Unknown — no frozen result recorded in the inventory. Validation colors and holds_up/weakens/breaks_down are native analytical labels, not automatic PASS/FAIL trading judgments.
Open this research tool in Backtesting
The rule
The report exposes headline, composite IC, quintiles, lens IC, plumbing, annual walk-forward, out-of-sample and transitions, plus the indicator drill-down with filters and sorting. It tests model relationships to forward S&P returns rather than prescribing one entry/exit portfolio.
The annual folds and 60/40 training/OOS split are evaluation design, not sample counts or historical performance. Trade fills, shorts/cash, sizing, stops/targets and execution costs are not applicable to a pure IC/bucket analysis unless a particular native section explicitly models them. Do not borrow a cost slider from a different engine.
Methodology
Evidence as of: the INVENTORY review dated November 5, 2026; this is the consolidation planning baseline, not a new observation window. Publication does not extend any study's sample.
The report uses historical composite/indicator inputs and forward-return analyses, preserving each section's own horizon, bounds, missing-data state and methodology. Source revisions, publication lags, overlapping horizons and effective N limit interpretation. Annual walk-forward and the OOS split remain different checks; neither is the Lab's fixed-parameter honesty panel.
No frozen horizon IC, bucket return, N, significance, exact dates/basis or headline verdict is inventoried. A negative IC must be read with the indicator's declared direction; a favorable color is not a profitable trade ledger. Missing data remains no_data rather than zero or failure. No report was regenerated for this article.
Results
| Study / variant | Inventory evidence | Window / basis | Strategy return / CAGR | Benchmark | Max drawdown | Closed trades / observation N | Significance / IC | Native verdict |
|---|---|---|---|---|---|---|---|---|
| Composite / quintiles / lens / plumbing / transitions | D17 / F31 / H25 | not recorded | not recorded | not recorded | not recorded | not recorded | not recorded | Unknown — no frozen result |
| Annual walk-forward / 60-40 OOS | D17 / H25 | not recorded | not recorded | not recorded | not recorded | not recorded | not recorded | Unknown — no frozen result |
| Indicator IC drill-down | D17 / F32 / H25 | not recorded | not recorded | not recorded | not recorded | not recorded | not recorded | Unknown — no frozen result |
The linked workspace retains the complete native output and notes when available. Missing evidence is not a zero return, a successful empty result or a failed strategy.
Verdict and limitations
Unknown — no result/verdict recorded in this inventory. Native green/yellow/red/no_data and holds_up/weakens/breaks_down labels remain unchanged. They are not automatically trading PASS/FAIL, positive-return claims or new gates. A later interpretation needs the entire dated report and direction/sign of IC, including weak, failed and unavailable sections.
Sources and live research
INVENTORY D17, F31–F32/H25. Sources to check: frontend/components/validation/ValidationBoards.tsx, IndicatorICPanel.tsx, frontend/lib/strategy-pages.ts, backend/app/scoring/validation.py, indicator_ic.py and backtest_analytics.py. Open: full dated report/config, evaluation window, source/basis and IC direction before quoting any number or conclusion. The workspace retains every section and indicator detail.