Research note
TA Strategy Backtest: a catalog has no universal PASS
Seventy-five registered experiments keep their parameters, metrics and ledgers. No fixed catalog-wide performance or passing verdict is recorded.
Backtest verdict: Unknown — no frozen result recorded in the inventory. Registry membership is not validation; every technical strategy needs its own captured request and native result.
Open this backtest in Backtesting
The rule
Choose a real registered strategy and its schema parameters, ticker and history window. Signals, filters, warmup, exits and volume requirements are strategy-specific. The shared full-results runner starts with 1,800 days, round-trip transaction cost 5 bps, slippage 2 bps, no stop/target, unit position size, no shorts and benchmark enabled. These are defaults, not a tested recommendation.
Changing cost, short support, stops, target, position size or indicator lengths defines a different experiment. The full view retains the native metric set, benchmark and trade ledger. A null profit factor stays n/a. The Lab-only SSO saved definition is not a runnable catalog option.
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 native technical engine defines signal/fill ordering and reports its data and configuration. The inventory establishes no frozen ticker/strategy/window, source price basis, closed-trade sample, significance or universal pass criterion for the page. Those must come from the exact request/result and strategy implementation, not an example URL.
Catalog and compact chart/stock aliases share the same engine, but a generic SMA crossover does not inherit the verdict of a separately specified Famous Timing study. Selecting the best of many strategies or parameter settings requires separate validation; an attractive in-sample ledger is insufficient.
Results
| Study / variant | Inventory evidence | Window / basis | Strategy return / CAGR | Benchmark | Max drawdown | Closed trades / observation N | Significance / IC | Native verdict |
|---|---|---|---|---|---|---|---|---|
| Registered technical experiments | D10 / F01 / F13 / H01 | 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. Neither the catalog size nor its twelve available metrics supplies a result sample or a PASS. Failure, unavailable data, empty ledgers and nullable statistics retain their native meanings. No new strategy, vote or execution capability is introduced.
Sources and live backtest
INVENTORY D10, F01/F13/H01 and the individual strategy definitions in Appendix A. Sources to check: frontend/components/BacktestRunner.tsx, BacktestResults.tsx, frontend/lib/taBacktest.ts, backend/app/scoring/technical/service.py and backtest.py. For any claim, capture ticker/id/parameters/fill/window and the native response. The Lab overview covers its separate authoring and frozen-definition evidence.