SignalQuarryGitHub
← T08 risk-filter enhancements

Historical study

T08: seven risk-filter combinations

7 frozen candidate rules were tested. None passed all qualification gates. Baseline and benchmark figures in the source report are not presented as candidate performance.

Completed; no qualifying challengerAdded 2026-09-23Test period: Not recorded in this metric bundle

Measured outcome

Performance at a glance

Figures use the source study's definitions. “Not available” means the source does not support that metric.

Total returnNot availableChange in simulated portfolio value over the test.
Annualized returnNot availableYearly compounded return; shown only when supported.
Worst drawdownNot availableLargest fall from an earlier portfolio high.
Sharpe ratioNot availableReturn relative to variability, under the source study's definition.
TradesNot availableCompleted simulated trades, when reported.

How this was tested

Assumptions and limits

Daily TECL and cash strategy research with documented cost, delayed execution, and qualification gates.

  • The source market-data archive is owner-held and is not redistributed; users need a separately licensed matching dataset to reproduce the run.
  • The historical periods were reused during strategy development; no untouched holdout remains.
  • All seven challengers failed at least one required gate; no challenger is promoted.
  • Original-environment reproduction is verified from the archived native ARM64 runtime and source hashes; portable parity against the migrated A code has not been claimed.
  • No separate feature-manifest file was recorded; feature definitions are bound by the source-runtime hash list and protocol.
  • The T08 baseline is a research comparator, not an activation decision; no live authorization follows from this study.