RESEARCH FRAMEWORK
Range mean reversion
Tests whether qualified boundary excursions revert toward a range midpoint.
Only the measurements the rule needs
Range boundariesRSIADX / DMIATR
Conditions before an entry exists
- Minimum two prior contacts per boundary
- Range width comfortably exceeds all-in costs
- Trend-strength filter remains below its declared threshold
ENTRY
After rejection closes back inside a pre-existing boundary.
STOP
Beyond the excursion with a fixed volatility buffer.
EXIT
Midpoint first; test opposite boundary only as a separate variant.
DO NOT TRADE
Fresh trend emergence, major scheduled events and repeatedly tested late-stage boundaries.
What has to happen before capital is at risk
1Timestamp range formation
2Never use future contacts
3Track breakout losses
4Bucket by range age
Report gross and net results, the full eligible sample, maximum drawdown, trade distribution and any rule changes. Then repeat on data that did not influence the design.
Use the same protocol every time.
The CSV test plan records eligibility, costs, execution, outcome and rule adherence without turning an example into a promise.
Download test-plan CSV ↓