Skip to content

RQF-ML Trading System

  • Set up in 5 minutes

    Install the RQF-ML system and get your first indicator running on TradingView

    Getting started

  • ICT Concepts

    Learn the ICT trading methodology - BoS, MSS, Order Blocks, Fair Value Gaps

    Concepts

  • PineScript v6

    Production-ready indicators with non-repaint compliance and webhook alerts

    PineScript

  • ML Pipeline

    Train models to predict setup quality and optimize entry timing

    Python/ML

What is RQF-ML?

RQF-ML (Range Qualification Framework with Machine Learning) is a professional-grade trading system that implements ICT (Inner Circle Trader) concepts with ML optimization for futures, crypto, and forex markets.

Core Pattern: RQF Sequence

BoS → Sweep → MSS → Entry
Stage Description What Happens
BoS Break of Structure Price breaks a swing point, establishing trend
Sweep Liquidity Grab Price wicks beyond level, closes back (trap)
MSS Market Structure Shift Price breaks trigger level, confirming reversal
Entry Discount Zone Price retraces to OB/FVG confluence for entry

System Architecture

graph TD
    A[TradingView Pine v6] -->|Webhook Alerts| B[Webhook Handler]
    B -->|Risk Check| C{Risk Layer}
    C -->|Paper| D[Paper Trading]
    C -->|Live| E[MT5/Exchange]

    F[Python Backtest] -->|Validate| G[ML Training]
    G -->|Deploy| A

    D -->|Metrics| H[Monitoring]
    E -->|Metrics| H
    H -->|Drift Detection| I[Alerts]

For Traders

For Developers

For Operations

Current Status

Strategy Version Phase Status
STRAT_02 (RQF) v12.0 Phase 1 MVP Paper Testing
STRAT_01 (Session) v9.1 Maintenance Paused
STRAT_03 (Advanced) v1.0 Research Development

Features

  • Pivot detection (dual-tier for labels and structure)
  • BoS/MSS with trend tracking
  • Sweep detection with S1, S2, S3... labeling
  • Equal Highs/Lows (EQH/EQL) liquidity pools
  • Order Blocks with Breaker conversion
  • Fair Value Gaps (FVG) in discount zones
  • Fibonacci retracement zones (discount/auction/premium)
  • Fibonacci extensions for targets (1.0, 1.272, 1.618, 2.0)
  • TP hit tracking and cleanup
  • HTF bias alignment (1H/4H arrows)
  • Confluence scoring (0-100 with A-D grades)
  • JSON webhook format for all signals
  • Entry, TP, SL, and invalidation alerts
  • Paper/Live routing
  • Risk layer integration
  • Kill switch support
  • Feature engineering from patterns
  • LightGBM/XGBoost models
  • Walk-forward validation
  • Drift detection
  • Automated retraining

Getting Help

Need Help?

  • Documentation Issues: Open an issue on GitHub
  • Feature Requests: Use the GitHub Discussions
  • Skills Questions: Run /skills-list in Claude Code