Price Action Patterns: Flag Breakout on Confirmed Market Structure (Runnable Notebook)

Date: 2026-06-01 IST
Sources: MQL5 Price Action Analysis Toolkit (Part 21 "Market Structure Flip Detector", Part 66 H&S, Part 69 Flags) by lynnchris. Archived .mq5: references/MQL5/lynnchris/implemented/. Core Rust: quantwave-core/src/indicators/{market_structure,geometric_patterns}.rs + synthetic generators in test_utils.rs. Visuals via docs/gen_pa_visuals.py.

This runnable Marimo notebook is the canonical executable reference for QuantWave's Price Action suite. It demonstrates the exact production pattern recommended in the four dedicated guides:

• Build only on confirmed MarketStructure bias (structure_count ≥ 2 before any BOS flip or pattern).
• Filter further with regime (HMM) + ML features (e.g. Hurst, Trendflex).
• Size dynamically using pole_length_atr (or height_atr) from the rich event structs.
• Consume PAEvent / FlagPattern / HsPattern / SRInteraction for event-driven logic.

What the Notebook Covers

• Synthetic data generators that exactly reproduce the invariants tested in Rust proptests (clean bull flag after confirmed bullish structure; perfect bear H&S; violation cases for robustness).
• Step-by-step: Market Structure bias + flip detection → Geometric (Flag / H&S) detection on the same swings → rich metadata extraction → position sizing math → simple trade log with R-multiples.
• Fallback pure-Python implementation (for browser / no-quantwave runs) that mirrors the Rust Next<T> logic.
• Notes on wiring the real quantwave Polars + streaming surfaces (identical results by design).

Run It

python -m marimo edit docs/examples/notebooks/pa_flag_breakout_strategy.py
# or: marimo run ... (after `pip install "quantwave[all]" marimo polars numpy`)

Companion Documentation (Read These First)

The notebook is intentionally code-first. For full explanations, field semantics, when-to-use, ML ideas, and annotated visuals, start with the dedicated pages:

• Market Structure (Swings + Confirmed BOS)
• Geometric Patterns (Flags + Head & Shoulders)
• S/R Interactions
• Using Rich PA Events & Metadata for Strategies and ML

All four + this notebook + the visual assets in docs/assets/pa-visuals/ were professionalized under epic quantwave-p1k6 (tasks n6e7 / za0u / 0ywt).

Parity & Fidelity

Every example here is designed to produce identical output to the Rust Next implementations and Polars .ta.* expressions (validated by the property tests in quantwave-core/tests/ and the generators in test_utils.rs). The synthetic vectors are the single source of truth for the documented behaviors.

Visuals

The notebook text describes the charts; the high-quality annotated PNGs (bos_flip.png, bull_flag.png, bear_head_shoulders.png, sr_interactions.png) live in the dedicated guides and were generated under the visual strategy in DOCUMENTATION_DECISIONS.md.

---

This notebook + the four dedicated guides together deliver the complete, professional, leak-free documentation for the PA / geometric features. No internal planning references remain in public content.