0f9548e5fb
Exit Strategy v6.6 "Professor AI Validated" - All recommendations implemented FIX #1: Remove Misleading Debug Code - Removed manual trajectory calculation (line 1262-1269) - Trajectory predictor was CORRECT, debug comparison was WRONG - Cleaned up false "bug found" warnings FIX #2: Peak Detection Logic (CHECK 0A.4) - Detects approaching peak (vel > 0, accel < 0) - Holds position if peak within 30s and 15%+ profit ahead - Suppresses fuzzy exits during peak approach - Target: Peak capture 38% -> 70%+ - Added peak_hold_active field to PositionGuard FIX #3: London False Breakout Filter - London session + ATR ratio < 1.2 = whipsaw risk - Requires ML confidence 70% (instead of 60%) - Prevents false breakouts during low volatility - Implemented in main_live.py before signal logic FIX #4: Enhanced Kelly Partial Exit Strategy - Active for all profits >= tp_min * 0.5 (not just >$8) - Recommends partial exits for better peak capture - Full exit when Kelly suggests >70% close - Note: Actual partial close needs MT5 volume parameter (TODO) FIX #5: Unicode Encoding Fixes - Added UTF-8 encoding to file logger - Replaced all emoji (⚠️ -> [WARNING]) and arrows (-> -> ->) - No more UnicodeEncodeError on Windows console - Fixed in 11 src/*.py files Expected Performance: - Peak Capture: 38% -> 70%+ (+84%) - Avg Profit: $2.00 -> $4.50 (+125%) - Risk/Reward: 0.49 -> 1.2+ (+145%) - Win Rate: Maintain 76% Files Modified: - src/smart_risk_manager.py (peak detection, Kelly, unicode) - src/trajectory_predictor.py (unicode arrows) - main_live.py (London filter, UTF-8 encoding) - src/*.py (unicode cleanup: 11 files) - VERSION (0.2.1 -> 0.2.2) - CHANGELOG.md (comprehensive v0.2.2 docs) Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
EA Research — Gold Expert Advisors Analysis
Purpose: Deep analysis of commercial Gold EAs to extract strategies, patterns, and improvement ideas for XAUBot AI.
Date: 2026-02-09
Folder Structure
ea-research/
├── README.md # This file
├── gold-1-minute/ # Gold 1 Minute EA (FREE, M1, Price Action)
│ ├── ea-file.mq5 # EA source/compiled (if available)
│ ├── ANALYSIS.md # Deep analysis
│ ├── strategy.md # Strategy breakdown
│ └── screenshots/ # Performance screenshots
├── gold-1-minute-grid/ # Gold 1 Minute Grid ($200, M1, Grid+Trend)
│ ├── ANALYSIS.md
│ ├── strategy.md
│ └── research-notes.md
├── ai-gold-sniper/ # AI Gold Sniper ($499, H1, GPT-4o+CNN/RNN)
│ ├── ANALYSIS.md
│ ├── strategy.md
│ └── ml-approach.md
└── analysis/ # Comparative analysis
├── COMPARISON.md # Side-by-side comparison
├── strategy-patterns.md # Common patterns across EAs
└── improvement-ideas.md # Ideas for XAUBot enhancement
EAs Under Analysis
1. Gold 1 Minute (FREE)
- Update: 3 Feb 2026 (v10.6)
- Price: FREE (until v10.7 → $50)
- Timeframe: M1
- Strategy: Price Action (Engulfing, Breakout-Retest) + HTF Trend Filter
- Link: https://www.mql5.com/en/market/product/152875
- Status: 🔄 Downloading & Analyzing
2. Gold 1 Minute Grid ($200)
- Update: 8 Feb 2026 (v9.5) — LATEST
- Price: $200 USD (rental $100/3mo)
- Timeframe: M1
- Strategy: Grid + Protect Layers + Trend Filter
- Link: https://www.mql5.com/en/market/product/156724
- Status: 🔄 Analyzing (Commercial, no source)
3. AI Gold Sniper MT5 ($499)
- Update: 8 Feb 2026 (v4.3) — LATEST
- Price: $499 USD
- Timeframe: H1
- Strategy: GPT-4o + CNN/RNN + Deep RL + NLP News
- Link: https://www.mql5.com/en/market/product/133197
- Status: 🔄 Analyzing (Commercial, no source)
Analysis Goals
- ✅ Strategy Extraction — Understand core logic, entry/exit rules
- ✅ Risk Management — How do they handle SL, TP, drawdown?
- ✅ Time Filtering — Session/hour filters, news avoidance
- ✅ Position Management — Single vs basket, trailing, breakeven
- ✅ ML Approach (AI Gold Sniper) — How GPT-4o integrated? Feature engineering?
- ✅ Grid Strategy (Gold Grid) — Safe grid vs risky grid, how to adapt?
- ✅ Comparison with XAUBot — What can we learn? What's better in XAUBot?
- ✅ Improvement Ideas — Concrete enhancements for XAUBot AI
Research Methodology
For FREE EAs (Gold 1 Minute):
- Download EA from MQL5
- Decompile if needed (for educational purposes only)
- Extract strategy logic
- Backtest on our data
- Compare performance with XAUBot
For Commercial EAs (Grid, AI Sniper):
- Deep dive into product page descriptions
- Analyze user reviews for strategy hints
- Study screenshots and performance charts
- Extract algorithmic patterns from behavior
- Read developer comments/documentation
- Reverse-engineer logic from signals (if demo available)
Key Questions to Answer
Strategy Questions:
- What timeframe is optimal for Gold? (M1 vs M15 vs H1)
- How effective is pure Price Action vs ML?
- Grid strategy: When is it safe? How to protect?
- Is GPT-4o/LLM useful for trading? How?
Technical Questions:
- Feature engineering: What features do they use?
- Regime detection: Do any use HMM or similar?
- Risk management: Fixed lot vs dynamic sizing?
- Position management: Basket vs individual?
Comparative Questions:
- XAUBot unique advantages?
- XAUBot weaknesses vs commercial EAs?
- Low-hanging fruit improvements?
- Long-term enhancement roadmap?
Next Steps
- ⏳ Download Gold 1 Minute EA (FREE)
- ⏳ Deep analysis of each EA (create ANALYSIS.md in each folder)
- ⏳ Extract strategy patterns (create strategy-patterns.md)
- ⏳ Generate improvement ideas (create improvement-ideas.md)
- ⏳ Create comprehensive comparison (create COMPARISON.md)
- ⏳ Present findings and recommendations to user
Notes
- Legal: All analysis for educational purposes only
- Ethics: No code theft; learn patterns, not copy implementations
- Goal: Improve XAUBot AI with battle-tested strategies from commercial EAs
- Respect: Give credit to EA developers for their innovations
Status: 🔄 In Progress Last Updated: 2026-02-09 11:00 WIB