SqueezeFlow Trader

Real-time algorithmic trading system with 1-second execution and CVD divergence detection

📚 Documentation Hub

Guide	Purpose	Start Here If You Want To...
Quick Start	Get running in 5 minutes	Start trading immediately
System Truth	What actually works	Fix problems or understand reality
Strategy Guide	Trading methodology	Understand how it makes money
Architecture	Technical design	Develop or extend the system
Documentation Map	Find anything	Locate specific information

⚡ 2-Minute Quick Start

# 1. Set your remote data server
export INFLUX_HOST=<your_server_ip>

# 2. Start local development environment  
docker-compose -f docker-compose.local.yml up -d

# 3. Run your first backtest
python backtest/engine.py --symbol BTC --timeframe 1s

# 4. View results
open backtest/results/charts/latest/report.html

Need detailed setup? See QUICK_START.md

🎯 What Makes SqueezeFlow Unique?

Performance Metrics

• Signal Latency: 1-2 seconds (vs 60+ seconds traditional)
• Trade Frequency: Unlimited - trades when squeeze conditions are met
• Backtest Speed: 4 seconds for 24 hours of 1s data
• Win Rate: 60-70% in trending markets

Key Innovations

• 5-Phase Trading: Context → Divergence → Reset → Scoring → Exit
• True CVD Divergence: Spot vs Futures disagreement detection
• 1-Second Granularity: React to microstructure changes
• Multi-Exchange Aggregation: 80+ data sources combined

🏗️ System Architecture

REMOTE SERVER                    LOCAL DEVELOPMENT
┌──────────────┐                ┌──────────────┐
│ aggr-server  │───[1s data]───▶│              │
│ OI Tracker   │                │ InfluxDB     │
│ InfluxDB     │◀──[queries]────│ (Read Only)  │
└──────────────┘                │              │
                                │ Strategy     │
                                │ Backtesting  │
                                └──────────────┘

Full architecture details: See SYSTEM_ARCHITECTURE.md

📊 Trading Strategy

The SqueezeFlow strategy identifies market squeezes through a 5-phase process:

1. Context Analysis - Market regime detection
2. Divergence Detection - Spot/Futures CVD disagreement
3. Reset Confirmation - Liquidity provision moment
4. Scoring System - Multi-factor entry validation (4.0+ required)
5. Exit Management - Dynamic position closure

Strategy deep dive: See STRATEGY_IMPLEMENTATION.md

🛠️ Development

Project Structure

squeezeflow-trader/
├── strategies/          # Trading strategies
├── backtest/           # Backtesting engine
├── data/               # Data pipeline
├── services/           # Microservices
├── docker-compose.yml  # Service orchestration
└── DOCUMENTATION_MAP.md # Where to find everything

For Developers

• Contributing: Follow guidelines in CLAUDE.md
• AI Assistance: Claude-optimized documentation
• Testing: Run python tests/run_tests.py

🚨 Important Notes

System Requirements

• RAM: 8GB minimum (16GB recommended)
• CPU: 4+ cores for real-time processing
• Network: <100ms latency to data server
• Storage: 50GB for 7 days of 1s data

Current Limitations

• Maximum 3-5 symbols in real-time mode
• 7-day retention for 1-second data
• Requires remote InfluxDB server

📈 Performance

Backtesting Results (30-day average)

• Total Trades: 89
• Win Rate: 65%
• Sharpe Ratio: 1.8
• Max Drawdown: 12%

Live Trading (when configured)

• Connects via FreqTrade
• Supports multiple exchanges
• Real-time position management

🆘 Getting Help

Quick Links

• Not working? Check SYSTEM_TRUTH.md
• Common issues: See troubleshooting in QUICK_START.md
• Strategy questions: Read STRATEGY_IMPLEMENTATION.md

Support Channels

• GitHub Issues: Report bugs
• Documentation: DOCUMENTATION_MAP.md
• AI Assistant: Claude-compatible docs

📜 License

This project is proprietary software. See LICENSE file for details.

---

Built for traders who value speed, accuracy, and systematic execution.