Binance Volume Trader Bot

A high-performance real-time trading bot that monitors Binance futures markets for volume imbalances and records price movements.

Features

• Real-time WebSocket streaming from Binance futures markets
• 100ms time window aggregation for ultra-short-term analysis
• Volume skew detection to identify buying/selling pressure
• Forward price measurement at multiple time horizons (50ms, 100ms, 250ms, 500ms)
• CSV data recording with automatic file rotation
• Prometheus metrics for monitoring
• Health check endpoints for service monitoring
• Automatic reconnection with exponential backoff
• Server time synchronization with Binance

Installation

1. Install dependencies:

pip install -r requirements.txt

2. Configure the bot by editing config/config.yaml

3. Run the bot:

python -m src.main --config config/config.yaml

Usage

Command Line Options

python -m src.main [OPTIONS]

Options:
  -c, --config PATH        Configuration file path [default: config/config.yaml]
  -l, --log-level LEVEL    Logging level (DEBUG/INFO/WARNING/ERROR) [default: INFO]
  -f, --log-format FORMAT  Log format (json/console) [default: console]
  --dry-run               Run without writing output files
  --help                  Show this message and exit

Configuration

Edit config/config.yaml to configure:

• Symbols: Trading pairs to monitor (e.g., BTCUSDT, ETHUSDT)
• Trigger conditions: Volume thresholds and skew ratios
• Time windows: Aggregation window size and horizons
• Storage: Output directory and file rotation settings

Output Data

The bot writes CSV files to data/records/ with the following columns:

• Timestamp and symbol information
• Volume data (buy/sell quantities and counts)
• Skew indicators (dominant side, skew ratio)
• Price data (baseline, min, max, first, last)
• Forward price changes at each horizon
• Metadata (measure type, latency)

Monitoring

Prometheus Metrics

Access metrics at http://localhost:9108/metrics

Key metrics include:

• Event processing rates
• Trigger detection counts
• Processing latency histograms
• Connection status
• Queue utilization

Health Check

Access health status at http://localhost:9109/health

The health check monitors:

• WebSocket connection status
• Processing performance
• Queue utilization
• Event rates

Architecture

The bot consists of several modules:

• WebSocket Manager: Handles Binance stream connections
• Window Aggregator: Aggregates trades into 100ms windows
• Trigger Detector: Identifies volume imbalance conditions
• Price Measurement: Tracks prices at future time points
• Data Recorder: Writes trigger events to CSV files
• Metrics Collector: Exposes Prometheus metrics
• Health Check: Provides service health endpoints

Development

Running Tests

pytest tests/

Code Formatting

black src/
flake8 src/

License

MIT License