Trading Modes¶

ArBot supports three distinct trading modes, each designed for different use cases and risk levels. Understanding these modes is crucial for successful arbitrage trading.

Overview¶

Mode	Purpose	Risk Level	API Keys Required	Real Money
Simulation	Learning & Testing	None	No	No
Live	Real Trading	High	Yes	Yes
Backtest	Strategy Analysis	None	No	No

Simulation Mode¶

Purpose and Use Cases¶

Perfect for: - Learning how arbitrage works - Testing configuration changes - Evaluating strategies risk-free - Training new users - Debugging system issues

What it does: - Uses real-time market data from exchanges - Simulates trade execution with paper money - Tracks hypothetical profit and loss - Records all trading decisions

How Simulation Works¶

graph TD
    A[Real Market Data] --> B[Arbitrage Detection]
    B --> C[Trade Signal Generated]
    C --> D[Simulate Order Execution]
    D --> E[Update Paper Portfolio]
    E --> F[Track P&L]
    F --> G[Display Results]

Data Flow: 1. Real Price Data: Fetches live prices from exchanges 2. Opportunity Detection: Identifies arbitrage opportunities 3. Simulated Execution: Pretends to place orders 4. Portfolio Tracking: Updates virtual balance 5. Performance Analysis: Records results for analysis

Simulation Features¶

Realistic Trading Simulation - Uses actual bid/ask prices - Includes trading fees in calculations - Accounts for slippage estimates - Respects minimum order sizes

Paper Trading Benefits

# Example simulation trade
{
    "symbol": "BTCUSDT",
    "buy_exchange": "bybit",
    "sell_exchange": "binance",
    "buy_price": 43250.50,
    "sell_price": 43295.80,
    "quantity": 0.1,
    "gross_profit": 4.53,
    "fees": 0.86,
    "net_profit": 3.67,
    "profit_percent": 0.0085,
    "executed_at": "2024-01-15T10:30:45Z",
    "mode": "simulation"
}

Configuration for Simulation¶

Recommended Settings:

{
  "trading_mode": "simulation",
  "arbitrage": {
    "min_profit_threshold": 0.005,
    "trade_amount_usd": 100.0,
    "max_symbols": 100
  },
  "ui": {
    "refresh_rate_ms": 1000
  }
}

No API Keys Required - Uses public market data only - No authentication needed - Works with exchange APIs read-only access

Live Trading Mode¶

Purpose and Use Cases¶

Use when: - You have tested thoroughly in simulation - You understand the risks involved - You have proper API keys set up - You're ready to trade with real money

High Risk Mode

Live mode trades with real money. Losses are real and can be significant. Only use after extensive testing in simulation mode.

Prerequisites for Live Trading¶

Technical Requirements: 1. Valid API Keys for each exchange 2. Sufficient Account Balance on exchanges 3. Stable Internet Connection (low latency preferred) 4. Proper Risk Management settings configured

API Key Permissions:

# Required permissions per exchange
Binance: "Enable Reading", "Enable Spot & Margin Trading"
Bybit: "Read", "Trade"
OKX: "Read", "Trade"
Bitget: "Read Only", "Trade"

Live Trading Flow¶

graph TD
    A[Market Data] --> B[Opportunity Detection]
    B --> C[Risk Check]
    C --> D{Risk OK?}
    D -->|Yes| E[Place Buy Order]
    D -->|No| F[Skip Opportunity]
    E --> G[Monitor Fill]
    G --> H[Place Sell Order]
    H --> I[Monitor Fill]
    I --> J[Record Trade]
    J --> K[Update Portfolio]

Execution Process: 1. Opportunity Identification: Find profitable spread 2. Risk Validation: Check all risk parameters 3. Order Placement: Submit real orders to exchanges 4. Execution Monitoring: Track order fills 5. Profit Realization: Complete arbitrage cycle 6. Record Keeping: Log all transactions

Live Trading Features¶

Real Order Execution - Places actual buy/sell orders - Manages order lifecycle - Handles partial fills - Implements timeout mechanisms

Risk Management Integration

# Live trading risk checks
def validate_trade(opportunity):
    checks = [
        check_balance_sufficient(),
        check_max_position_size(),
        check_concurrent_trade_limit(),
        check_max_drawdown(),
        check_api_connection(),
        check_market_conditions()
    ]
    return all(checks)

Real-Time Portfolio Management - Tracks actual account balances - Monitors open positions - Calculates real P&L - Updates risk metrics

Configuration for Live Trading¶

Critical Settings:

{
  "trading_mode": "live",
  "arbitrage": {
    "min_profit_threshold": 0.01,
    "trade_amount_usd": 100.0,
    "max_spread_age_seconds": 2.0,
    "use_trend_filter": true
  },
  "risk_management": {
    "max_drawdown_percent": 5.0,
    "stop_loss_percent": 2.0,
    "max_concurrent_trades": 3,
    "balance_threshold_percent": 10.0
  }
}

Environment Variables:

# Required for live trading
BINANCE_API_KEY=your_binance_key
BINANCE_API_SECRET=your_binance_secret
BYBIT_API_KEY=your_bybit_key
BYBIT_API_SECRET=your_bybit_secret

Backtest Mode¶

Purpose and Use Cases¶

Ideal for: - Strategy development and validation - Historical performance analysis - Parameter optimization - Risk assessment - Academic research

What it provides: - Historical performance metrics - Strategy comparison capabilities - Risk analysis over time - Market condition sensitivity analysis

Backtest Data Sources¶

Database Source - Uses historical data stored in ArBot database - Includes actual spreads and opportunities found - Limited to periods when bot was running

CSV Source - Import external historical data - Support for OHLCV format - Custom data preprocessing

Backtest Configuration¶

Basic Setup:

{
  "trading_mode": "backtest",
  "backtest": {
    "start_date": "2024-01-01",
    "end_date": "2024-03-31",
    "initial_balance": 10000.0,
    "data_source": "database"
  }
}

Advanced Parameters:

{
  "backtest": {
    "start_date": "2024-01-01",
    "end_date": "2024-12-31",
    "initial_balance": 10000.0,
    "data_source": "csv",
    "csv_path": "data/historical_prices.csv",
    "commission": 0.001,
    "slippage": 0.0005,
    "market_impact": 0.0001
  }
}

Backtest Features¶

Comprehensive Analysis - Win/loss ratios - Maximum drawdown - Sharpe ratio calculation - Monthly/quarterly performance - Risk-adjusted returns

Strategy Optimization

# Example backtest results
{
  "period": "2024-01-01 to 2024-03-31",
  "total_trades": 1247,
  "winning_trades": 1089,
  "losing_trades": 158,
  "win_rate": 87.3,
  "total_return": 24.6,
  "max_drawdown": 3.2,
  "sharpe_ratio": 2.1,
  "profit_factor": 3.8,
  "average_trade": 0.0197,
  "best_trade": 1.24,
  "worst_trade": -0.87
}

Switching Between Modes¶

Safe Mode Transitions¶

Simulation → Live: 1. ✅ Test extensively in simulation 2. ✅ Configure all risk parameters 3. ✅ Set up API keys and permissions 4. ✅ Start with small position sizes 5. ✅ Monitor closely for first few trades

Live → Simulation: 1. ✅ Stop all active trading 2. ✅ Wait for open positions to close 3. ✅ Switch mode in configuration 4. ✅ Restart application

Any Mode → Backtest: 1. ✅ Ensure historical data available 2. ✅ Configure date range 3. ✅ Set initial balance 4. ✅ Run analysis

Mode-Specific Considerations¶

Simulation Mode Best Practices: - Use realistic position sizes - Test different market conditions - Monitor for extended periods - Validate all configuration changes

Live Trading Best Practices: - Start with conservative settings - Monitor constantly during initial period - Have stop-loss mechanisms ready - Keep detailed trading logs

Backtest Mode Best Practices: - Use sufficient historical data - Account for market regime changes - Validate with out-of-sample data - Consider transaction costs realistically

Performance Comparison¶

Expected Results by Mode¶

Metric	Simulation	Live Trading	Backtest
Accuracy	High	Moderate	Variable
Execution	Instant	Market-dependent	Instant
Costs	Estimated	Actual	Estimated
Slippage	Estimated	Actual	Estimated
Market Impact	None	Actual	Estimated

Factors Affecting Live Performance¶

Network Latency: - Geographic distance to exchanges - Internet connection quality - VPS vs local execution

Market Conditions: - Volatility levels - Liquidity availability - Competition from other arbitrageurs

Execution Quality: - Order fill rates - Slippage experienced - Fee structures

Troubleshooting by Mode¶

Simulation Issues¶

No opportunities found: Lower profit threshold
Unrealistic profits: Check fee calculations
Performance issues: Reduce max symbols

Live Trading Issues¶

Orders not executing: Check API permissions
High slippage: Reduce position sizes
Connection errors: Verify network stability

Backtest Issues¶

No historical data: Check data source
Incomplete results: Verify date range
Performance discrepancies: Review parameters

Mode Selection Guide

New users: Start with Simulation
Strategy testing: Use Backtest + Simulation
Production trading: Live (after thorough testing)
Research/analysis: Backtest with historical data

Last update: July 12, 2025
Created: July 12, 2025