Yes — using fixed balance mode in MagicTradeBot during backtesting is highly recommended. This ensures that all DCA and position sizing calculations are based on a consistent account balance, allowing for repeatable, reliable, and comparable results across different market scenarios.

🔹 1️⃣ How It Works

1. Fixed Balance Mode

   • You manually set a balance value in the backtesting configuration (e.g., $5,000)
   • The bot uses this fixed balance as the reference for all initial and DCA order sizing
   • Ignores your live or fluctuating account balance

2. Consistent DCA Calculations

   • Ensures that total_percent_investment_per_trade, size_multiplier, and max_orders produce identical order sizes in each backtest run
   • Prevents variations caused by previous trade outcomes affecting subsequent position sizing

3. Repeatable Strategy Testing

   • Provides accurate comparison between different DCA strategies, settings, or money management sequences
   • Critical for testing across diverse symbols or market volatility conditions

🔹 2️⃣ Practical Example

Backtesting Settings:

auto_calculate_initial_amount: true
balance: 5000  # fixed balance for backtesting
total_percent_investment_per_trade: 5
max_orders: 3
size_multiplier: 1.2

• Max DCA exposure = 5% of $5,000 → $250
• Initial order = $75
• DCA 1 = $90
• DCA 2 = $85 (scaled to fit total $250)

Using a fixed balance ensures these order sizes are consistent across all backtesting simulations, regardless of hypothetical PnL from prior trades.

🔹 3️⃣ Key Benefits

🔹 4️⃣ Key Takeaways

• Fixed balance mode is ideal for backtesting and demo simulations
• Ensures predictable and repeatable DCA sizing
• Makes strategy evaluation accurate across different market conditions
• Supports testing of risk management, money management sequences, and leverage effects without interference from fluctuating balances

In short, fixed balance mode provides a controlled environment for evaluating DCA strategies, letting you see exactly how your settings would perform in live markets without unpredictable variables.