Freqtrade Basics: A Comprehensive Guide to Key Concepts

This guide covers fundamental concepts about how Freqtrade operates and its core functionalities.

Freqtrade Terminology

• Strategy: Your trading algorithm that instructs the bot on actions to take.
• Trade: An open position.
• Open Order: Active exchange orders awaiting completion.
• Trading Pair: Tradable currency pairs (e.g., spot pairs like XRP/USDT, futures like XRP/USDT:USDT).
• Timeframe: Candlestick duration (e.g., "5m", "1h").
• Technical Indicators: Tools like SMA, EMA, RSI, etc.
• Limit Order: Executes at specified price or better.
• Market Order: Guarantees execution, potentially affecting price volatility.
• Current Profit: Unrealized profit (used internally).
• Realized Profit: Actualized profit (relevant for partial fills).
• Total Profit: Combines realized/unrealized profits, calculated against total investment.

Fee Management

Freqtrade incorporates fees in all profit calculations:

• Backtesting/Hyperopt: Uses exchange default fees.
• Live Trading: Applies actual fees (including discounts like BNB).

👉 Optimize your trading strategy with low fees

Trading Pair Nomenclature

Follows CCXT conventions:

Spot Pairs

Format: base/quote (e.g., ETH/USDT).

Futures Pairs

Format: base/quote:settle (e.g., ETH/USDT:USDT).

Bot Execution Logic

The iterative loop runs every few seconds (internals.process_throttle_secs):

1. Initialization:

   • Fetch open trades from storage.
   • Identify tradable pairs.
   • Download OHLCV data (including informational pairs).

2. Strategy Analysis:

   • Call populate_indicators().
   • Evaluate entry/exit trends via populate_entry_trend() and populate_exit_trend().

3. Order Management:

   • Check order timeouts (check_entry_timeout(), check_exit_timeout()).
   • Adjust prices via adjust_entry_price().

4. Position Handling:

   • Validate stop-loss, ROI, exit signals.
   • Determine exit pricing (exit_pricing or custom_exit_price()).
   • Call confirm_trade_exit() before order placement.

5. Entry Validation:

   • Set entry price (entry_pricing or custom_entry_price()).
   • Calculate leverage (leverage()) for margin/futures.
   • Determine stake size (custom_stake_amount()).
   • Confirm via confirm_trade_entry().

Backtesting/Hyperopt Execution

Simulates core logic without live interactions:

• Load historical data for configured pairs.
• Call bot_start() once.
• Compute indicators (populate_indicators() per pair).
• Evaluate entry/exit signals (populate_entry_trend(), populate_exit_trend()).

• Simulate trades candle-by-candle:

  • Check timeouts (unfilledtimeout or callback-based).
  • Adjust entry prices (adjust_entry_price()).
  • Validate entries/exits (confirm_trade_entry/exit()).
  • Manage position sizing (custom_stake_amount()).
  • Handle partial exits (adjust_trade_position()).

👉 Master backtesting techniques for better results

FAQ

Q: How often are callbacks triggered in backtesting vs. live trading?
A: Backtesting calls each callback once per candle; live trading triggers them per iteration (~5s intervals).

Q: Does Freqtrade support custom fee structures?
A: Yes, use --fee parameter in backtesting/hyperopt to override defaults.

Q: What happens if a trading pair uses incorrect nomenclature?
A: The bot may fail to recognize it, causing errors like "pair unavailable."

Q: How is total profit calculated?
A: Sum of realized/unrealized profits relative to total investment.

Q: Can I adjust open positions dynamically?
A: Yes, via adjust_trade_position() callback for live trading.