Momentum Trading Strategy for Crypto Perpetual Futures ( BTC, ETH, SOL ... )

Here is the document

Overview

This is a momentum-based algorithmic trading system specifically designed for the ETH/USDT perpetual futures market on a 15-minute timeframe. The strategy combines multiple technical indicators to identify high-probability entries and exits while managing risk with dynamic stop-losses. Parameter optimization is done using Optuna to enhance performance across varying market conditions.

Video

FVRA_backtest_1.mp4

Strategy Components

Market and Timeframe

• Asset: ETH/USDT (Perpetual Futures)
• Timeframe: 15-minute candlesticks
• Leverage: 10x
• Position Size: 20% of account balance per trade
• Trading Fee: 0.06% per trade (entry and exit)

Key Indicators

• Fisher Transform (FS)
  • Identifies potential price reversals by transforming price data to Gaussian distribution
  • Parameters: Length = 10 periods
  • Entry: Bullish crossover (FS crosses above trigger)
  • Exit: Opposite crossover when momentum exceeds thresholds
• Average True Range (ATR)
  • Measures volatility to set dynamic stop-loss levels
  • Parameters: Length = 14 periods; Stop-loss multiplier applied to ATR
  • Stop-loss is dynamically trailed based on ATR
• Relative Strength Index (RSI)
  • Gauges overbought/oversold conditions to filter entries/exits
  • Parameters: Length = 14 periods
  • Entries allowed during neutral momentum
  • Exits triggered on strong directional momentum
• Volume Oscillator
  • Confirms trade entries and exits with volume momentum
  • Parameters: Short EMA = 5 periods; Long EMA = 10 periods
  • Entries require positive volume momentum
  • Exits confirmed on elevated volume momentum

Parameter Optimization

Parameters for all indicators are optimized through Optuna, an open-source hyperparameter optimization framework. The optimization seeks to maximize performance metrics such as total PnL and profit factor over historical data.

Trading Rules

Entry Conditions

Long Entry

• Bullish Fisher Transform crossover (FS > trigger)
• Low momentum in FS or trigger
• Neutral RSI momentum
• Positive volume momentum

Short Entry

• Bearish Fisher Transform crossover (FS < trigger)
• Low momentum in FS or trigger
• Neutral RSI momentum
• Positive volume momentum

Exit Conditions

• Stop-Loss: Set dynamically using ATR and trailed to lock in profits
• Signal Exit: Triggered by opposite Fisher crossover, high momentum, strong RSI, and elevated volume momentum
• Close any open positions at the last candle's close price if data ends

Backtesting Framework

• Initial balance: $1,000
• Position sizing: 20% with 10x leverage
• Data: Historical ETH/USDT 15-minute candle data (CSV input)
• Outputs:
  • CSV report with trade details (entry/exit, PnL, etc.)
  • Interactive Plotly chart showing price, indicators, and equity curve
  • Summary metrics like total PnL, win rate, profit factor, expectancy

Risk Management

• Leverage capped at 10x
• Position size limited to 20% of balance
• ATR-based dynamic stop-loss adapts to volatility
• Trading fees (~0.06%) included for entry and exit

Performance Metrics (Backtest: Jan 1, 2025 – Jul 19, 2025)

• Initial Balance: $1,000
• Final Balance: $15,291.61 (+1429.16%)
• Total Trades: 293
• Win Rate: 48.1%
• Profit Factor: 1.41
• Average Win: $351.20
• Maximum Win: $4,500.18
• Maximum Loss: -$1,045.67
• Expectancy: $48.78

While the win rate is near-even, larger winning trades significantly drive profitability.

Usage Instructions

1. Provide historical ETH/USDT 15-minute candle data in a CSV file with columns: timestamp, open, high, low, close, volume.
2. Run the Python script to compute indicators, execute backtesting, and generate reports.
3. Analyze the output CSV and interactive Plotly charts for performance insights.

Conclusion

This strategy offers a systematic and automated approach to trading ETH/USDT perpetual futures on a short timeframe. It effectively combines momentum-based indicators with rigorous risk management and leverages parameter optimization for robust results. The demonstrated backtest performance highlights significant profitability potential with clear, rule-based entry and exit signals.