Documentation - errational

🚀 Quick Start Guide

Follow these simple steps to create your first trading strategy in minutes:

Sign Up / Login

Create a free account to get started. You'll receive 50 free strategy runs to test the platform.

Create Free Account

Describe Your Strategy

Use natural language to describe your trading strategy. No coding required!

Example queries:

• "Buy when RSI is below 30, sell when RSI is above 70"

• "Buy when 50-day MA crosses above 200-day MA"

• "Use Bollinger Bands for entry and exit signals"

• "Momentum strategy based on 14-day RSI"

Tip: Be specific about entry and exit conditions. The clearer your description, the better the AI can generate your strategy.

Configure Parameters

Fill in or use the auto-populated values:

Ticker Symbol: Select from available NSE stocks (e.g., RELIANCE, TCS, INFY)
Date Range: Auto-suggested based on data availability (or enter custom dates)
Initial Cash: Starting capital (default: ₹10,000)
Trading Fees: Commission per trade (default: 0.1%)

Tip: The suggested values are automatically populated based on your ticker selection. You can modify them or use the defaults.

Hit Execute Strategy!

Click the "Execute Strategy" button and let our AI do the work. The system will:

Generate Python code for your strategy
Backtest against historical data
Calculate performance metrics
Generate visualizations

Processing time: Usually takes 5-15 seconds depending on strategy complexity.

Explore Results!

View comprehensive backtest results including:

📊 Performance Metrics

• Total Return
• Sharpe Ratio
• Max Drawdown
• Win Rate

📈 Visualizations

• Portfolio value chart
• Drawdown analysis
• Trade signals overlay
• Returns distribution

You can also:

Save strategies to your library
Reuse strategies with different tickers
Run batch backtests across multiple tickers
Compare multiple strategies side-by-side

⚡ Advanced Features

📚 Strategy Library

All your strategies are automatically saved to your personal library. You can:

• View and manage all saved strategies
• Reuse successful strategies on different stocks
• Delete underperforming strategies
• Track your strategy evolution over time

🔄 Batch Backtesting

Test the same strategy across multiple stocks or parameter configurations:

Open a saved strategy from your library
Click "Batch Backtest" button
Select multiple tickers or parameter variations
Run all tests simultaneously
Compare results in a unified view

♻️ Strategy Reuse

Found a winning strategy? Reuse it instantly without consuming LLM credits:

• Click "Reuse" on any saved strategy
• Modify ticker, dates, or capital parameters
• Execute immediately (no AI generation needed)
• Saves your LLM quota for new strategy exploration

💡 Best Practices

✅ Writing Good Strategy Queries

• Be specific about entry and exit conditions
• Mention exact indicator thresholds (e.g., "RSI below 30" not just "RSI oversold")
• Specify position sizing if non-standard
• Include stop-loss or take-profit conditions if desired

📊 Choosing Test Parameters

• Use at least 1-2 years of data for reliable results
• Consider market cycles (bull/bear markets) in your date range
• Account for realistic trading fees (typically 0.05-0.2% per trade)
• Test with capital amounts similar to what you'd actually trade

🎯 Interpreting Results

• Look beyond just returns - check Sharpe ratio and max drawdown
• Higher Sharpe ratio (>1.5) indicates better risk-adjusted returns
• Max drawdown shows worst-case scenario - ensure you can tolerate it
• Win rate above 50% is good, but focus on risk-reward ratio too

⚠️ Avoiding Overfitting

• Don't over-optimize for historical data
• Test strategies on different stocks and time periods
• Simpler strategies often perform better in live trading
• Past performance doesn't guarantee future results

Still have questions?

Check out our Help Center for FAQs and troubleshooting tips

Visit Help Center