Strategy Compare

The strategy-compare skill allows AI agents to evaluate and rank multiple trading strategies against each other using historical data. It leverages the VectorBT library to calculate performance metrics like Sharpe ratio, drawdown, and total return for side-by-side analysis.

When to use it

• You need to determine which of several algorithmic approaches performs best under specific market conditions.
• You want to validate a new strategy by comparing its risk-adjusted returns against established benchmarks.
• You are optimizing a portfolio by identifying the most efficient strategies based on custom criteria.

Key capabilities

• Executes backtests for multiple strategies simultaneously within a single session.
• Calculates and compares standard performance metrics including profit factor, max drawdown, and annualized return.
• Generates structured output suitable for ranking strategies or feeding into further decision-making logic.

Example prompts

• "Compare the performance of a moving average crossover strategy against an RSI mean-reversion strategy using 5 years of data."
• "Rank three different volatility-based trading models by their Sharpe ratio and maximum drawdown."
• "Analyze how a trend-following bot performed compared to a buy-and-hold approach during the last market correction."

Tips & gotchas

Ensure you provide sufficient historical data context for accurate backtesting results, as short timeframes may yield misleading comparisons. Be aware that this skill relies on the VectorBT library, so complex custom indicators may require specific formatting or pre-processing before comparison.