Crypto Perp Trend-Pullback Backtester

Backtests a trend-following, pullback-entry strategy on Binance USDⓈ-M perpetual futures (1h candles) and writes an honest report:

• Results after an out-of-sample split shown first
• Funding payments included
• Strict-fill stress check (worse fills and stop slippage)
• Random-direction significance test
• Calendar-year breakdown and buy & hold comparison
• "state" mode: current trend facts of the strategy's EMA filter (direction, last change, volatility), with no price levels

It does not place orders, needs no exchange API keys and gives no buy/sell recommendations or trade levels. It runs locally in your Claude Code with Python 3.8+ (standard library only) and public Binance market data.

Disclaimer
All outputs of this tool, including backtest results, statistics, reports and trend information, are for research and educational purposes only. They do not constitute financial, investment or trading advice, and they are not a recommendation to buy or sell any asset. Backtest results are hypothetical: they rely on historical data and simplifying assumptions and do not guarantee future performance. Actual results can differ materially, including losses. Trading cryptocurrencies and leveraged perpetual futures involves substantial risk, including liquidation and the loss of your capital. You are solely responsible for your own investment and trading decisions.