Lab
Choose a market and a strategy. The calculation runs in your browser, and nothing goes to the server. Language reference
The rule stays the same, and the history is split into two parts. If the result falls apart on the second, “new” part, the rule is most likely fitted to the past.
A grid search over two parameters. A robust rule gives an even “green zone”; a lone bright cell among red ones is a sign of chance luck.
The same run with fees and slippage two and four times higher. Short-term strategies with frequent trades usually do not survive this check.
A backtest is a calculation by set rules on historical data: trades at the next candle’s open, the stop before the take-profit, a fee and slippage on every trade. It does not account for taxes, dividends on individual stocks, return on cash out of the market, the cost of shorting or liquidity, and it is easy to fit to history. Past returns do not guarantee or promise future returns. The calculation runs in your browser and is not individual investment advice.