Why Backtesting Can Be Deceptive
Your backtest shows +150% returns? Before you pop the champagne — check whether you've fallen into one of the classic traps.
Mistake 1: Overfitting
You optimize parameters until they fit historical data perfectly. The problem: the strategy only works for the past, not the future. Solution: split your data. Optimize on 70%, test on the remaining 30% (out-of-sample test).
Mistake 2: Survivorship Bias
You only test coins that still exist today. But what about the hundreds of coins that were delisted from exchanges in 2024? If you only test survivors, your results are systematically too optimistic.
Mistake 3: Unrealistic Slippage
In a backtest, you buy at exactly the displayed price. In reality, there's slippage — especially with illiquid pairs or large orders. At BotTrade.app, standard spot fees (0.1%) are included in backtests. Slippage is not included in default results — use our fee simulator for a more realistic picture.
Mistake 4: Ignored Fees
Trading fees of 0.1% per trade sound like nothing. With 500 trades per month, that adds up to 100% of your capital per year. Every serious backtest must account for fees.
Mistake 5: Testing Period Too Short
A backtest over 2 weeks during a bull market tells you nothing. Test over at least 6 months, ideally across different market phases (bull, bear, sideways).
How BotTrade.app Gets It Right
Our backtesting engine accounts for standard spot trading fees (0.1% per trade, standard spot), various time periods (1, 3, 6, 12 months), and shows you not just PnL but also Sharpe Ratio, Max Drawdown, and Win Rate. For slippage and funding rates, use the integrated fee simulator.
Test your strategies with professional backtesting — free on BotTrade.app.