Backtesting is the practice of applying a trading strategy’s rules to historical price data to estimate how it would have performed in the past. It is the cheapest way to find out whether an idea has an edge: a strategy that loses on history has no claim on the future, while one that wins on history has at least earned a forward test.
A backtest is only as honest as its assumptions. Realistic results require including commissions and slippage, avoiding look-ahead bias (using information that was not available at the time), and testing enough trades across enough market conditions to mean something. Costs are routinely the difference between a paper edge and a real loss — Secuora’s published research at /strategy ran twelve months of popular baseline strategies on BTC and ETH, and every one of them lost money once fees were applied.
Backtesting comes in two main forms: automated (code executes the rules) and manual bar replay (the trader executes them candle by candle), which also trains execution.