Evidence before automation
Backtesting vs forward testing
A backtest applies a strategy to historical data. A forward test observes it on data that was not part of the optimization process, either prospectively or through a platform's forward-testing split.
Risk note: Automated trading can lose money. Historical or simulated performance does not guarantee future results.
Why use both?
Backtests are efficient for exploring behavior over long periods. Forward testing adds evidence about how the strategy behaves outside the development sample and, in live/demo observation, under current execution conditions.
Neither removes uncertainty
A strong backtest can fail forward. A good forward period can also be too short or unrepresentative. Evaluate sample size, market regimes, costs and drawdown rather than looking for one magic test.