What a backtest can tell you
Backtests are useful for testing rules across historical market regimes, estimating drawdown and finding obvious weaknesses. They are research tools, not proof of future profitability.
The overfitting problem
A strategy can be tuned so closely to historical data that it captures noise rather than a durable trading edge. The more parameters are adjusted to improve past results, the more important out-of-sample testing becomes.
Why live results matter
Live trading introduces spreads, slippage, latency, liquidity and operational behaviour that a simulation may model imperfectly. A credible evaluation therefore gives more weight to a sufficiently long live record than to an attractive equity curve from a backtest.
Use an evidence ladder
Marketing screenshots sit near the bottom. Detailed backtests are more useful. Forward tests add another layer. A verifiable live account with meaningful duration and trade count is stronger evidence, while still never guaranteeing future results.