Interactive Backtest Visualizer

A backtest is a simulation of how a trading strategy would have performed on historical market data. It is the cheapest and fastest way to build confidence in an idea, but it is also the easiest place to lie to yourself. Learning to read a backtest sceptically is one of the highest-leverage skills a retail algo trader can develop.

Start with the shape of the equity curve. A healthy strategy has visible pullbacks, flat stretches, and occasional sharp drawdowns. A suspiciously straight line going up-and-to-the-right is almost always a sign of curve-fitting — the strategy has been over-optimised to the specific past it was tested on. Strategies that look too clean in backtest very rarely survive contact with live markets.

Next, check the max drawdown. This is the worst peak-to-trough loss the strategy ever experienced in the backtest. If the max drawdown is 15% and your personal loss tolerance is 8%, the strategy is wrong for you regardless of its average return. You will almost certainly switch it off at the worst possible moment, locking in losses that the strategy would have recovered from.

Finally, look at secondary metrics — Sharpe ratio, Sortino ratio, profit factor, and win rate. Sharpe above 1 is a reasonable baseline; above 2 is excellent; anything above 4 on a retail forex backtest should be treated as a red flag for overfitting or data leakage.

Overfitting. Adjusting parameters until the backtest looks great. A strategy with 40 hand-tuned inputs will always look perfect in-sample and collapse out-of-sample. Defend against this with walk-forward validation and a strict out-of-sample hold-out period.

Survivorship bias. Testing only on assets or bots that still exist today ignores every strategy that blew up. In forex this is less severe than in equities, but relevant when evaluating bot vendor track records.

Look-ahead bias. Using information the strategy could not actually have known at the time — e.g. using the day’s closing price as an input to trades taken during the same day. This is the single most common source of inflated backtests.

Ignoring transaction costs. A high-frequency strategy with 5,000 trades per month dies instantly once realistic spread and commission are applied. Always subtract at least 0.5 pip of round-trip cost per trade on majors.

Slippage assumptions. Retail brokers fill market orders at worse prices than the chart implies, especially during news events. A backtest that assumes perfect fills will always overstate returns.