Most backtesting software will happily show you a beautiful equity curve that would never have existed in real life. The best futures backtesting software isn't the one with the nicest charts — it's the one that tries hardest to disprove your strategy before the market does.
The features that actually matter
- Zero lookahead / no repainting — the backtest must never use data it couldn't have known at the moment of the trade. This one flaw turns flat strategies into rockets and is the number-one reason backtests lie.
- Realistic fills and costs — next-bar fills, commissions, and slippage modelled. A curve that ignores costs is fiction.
- Walk-forward analysis — optimise on one slice, test on the next, roll forward. Catches curve-fitting a single in-sample test can't.
- Monte-Carlo — reshuffle the trade sequence into thousands of alternate histories to separate skill from luck.
- An untouched holdout — data the optimiser never saw, replayed once as an honest final exam.
- Overfitting checks — deflated Sharpe and probability-of-backtest-overfitting gates that flag a result too good to be true.
Pretty curves are the trap
If a tool makes it easy to tweak parameters until the curve looks perfect but hard to test whether that curve survives out-of-sample, it's optimising for the wrong thing. You want software that makes overfitting obvious, not invisible.
What good looks like
TapeScript was built around this idea: every strategy is backtested with zero lookahead on real futures history, then put through 30+ research tracks — sweeps, walk-forward, Monte-Carlo, session and regime analysis, and an untouched holdout — and graded A–F. It also simulates your strategy against 10 prop firms. The goal is a number you can trust, not a screenshot you can't. See it work →