Every bar has a high and a low, but Pine Script has no idea which came first. If price hit your stop and your target in the same bar, it guesses — and it tends to guess in your favour. Every strategy that's ever gone against you before recovering looks better than it is because of this. There is no fix. It's a fundamental limit of bar data.

By default, Pine Script signals on bar close and fills at the next bar's open. That open existed for a fraction of a second. You'd never catch it in real trading. Strategies that rely on opening-range entries or fast reactions look cleaner in backtests than they'll ever be live.

Pine Script recalculates the current bar in real time as price moves. But in the backtest, every bar is already closed and final. A strategy that reacts to a bar "closing above a level" behaves completely differently live — because live, that bar keeps changing until it actually closes. What you see in the backtest isn't what the strategy experienced.

In Pine Script, if your stop is at 19,850 you get filled at 19,850. In real futures markets, if price drops through 19,850 in a fast move, you get filled wherever the market is when your order hits — which could be 19,843. On a strategy with a tight stop, that gap alone can flip a winner to a loser. Pine Script doesn't model this at all.