Backtests ignore slippage unless told otherwise
Posted: Sat Oct 31, 2026 7:32 am
Most testers assume orders fill at the requested price. Real markets do not behave that way, and for a strategy aiming at small targets, slippage is the difference between an edge and a loss.
Adding a fixed slippage penalty, such as 0.2 or 0.3 pips per trade, is a simple adjustment. A more refined approach scales the penalty with volatility or time of day, charging more during news and thin hours.
Sensitivity analysis is useful too. Running the same test with slippage set at zero, 0.2, 0.5, and 1 pip shows how fragile the strategy is. If a small penalty destroys the profit, the edge was thinner than it looked.
Comparing live results with the backtest provides calibration. If live trades average 0.4 pips of slippage, using that figure in future tests makes them more realistic.
Strategies that survive harsh assumptions deserve more confidence than those that only work under ideal conditions.
How is slippage handled in tests, and what penalty does the live record suggest?
Adding a fixed slippage penalty, such as 0.2 or 0.3 pips per trade, is a simple adjustment. A more refined approach scales the penalty with volatility or time of day, charging more during news and thin hours.
Sensitivity analysis is useful too. Running the same test with slippage set at zero, 0.2, 0.5, and 1 pip shows how fragile the strategy is. If a small penalty destroys the profit, the edge was thinner than it looked.
Comparing live results with the backtest provides calibration. If live trades average 0.4 pips of slippage, using that figure in future tests makes them more realistic.
Strategies that survive harsh assumptions deserve more confidence than those that only work under ideal conditions.
How is slippage handled in tests, and what penalty does the live record suggest?