Slippage in backtests: the number most people leave at zero
Posted: Thu Oct 29, 2026 7:00 am
Slippage is the gap between the expected price and the actual fill, and in a backtest it is usually set to zero. That makes results look better than real trading can deliver.
For a scalper, slippage can be a major part of the cost. An average of 0.3 pips per side adds 0.6 pips per round trip. On a strategy aiming at three pips per trade, that is twenty percent of the target.
Modelling slippage realistically requires data. Live trading logs provide the best source: the difference between requested and filled prices, split by order type, time of day, and market conditions.
A simple approach is to add a fixed slippage figure in the backtest, based on observed averages, and increase it during news windows and thin hours.
Strategies that survive this treatment have a much better chance of surviving contact with the live market. Those that do not are better discovered early.
How much slippage do you build into your tests, and where did that number come from?
For a scalper, slippage can be a major part of the cost. An average of 0.3 pips per side adds 0.6 pips per round trip. On a strategy aiming at three pips per trade, that is twenty percent of the target.
Modelling slippage realistically requires data. Live trading logs provide the best source: the difference between requested and filled prices, split by order type, time of day, and market conditions.
A simple approach is to add a fixed slippage figure in the backtest, based on observed averages, and increase it during news windows and thin hours.
Strategies that survive this treatment have a much better chance of surviving contact with the live market. Those that do not are better discovered early.
How much slippage do you build into your tests, and where did that number come from?