Many strategy tests treat transaction costs as a small adjustment. In practice, costs are part of the edge model. A strategy with a thin expected value can become weak or negative after spread expansion, commission, slippage or delayed execution are included.
A positive gross expectancy is not enough. The strategy must remain positive after realistic execution costs and stressed cost assumptions.
Expectancy is a small number
For many automated systems, the average edge per trade is much smaller than the largest winning trade shown in a report. That average edge can be reduced by only a few points of extra spread or slippage. This is especially important for systems that trade frequently, use small targets or operate during sessions where liquidity conditions change.
The right question is not only whether the strategy was profitable in the model. The question is whether the remaining edge is large enough after realistic friction.
Spread expansion changes the break-even threshold
Wider spread increases the cost of entering and exiting positions. When the average win is not much larger than the average loss, this can shift the break-even win rate meaningfully. A strategy that looks stable at one spread assumption can become fragile when the spread is stressed.
This is why execution cost stress should be evaluated before increasing exposure. Cost sensitivity is not an afterthought; it is part of the validation layer.
Calculate expectancy at baseline costs and again under stressed spread, slippage and commission. If the result is only marginally positive, the strategy should not be treated as robust.
Slippage has asymmetric impact
Slippage often appears only during the conditions that already stress the strategy: fast movement, poor liquidity, news windows or abrupt volatility expansion. That means its impact may not be evenly distributed across trades. It can cluster around losses or around exits that matter most.
A simple average slippage value is useful, but the more important review is whether the strategy still works when slippage appears at the wrong time.
Execution delay can change the trade
Delay is not only a technical inconvenience. For short-horizon or breakout systems, delayed execution can produce a different entry price, a different stop distance and a different realized payoff. The trade captured by the backtest may not be the trade executed in the terminal.
A dashboard or report should therefore distinguish signal quality from execution quality. A correct signal can still produce a poor result when execution deteriorates.
What to stress
- Spread: baseline, session average and stress-window expansion.
- Commission: round-turn cost per trade and account-type differences.
- Slippage: average and adverse-case assumptions.
- Execution delay: impact on entry price, stop distance and target distance.
- Trade frequency: whether costs accumulate faster than gross edge.
- Break-even win rate: how much the threshold moves under stress.
Conclusion
Execution costs do not merely reduce profit. They can change the classification of a strategy. A candidate with positive gross expectancy but weak cost tolerance should remain under review until execution stress confirms that the edge is not dependent on ideal assumptions.