Definition
Slippage is the difference between the price at which a trader expected an order to execute and the price at which it actually filled. It has two components. The first is the bid-ask spread crossed by an aggressive order — the half-spread cost of demanding immediacy. The second is market impact, the price drift caused by the order itself consuming available liquidity at successive levels of the order book.
In backtesting, slippage is the variable that quietly converts profitable simulations into losing live deployments. A strategy that books a four-basis-point edge per trade on paper will not survive a six-basis-point round-trip cost, no matter how tidy the equity curve looks.
Why it matters
How it works
Realistic slippage modelling requires more than subtracting a flat number of basis points. The cost depends on order size relative to typical volume at the chosen venue, on the time of day (open and close are wider and more volatile), on instrument liquidity tier (a large-cap stock at midday is cheap; a small-cap at the open is expensive), and on whether the order is marketable or rests passively. Passive orders avoid spread cost but accept adverse-selection risk — they get filled when the market is moving against them and skipped when it moves in their favour.
The practical defence is twofold. During backtesting, model slippage as a function of order size and instrument, not a constant. During live deployment, measure realised slippage per trade, compare to the model, and feed the residuals back into the next iteration. A strategy whose realised slippage steadily exceeds the model has either grown beyond the liquidity it was designed for or is signalling at the same moments other participants are, and the edge is being competed away at the touch.