What is Slippage? Definition, Formula, and Example
Slippage is the dollar or percentage difference between the price a trader expects an order to fill at and the price it actually fills at, driven by the bid-ask spread, order size relative to available liquidity, and how much the market moves between order submission and execution.
What is slippage?
Slippage is the gap between the price you expected to pay (or receive) when you clicked "submit" and the price your order actually filled at. It happens because a quoted price is never a guarantee — it's a snapshot of the best bid or ask at that instant, and by the time your order reaches the exchange, that quote can have moved, been partially filled by someone else, or simply not existed for the full size you wanted. Slippage isn't a fee or a broker markup; it's a structural feature of trading in a market where prices update continuously and liquidity at any single price level is finite.
How slippage is calculated
Slippage = fill price − expected price, expressed in dollars or as a percentage of the expected price. The sign convention flips by side: for a buy order, a fill price *above* the expected price is negative slippage (you paid more); for a sell order, a fill price *below* expected is negative slippage (you received less). The reverse — filling better than expected — is called positive slippage or price improvement. Two structural drivers set the size of slippage: the bid-ask spread (the minimum slippage a market order can experience is roughly half the spread, since a market buy crosses the spread to hit the ask) and market impact (an order larger than what's resting at the best price has to "walk the book," filling progressively worse prices for the remaining size).
Worked example
AAPL trades with a bid-ask spread of roughly one cent on a $308.63 stock — a market order for 100 shares fills within a fraction of a percent of the last trade, because depth at the top of book is deep relative to typical retail order size. Compare that to a thin micro-cap quoted $2.00 bid / $2.20 ask — a 10% spread. A market buy there fills near $2.20, and if the order is larger than what's resting at $2.20, the next shares fill at $2.25, $2.30, and so on. GameStop's late-January 2021 squeeze is the extreme, real-world version of this: with the stock halted repeatedly intraday and reopening auctions resetting the book each time, market orders placed just before a halt could fill tens of dollars away from the last printed trade — slippage measured in double-digit percentages rather than fractions of a cent.
When traders use it
Slippage is why institutional and algorithmic traders route large orders through execution algorithms — TWAP and VWAP-tracking algos slice a big order into smaller pieces over time specifically to minimize market impact and hold average fill price close to the benchmark. Discretionary traders manage it by using limit orders instead of market orders in illiquid names, checking Level 2 depth before sizing an order, and avoiding market orders in the first and last minutes of the session or immediately after news, when spreads widen and depth thins. Backtested strategies that ignore slippage systematically overstate real-world returns, especially for strategies trading small-cap or low-volume names.
Limitations and misconceptions
Slippage is not inherently bad — positive slippage happens as often as negative slippage in liquid names, and over a large sample the two roughly offset unless an order is consistently large relative to available size. It's also not the same thing as a bid-ask spread cost, a commission, or a stop order gapping past its trigger price (that's a related but distinct risk on stop-loss orders), though all three degrade the difference between theoretical and realized returns. Slippage estimates from a backtest are also only as good as the liquidity assumptions built into the simulation — a backtest that fills every order at the last traded price, with no size or spread modeling, will systematically understate real slippage on anything but the most liquid names.