What Is Slippage in Trading? The Hidden Cost That Kills Backtests

Slippage is the difference between the price you expected to trade at and the price you actually got. You click to buy a stock at $50.00, the order fills at $50.04, and that four-cent gap is slippage. It happens because prices move in the moment between your decision and your fill, and because the size of your order can push the price against you. Slippage is a real, recurring cost, and leaving it out of a backtest is one of the fastest ways to believe in a strategy that does not exist.

Most new traders track commissions to the penny and never measure slippage at all, which is backward. Commissions on stocks and ETFs are often zero now. Slippage never is, and on an active strategy it usually dwarfs whatever commission remains.

What causes slippage?

Slippage comes from three forces, and knowing which one is hitting you tells you how to reduce it.

• The bid-ask spread. Every stock has a price to buy (the ask) and a lower price to sell (the bid). A market order crosses that gap immediately, so you pay the spread the moment you trade. On a liquid name like SPY the spread is a penny; on a thin small-cap it can be many cents or more.
• Price movement (latency). Between the instant you decide and the instant your order reaches the market, the price keeps moving. In fast or volatile conditions, especially around news or the open, the quote you saw is already gone.
• Market impact. A large order eats through the available shares at the best price and reaches for worse prices to complete. The bigger your size relative to the stock's normal volume, the more your own order moves the price against you.

Slippage vs spread: what is the difference?

The spread is one component of slippage, not a synonym for it. The bid-ask spread is the fixed gap between the buy and sell price at a single moment, and crossing it with a market order is a guaranteed cost you can see before you trade. Slippage is the broader gap between your expected price and your realized price, which includes the spread plus any price movement and market impact that occur while the order is being filled. On a liquid stock traded in small size, most of your slippage is just the spread; on a thin stock or a large order, movement and impact can be far larger than the spread alone.

How much slippage should you assume?

For a liquid US large-cap or a major ETF traded in retail size, a common working assumption is 0.05% to 0.1% per trade, which covers the spread and a little movement. For less liquid names, faster strategies, or larger size, it climbs quickly. The table below gives realistic starting points, but the honest answer is that you should measure your own fills rather than trust a default forever.

Past performance does not guarantee future results. For educational and informational purposes only. Not financial advice. Consult a licensed advisor.

Why slippage in backtesting matters so much

A backtest that ignores slippage tests a strategy nobody can actually trade. The problem compounds with trading frequency: a rule that turns over 200 times has 200 chances to pay slippage, so even a modest 0.1% per trade adds up to roughly 18 to 20% of cumulative drag over the test. Fast strategies are hit hardest of all. A high-frequency mean-reversion rule can look brilliant at zero cost and lose money the moment realistic slippage is charged, because the edge per trade was always smaller than the cost per trade.

This is why a paper edge and a live edge so often diverge. The backtest was run frictionless, the trader saw a beautiful equity curve, and then real fills quietly bled the account. The fix is not complicated: charge realistic costs in the test, and be especially skeptical of any strategy whose profit depends on trading often. If the edge cannot survive a fair slippage assumption, it is not an edge. The broader list of ways a test flatters itself is covered in common backtesting mistakes, and slippage sits near the top of it.

How to reduce slippage

You cannot eliminate slippage, but you can control most of it with a few habits.

1. Use limit orders where you can. A limit order caps the price you will accept, so it removes the movement and impact components at the cost of occasionally not filling. For anything less than highly liquid, a limit order is the single biggest lever.
2. Trade liquid instruments. Spreads on major ETFs and large-caps are pennies. The same strategy on a thin small-cap can lose its entire edge to the spread alone.
3. Avoid the first and last minutes. The open and close are the most volatile windows of the day, with the widest spreads. Unless your strategy specifically needs them, waiting a few minutes reduces slippage meaningfully.
4. Size relative to volume. If your order is a large fraction of a stock's average volume, you are the market impact. Scaling in, or simply trading a more liquid name, keeps your own footprint small.
5. Measure your realized slippage. Compare your intended price to your fill on every trade and keep the running average. Once you know your real number, you can plug it into your backtests instead of guessing.

Where slippage fits in the bigger picture

Slippage is one input into whether a strategy actually makes money, and it interacts with the others. A strategy with a large edge per trade, measured as expectancy, can absorb meaningful slippage and stay profitable; a marginal one cannot. That is why costs and expectancy have to be read together, and why what is expectancy in trading is the natural companion to this page: expectancy tells you how much room you have, and slippage tells you how much of it the market takes back.

The practical move is to stop treating costs as an afterthought and bake them into every test from the start. Agenttrading does this by default: type a thesis in plain English, and it backtests the rule on 20+ years of split- and dividend-adjusted daily data with a 0.1% cost per trade assumed by default, then stamps an honest verdict, HELD UP, MIXED, or UNDERPERFORMED, against simply holding. Pro users can raise the cost and slippage assumption to match their own instruments and size. You can see the mechanics on backtesting software, or drop a rule into the trading strategy tester and watch a realistic cost assumption change the verdict in real time. A strategy that only survives at zero slippage is telling you something, and it is better to hear it on the bench than on your statement.