What is slippage in trading?
Slippage is the gap between the price you expected and the price your order actually filled at. It appears whenever you demand immediacy — market orders, stop-losses triggering, fast markets — and it is largest exactly when you most need it to be small: during crashes and in thin altcoins.
Why does slippage happen at all?
An order book only holds so much size at each price. A market buy for more than the size resting at the best ask absorbs the next, worse-priced level, then the next, until it is filled — so the average price you paid lands above the quote you clicked. Nobody charged you for this. You asked to be filled immediately, and immediacy is bought by paying whatever price is standing there.
Which means slippage is not a property of the market. It is a property of your order measured against the depth underneath it. The same $10,000 is a rounding error in BTC and a visible dent in a thin altcoin.
How much slippage is normal for my order size?
There is no single honest number, only a calculation. Here it is on a book we state in full, so you can redo it with your own pair.
The book we are assuming: a mid-cap altcoin where about $8,000 of sell orders rest at each 0.05% price step above the best ask. That ladder is an assumption we are declaring, not a measurement of any live market — the point is the shape of the answer. Open the book on your own pair and substitute your figures.
| Market buy | Levels it eats | Average fill above quote | Slippage cost |
|---|---|---|---|
| $1,000 | fits inside the first level | 0.00% | $0 |
| $10,000 | 1.25 levels | 0.01% | $1 |
| $100,000 | 12.5 levels | 0.29% | $288 |
Working for the last row: $8,000 × 12 levels = $96,000 filled at steps 0.00% through 0.55%, averaging 0.275%; the final $4,000 fills at 0.60%. ($96,000 × 0.275% + $4,000 × 0.60%) ÷ $100,000 = 0.288%, or $288.
The line worth remembering: the order got 10× bigger and the cost got 288× bigger. The reason is that the slippage percentage rises roughly in step with your size, and the dollar bill is that percentage multiplied by the size again — so the cost grows with the square of your order. Doubling your position does not double what slippage takes from you. It roughly quadruples it.
Now change one input only — the depth — and keep the order identical. On a book like BTC/USDT, where a single 0.01% step can hold several hundred thousand dollars, that same $100,000 never leaves the first level: 0.00%, $0. Same trader, same size, same urgency, and the cost falls from $288 to nothing. "Trade liquid pairs" is not a slogan. It is the largest lever you have over this number, and it costs you nothing to pull.
When does a stop-loss get slipped?
Whenever it fires. A stop-loss is not an order sitting in the book — it is an instruction that creates a market order the moment price touches your trigger. So a stop is a market order scheduled to arrive at the exact moment the book is thinnest, because the same move that triggered you triggered everyone else at that level too. In a liquidation cascade, the bids near your stop have already been eaten by the people ahead of you.
When is this estimate wrong?
The table above assumes a tidy ladder that stands still while you eat it. Real books break that assumption in four ways, and you should know which way you are wrong before you trust any slippage number:
- Real books are lumpy, not uniform. Depth clusters near the touch and goes hollow further out. Inside the first few levels the table overstates your cost; past them it can badly understate it.
- Depth is a snapshot, and it runs. The model assumes resting orders wait for you. Market makers pull quotes when volatility spikes — so the depth you measured at rest is not the depth you get at the moment you most need it.
- Slicing changes the answer. Break the $100,000 into ten $10,000 orders spaced out, and makers replenish between them. You may pay far closer to 0.01% than 0.29%. You trade slippage for time risk — price can move away while you wait.
- Gaps and halts have no ladder at all. If the venue stops matching, or price jumps a range with no resting orders inside it, there is no arithmetic to do. Your fill is wherever the market reopens.
So use this table to size positions and to sanity-check a backtest — not to promise yourself a fill.
How do traders reduce slippage?
Use limit orders when entry timing isn't critical; trade liquid pairs and liquid hours; keep size small relative to book depth; and budget slippage into your risk math — a plan risking 1% that routinely slips 0.3% is actually a 1.3% plan. Slippage is also the quiet killer of over-backtested strategies: the chart's fills were free, yours aren't.
FAQ
Is slippage a fee the exchange charges? No — it isn't charged by anyone. It is a market cost that comes from consuming order-book depth; the exchange's actual fees come on top of it.
How much slippage is normal? There is no normal — only your size against the book’s depth. Work the ladder above for your own pair before you assume a number.
Can slippage ever help me? Yes, rarely: limit orders in fast markets can fill at better prices than requested ("positive slippage"). Plan around the negative kind anyway.
Slippage is decided by the venue you chose
The same order slips differently depending on the depth resting under it. Choosing a liquid venue is the part of slippage you actually control.
Compare the three exchanges we use →
That page carries referral links and says so up front — including what each venue does badly.
Every key term, one roadmap
The whole slide course — ten free PDF parts, 351 pages.
Slippage is a symptom of order-book depth. The arithmetic of a market order walking four price levels is worked out in how the crypto market actually works. For how to avoid paying it — marketable limits, when a limit order is the wrong tool, and why stop-market beats stop-limit for protective exits — see market, limit and stop orders. And for the step before either of those — measuring how much depth is actually resting under you, and sizing the position so slippage stays small — see liquidity and spread.