The cost everyone checks after the trade
The figures come from a single order-book snapshot taken when the page loaded, and they assume every resting order stays put while your order consumes it. Neither holds in practice. Market makers pull quotes when they see size arriving, so a large order often fills worse than this table suggests — the numbers here are closer to a floor than a forecast.
The book also cannot show hidden and iceberg liquidity, which works the other way and can make large fills better than modelled. Both effects grow with order size, so treat the small rows as reliable and the large ones as indicative. Splitting an order across time is the standard response to everything on this page.
Slippage is measured against the best available price rather than the mid, so it excludes the spread you cross when taking. Add the spread figure above to get the full round-trip cost of entering and exiting immediately, and add the taker fee twice on top.
Exchange fees are advertised, compared and argued about to the second decimal place. Slippage is neither advertised nor easy to look up, and on anything other than the deepest pairs it is the larger of the two by a wide margin. A market order does not execute at the price on the screen; it consumes resting orders one level at a time, and the average of those levels is what you actually pay.
The relationship is not linear either. Doubling the order size more than doubles the slippage, because the book thins out as you move away from the mid. That is why the same trade that costs a basis point at retail size can cost fifty at institutional size, and why the useful question is never "what is the spread" but "how much money is resting within one percent of here".
It is also charged twice. Every position pays it entering and pays it again leaving, so a round trip on a thin pair can start several percent underwater before the market has done anything at all. For short-horizon strategies this is frequently the difference between a backtest that works and a live account that does not.
⚠️ Not investment advice. Figures come from a single live order-book snapshot and assume resting orders do not move, which is not how markets behave under size. Treat them as a lower bound on real execution cost. All decisions and risks are your own.
It is the gap between the price on the screen and the price you actually get. A market order consumes resting orders one level at a time, so the average of those levels — not the best quote — is what you pay. This page walks the live order book to compute it for a range of order sizes.
On thin pairs, by a wide margin. A typical taker fee is around 10 basis points, and the table shows where slippage overtakes that for the pair you selected. People compare exchange fee schedules closely and almost never check this, even though it is charged on both entry and exit.
Because the order book thins out as you move away from the mid. The first levels hold the most liquidity, so each additional unit of size reaches further into progressively emptier price levels. Doubling the order more than doubles the cost.
The depth table sums resting orders within bands of the mid — 0.1%, 0.5%, 1% and 2%. That figure is far more useful than the number of price levels, because it tells you directly how much size the market absorbs before the price moves.
Not reliably. Resting orders are cancellable and are frequently cancelled the moment someone tries to trade through them. Read the imbalance as which side is cheaper to push, not as a prediction of direction.
They are closer to a floor. The snapshot assumes every resting order stays put while your order consumes it, and market makers pull quotes when they see size arriving. Hidden and iceberg liquidity works the other way, so the small rows are reliable and the large ones indicative.
Trade deeper pairs, use limit orders rather than market orders, and split large orders across time. The depth table shows exactly which size your chosen pair starts to struggle with, which is the number to split below.