What is the bid-ask spread?

The bid-ask spread is the difference between the lowest ask and highest bid in the same market at the same time. In an order-book-based crypto market, it describes the distance between the closest buy and sell quotes.

How is the bid-ask spread calculated?

The order book lists unfilled buy and sell quotes. Buy quotes are bids, with the highest called the best bid. Sell quotes are asks, with the lowest called the best ask.

Bid-ask spread = best ask − best bid

For example, suppose a spot pair’s highest bid is 99.9 USDT and lowest ask is 100.1 USDT. The spread is 100.1 − 99.9 = 0.2 USDT. USDT is the quote asset here, and the numbers are illustrative.

To express it as a percentage, divide the spread by the midpoint, the average of best bid and best ask. Here, the midpoint is (99.9 + 100.1) ÷ 2 = 100, and the relative spread is 0.2 ÷ 100 × 100% = 0.2%.

It can also be expressed in basis points (bp, plural bps): 1 basis point = 0.01%, so 0.2% = 20 basis points here. Percentages and basis points express relative size and are more meaningful than directly comparing absolute differences across assets with different prices. The denominator must still be consistent.

Why is there a gap between bid and ask?

The prices buyers will pay and sellers will accept do not always agree. During continuous order matching, orders with suitable prices and conditions execute. Unmatched quotes remain in the book, potentially leaving a spread between the closest prices on each side.

In the example, buyers offer at most 99.9 and sellers ask at least 100.1, so these quotes cannot yet trade. If a later buy accepts the ask of 100.1, it can match that sell order without requiring the original bidder to change their quote at the same time.

New orders, cancellations and executions continually change available quotes, so a pair’s spread changes over time.

What does a narrower or wider spread mean for execution?

A narrow spread means the quotes for immediate buying and selling are closer; a wide spread means they are farther apart. For the same pair, with sufficient quantity and unchanged prices, a wider spread means a larger price loss from buying immediately and then selling immediately.

With the same 99.9/100.1 assumptions, sufficient quantity at both best quotes, unchanged prices and no fees, buying one unit at 100.1 and selling it at 99.9 produces a difference of 0.2 USDT: one spread. This is the outcome of a buy followed by a sell, not an extra 0.2 USDT charged by the platform.

Quantity at the best quotes matters too. Order-book depth describes how much buy and sell quantity is available at each price. If only 1 token is offered at the lowest ask, buying more may require higher price levels. If nearby ask levels contain larger quantities, an order of the same size is more likely to complete near that price.

The spread therefore measures the distance between buy and sell quotes, while depth measures how much trading can be accommodated near those prices. Close quotes do not establish sufficient executable quantity.

How is the spread related to slippage?

Both depend on available quotes, but the spread cannot be directly converted into slippage. The spread is visible before execution as the difference between quotes. Slippage compares an actual execution price with a reference price.

Again suppose the best bid is 99.9 and best ask 100.1 USDT, a spread of 0.2. Using the ask of 100.1 as the buy reference, consider two independent hypothetical orders of different sizes starting from the same quotes. Assume unchanged quotes and completed orders:

• Small order: the quantity does not exceed that available at the best ask, so all executes at 100.1. There is no slippage relative to the reference, but the spread remains 0.2.
• Larger order: quantity at the best ask is insufficient, so higher levels are used. If the average price is 100.3, adverse slippage is 0.2 USDT per unit relative to the reference.

A narrow spread therefore does not guarantee low slippage: order quantity and depth matter. Likewise, a wide spread does not mean a small order must deviate from the pre-submission quote. If quotes change before execution, even a small order may have slippage.

What spread is reasonable?

A fixed percentage cannot judge every crypto market without considering the pair and market conditions. Relative spreads and basis points are ways to compare numbers, not industry ratings with a universal pass threshold.

Comparisons can first examine relative spreads for the same pair at similar times and under similar conditions, together with depth near the best quotes. Across venues, the product, quote asset and sampling time must also align. A single snapshot does not represent performance over a period.

The example’s 20 basis points only means the quotes differ by 0.2% of the midpoint. It alone cannot establish low trading costs or sufficient depth. To understand the price achievable for a particular amount, the expected or actual average price for that quantity is also needed, not just the spread.

How do spread, slippage and fees affect costs?

Spread and slippage are reflected in execution prices; trading fees are charged under the applicable rules. Platforms calculate fees from actual executions and rates, rather than separately deducting spread or slippage. An execution exactly at the reference with no slippage may still incur fees; zero fees do not ensure the execution price meets expectations.

The three numbers cannot simply be added. In the earlier buy example, an average of 100.3 versus an ask reference of 100.1 gives slippage of 0.2 per unit. Against the midpoint of 100, the deviation is 0.3, already including the 0.1 distance from midpoint to ask.

Different references include different components. Establish whether the comparison concerns one buy or a complete buy-and-sell sequence, and whether quotes already include fees, to avoid double counting.

Related terms

Explore limit orders, market orders and takers, or return to the crypto glossary.

References

• OKX: Trading fee rules FAQ
• OKX: Basic order types