Trading & AMMs

Order Book

Real-time list of buy and sell orders organized by price levels

Definition

An order book is the list of resting limit orders for a market, sorted by price: bids below, asks above, with the gap between the best of each forming the spread. A market order consumes the book from the best price outward, which is why a large order executes at a worse average price than the top of book suggests. Order books give makers precise control over price, size and cancellation, and let a professional market maker quote both sides with far less capital than an AMM needs for the same depth. The trade-off is that they require constant active participation and a cheap, fast venue to quote and cancel on, which is why almost all crypto order books live on centralised exchanges or on purpose-built app-chains such as Hyperliquid and dYdX rather than on Ethereum mainnet.

An order book is the live list of who wants to buy and sell at what price. Your market order eats through it from the best price outward, so big orders get worse average prices.

Example

An ETH book shows a best bid of $2,999.50 for 40 ETH and a best ask of $3,000.50 for 25 ETH: a $1 spread, about 3.3 basis points. A 100 ETH market buy takes the 25 at $3,000.50, then the next level of 30 at $3,002, then 45 at $3,004.50, for an average fill around $3,003 rather than the $3,000.50 shown at the top. That $2.50 per ETH is the cost of size, and it is the order book's equivalent of an AMM's price impact. The same trade on a thin book with 3 ETH at each level would walk far further.

How it works

Limit orders rest at chosen prices. The best bid and best ask define the spread. Market orders consume resting liquidity level by level until they are filled.

Why it matters

It is still where most crypto volume trades, and it is the model against which every AMM design is implicitly compared. Reading depth tells you what your exit will actually cost.

What to check

Look at depth within a reasonable band of the mid price, not just the top of book. Watch whether depth survives volatility, check the venue's custody and matching model, and discount suspiciously tight spreads on small exchanges.

Risks to Consider

Displayed depth can be withdrawn in a single block or a single millisecond, so the book you see is a quote, not a commitment
Spoofing and wash trading make thin books look deep and liquid, particularly on smaller centralised venues
Professional market makers have a latency and information advantage that retail order flow simply pays for
Off-chain matching means you trust the operator for custody, for fair ordering and for the book being real at all

Common Questions

Order book or AMM?

Order books win on capital efficiency and on price control: a market maker can quote a tight two-sided market with a fraction of the inventory an AMM pool needs, and a trader can name an exact price. AMMs win on being permissionless and always available: there is no one to stop quoting, and a long-tail token can have a market from day one. In practice liquid majors are better on books and long-tail assets only exist on AMMs.

Why are there so few on-chain order books on Ethereum?

Because market making is a high-frequency activity and mainnet is not. A maker updates quotes constantly as the reference price moves, and each update on Ethereum costs gas and waits for a block. That makes tight quoting economically impossible, so on-chain books migrated to venues with near-zero fees and sub-second blocks. Attempts like Serum on Solana and Hyperliquid's own chain exist precisely because the environment, not the design, was the constraint.

Is the depth I see on the book real?

Partly. Some of it is genuine resting interest and some is quotes that will be pulled the moment the market moves, which is exactly when you need them. A practical check is to watch how the book behaves during a fast move: depth that vanishes on volatility was never liquidity you could use. On smaller venues, also assume some of the visible volume is wash traded and discount accordingly.