Rank centralized exchanges for liquidity using the same trading pair, order size, direction, and observation time. Total reported volume alone cannot establish where your order will execute best. Broad market-quality benchmarks can guide research, but a current pair-level order book is needed for an actionable comparison.

A published benchmark provides context, not today's fill price

CoinDesk's benchmark page, accessed on 19 November 2026, labels its displayed dataset May 2026. Among four familiar candidates, its Market Quality column shows Binance 22.6, Coinbase 20.9, Kraken 19.1, and OKX 17.3. Those are selected entries ordered by that column, not a complete global liquidity leaderboard.

The distinction matters: Market Quality combines several inputs, including liquidity-related measures. Its score is not a quantity of immediately available bids or asks, and the page's overall ranking includes other categories. Do not rename either measure a live execution ranking.

Use a benchmark to identify candidates worth checking. For a purchase today, inspect the actual market available to your account. A historical score cannot establish the current spread or depth for a specific altcoin, payment currency, or trade size.

Selected entries from CoinDesk’s May 2026 dataset, sorted by its Market Quality column. This composite measure is not a live pair-level liquidity rank.
Selected candidateMarket Quality score
Binance22.6
Coinbase20.9
Kraken19.1
OKX17.3
Buy order: consume asks from best price outwardPrice P1Quantity Q1Price P2Quantity Q2Price P3Quantity Q3Average fill = total value / total quantitySnapshot estimate; resting orders may change.
Symbolic order-book calculation: consume only the needed ask quantities and divide total value by total quantity. P and Q are variables, not measured market data; the result is a snapshot estimate.

Know what each liquidity measure can tell you

The spread is the gap between the best available bid and ask. A narrow spread can indicate a relatively small immediate crossing cost at the top of the book, but says little about how much volume is available behind that quote.

Depth describes available orders across price levels. Specify the price range or intended size when discussing it. A large quantity far from the current price may not help a modest purchase near the best ask.

Volume describes trading activity over a period. It does not by itself show the orders available now, and an exchange-wide figure can hide important differences between pairs. Treat volume as a separate observation rather than a substitute for depth.

For the decision at hand, the useful measure is often estimated average execution price for your order, followed by trading fees and other route costs. This converts several market observations into an outcome you can understand.

Compare the same pair, not similar-looking symbols

Begin with the exact base and quote asset. A bitcoin market quoted in one stablecoin is not identical to a bitcoin market quoted in dollars or another token. Quote conversion, issuer exposure, and funding access can affect the practical result.

Also distinguish spot from derivatives. Perpetual-futures liquidity does not establish the spot market's ability to fill your purchase. A large derivatives volume should not be used to rank a cash-funded spot route.

For each candidate, record the market identifier, buy or sell direction, intended input or output quantity, and timestamp. If the account cannot access that market or transfer the required funds, mark it unavailable rather than assigning an artificially poor liquidity score.

Kraken's public endpoint documentation illustrates where order-book data can be obtained. Other providers expose market data through their own interfaces or APIs. Use the correct documentation and preserve the scope of the data actually collected.

Estimate what the order would consume

  1. For a buy, inspect asks; for a sell, inspect bids. Work outward from the best executable price in the correct direction. A depth chart without side labels is not enough.
  2. Take only the quantity needed from each price level until the intended order amount is covered. If the available snapshot is insufficient, report insufficient observed depth rather than assuming the remaining quantity fills at the last displayed price.
  3. Calculate the value at each consumed level, add those values, and divide by the total base-asset quantity. This gives an estimated volume-weighted execution price for that snapshot.
  4. Compare the estimate with a clearly stated reference price. Label whether the comparison uses the best quote or midpoint; changing the reference changes the measured difference.
  5. Add the applicable trading fee and identify restrictions or missing costs separately. The snapshot estimates market execution; it does not prove an order was submitted or filled.

Account for the order book changing

Displayed liquidity can change between observation and order arrival. Orders can be canceled, consumed, or replaced. A calculation from a snapshot is therefore an estimate, not a promise that the same quantity will remain available.

Record how fresh the data is and whether the snapshot covers enough levels. A delayed page, truncated API response, or out-of-sync local book can make a precise calculation misleading. If you cannot validate the data, describe the limitation instead of publishing decimal-heavy certainty.

A limit order can constrain price but may not fill immediately or completely. A market order prioritizes execution under its rules but can consume several levels. Neither order type makes a thin market deep.

For a small beginner trade, a top-of-book difference may be more relevant than distant depth. For a larger order, deeper levels can dominate. That is why an exchange's rank can change with the task.

Turn observations into a limited, useful ranking

Build a comparison with candidate, market, observation time, intended size, spread, estimated execution, trading fee, and missing information. Order only the available candidates using the same stated metric.

If one candidate has an excellent execution estimate but unsuitable funding or withdrawal conditions, keep those facts visible. Liquidity ranking and overall exchange choice are related decisions, not the same decision. A trade cannot start with funds you cannot deposit or finish through a route you cannot use.

Repeat observations if the decision requires more than a single instant. Describe the sampling approach honestly; a few observations do not establish typical behavior across every market condition. Avoid using one quiet-period snapshot to claim resilience during stress.

The result should state its boundary clearly: best estimated execution among the observed candidates for this pair, size, and time. That is narrower than a universal winner, but it gives a beginner evidence they can actually use.

Separate buying liquidity from selling liquidity in the final record. A market can have substantial asks near the current price while offering less bid depth for the quantity you later want to sell. If your decision concerns a round trip, calculate each direction independently using the relevant observation. Do not assume the same ranking survives an opposite trade or a larger amount. Also record the quote currency you can actually use; converting first through another market adds an execution step that belongs in the complete route, even if the final asset pair itself appears liquid.

Reader questions

Does a large market-cap token automatically have a deep market on every exchange?

No. Liquidity is specific to the venue and pair. A widely followed asset can still have a thin local market or an inconvenient quote currency.

Can a chart's trading volume show where all current limit orders sit?

No. Historical traded activity and resting orders are different data. Use an order-book or depth source for the latter.

Should I place an order just to measure whether the book is real?

Do not submit unwanted orders as a research shortcut. Public market data and genuine quote previews can support a limited estimate without creating an unintended trade commitment.

Sources and further reading