Market Pair Ranking by Liquidity
At CoinMarketCap, one of our main objectives of listing the range of exchanges in the space is to allow our users to make more informed decisions in selecting exchanges to trade on.
To help our users identify the most liquid market for every single cryptoasset listed on our website, we have designed a “Liquidity” metric to accurately reflect the difference in liquidity between exchanges and market pairs.
What is Liquidity?
Liquidity refers to the ease of being able to trade in and out of an asset, or in our case, cryptoassets. A liquid market is one where the price of buying and selling is very similar to each other. This allows investors and traders to trade in and out of the cryptoassets cheaply, as traders incur less slippage on price when trading on more liquid markets.
In traditional markets with regulatory oversight, volume is an effective metric to determine liquidity. However, this is not the case for the cryptoasset market because not every market is subject to the same regulatory environment. Some cryptoasset exchanges have resorted to inflating volumes (i.e. trading against yourself) in order to create the impression of trading activity on their exchange, since 2017.
Thus, we have decided to rethink the use of volume as the default metric for ranking market pairs and exchanges. Instead, we introduce a new metric to track and measure what matters most to traders - Liquidity.
Liquidity by CoinMarketCap
Liquidity by CoinMarketCap works by taking a few variables into consideration:
- The distance of the order from the mid-price
- The size of the order
- The relative liquidity of the asset in question
In measuring liquidity effectively, it is important to consider the distance of the order from the mid-price (calculated by [best bid + best ask] / 2 ). The further the order from the mid-price, the less weight should be given to the order as it becomes less relevant for traders. For example, an order placed at 0.5% from the mid-price would be given a heavier weight than an order that is placed >1.5% from the mid-price. This is because the resting order nearer the mid-price has a higher probability of getting filled, thus needs to be accorded more weight than the order further away.
The size of the order is important as liquidity seeks to measure the depth of the order book. The larger the orders on the order book, the higher the Liquidity score of the market.
In addition, the Liquidity metric has been designed to measure the liquidity of the market in an adaptive manner. Liquidity between different markets and cryptoassets can differ a lot - the metric takes this into account and the algorithm adapts to this difference to return the fairest result possible. This means that the more liquid the market pair is, the more aggressively orders get discounted as they move further from mid-price. This is in line with the idea that we want to accord more weight to orders that have a higher probability of getting filled as they represent liquidity that matters to traders.
The end result of adopting such a methodology is the ability to return a singular number that reflects the effective liquidity of any market pair. The weight given to the orders varies depending on the absolute order-book depth of the market pair in question and the distance it rests from the mid-price. This allows us to fairly compare the liquidity between all markets we track, regardless if it is a liquid asset like Bitcoin or the 1000th ranked coin/token. We refrained from using a static % depth as we recognized the difference in absolute liquidity between the various cryptoassets.
The adaptive methodology applied makes our metric difficult to “game” as orders would need to be placed very close to the mid-price or risk being counterproductive to the Liquidity score we return. We hope this results in more liquid markets for the entire cryptoasset space as exchanges and market makers prioritize posting effective liquidity instead of increasing volumes.
Lastly, the reported Liquidity is calculated by polling the market-pair at random intervals over a 24-hour period and averaging the result. Orderbook depth of any given market changes constantly due to immediate market conditions. In order to make Liquidity fair with respect to differences in global time zones, a rolling average over a 24-hour period is implemented.
How do I interpret the Liquidity metric?
The algorithm returns a number that is indicative of the average liquidity of any given market pair over a 24 hour period. This number can be used to compare liquidity between different market pairs, regardless of Base or Quote asset. The higher the result, the better the liquidity of the market pair.
Our Liquidity metric is more than a mere summation of order book depth. As we realize that working orders only matters insofar as the probability that they get filled, orders further away from the mid-price of any market should be given less weight in determining the liquidity of the market. Thus, we believe that our Liquidity metric is a fairer way of summarising the liquidity of any market pair, regardless of Base or Quote asset, than any other “static polling” of order book depth.
Frequently Asked Questions
Why was Volume used in the first place, and what has changed?
In traditional markets with proper regulatory oversight, volume is an effective metric to determine liquidity. A market with high volumes can usually be assumed to be liquid as many traders congregate there to trade.
As discussed, this assumption has become flawed in the case of cryptoasset market due to the prevalence of “wash trading” since 2017. While the act of “wash trading” is illegal in traditional markets, many cryptoasset exchanges are located and licensed in countries that do not enforce strict regulations over this act. This makes “volume” a problematic metric when used to rank market pairs and exchanges
Thus, we have decided to rethink the use of "volume" as the default metric for ranking. Instead, we have designed a new Liquidity metric that we believe to be superior in its accuracy.
Why is our Liquidity metric better than the oft-used Volume metric?
Our Liquidity metric takes into account a wider range of key variables from the order book, such as the distance of the order from the mid-price, the size of the order and the relative liquidity of the asset in question. The weightage for each variable depends depend on the extent of their roles in determining liquidity.
In addition, the Liquidity metric has been designed to poll the assets in an adaptive manner, rather than to assign a singular depth estimate of 2%. This is important as there are over 3000 cryptoassets listed on CoinMarketCap and the difference in relative liquidity among the assets can vary a lot.
In summation, our Liquidity metric is more than a mere summation of order book depth. We accord heavier weight to orders that have a higher probability of getting filled, by studying the distance of it from the mid-price and observing the absolute liquidity of the market pair in question. This allows our algorithm to be adaptive to the differences in orderbook depth across any market pair.
Is volume still valid as a metric in cryptoasset?
This would depend on the reason you are using volume for, as well as the amount of regulatory oversight that the exchange is subjected to. Volumes are still relevant on exchanges that have strict “anti-wash trading” rules and are subject to strict regulatory oversight. Hence, volumes can still be used on those exchanges to determine the amount of trading interest in the market.
However, we think that using volume as a substitute for liquidity is inefficient and would recommend the use of our Liquidity metric to compare liquidity between market pairs.