An In-Depth Look at CoinMarketCap’s Newly Improved Liquidity Score for Finding Best Crypto Exchanges

Celý článek

CoinMarketCap first launched the Liquidity Metric in November 2019 to help users find the best crypto exchanges, by calculating and presenting the available liquidity on exchanges. In the past five months, we have received a great deal of feedback about the metric from both the industry and our users. 

Given the ever-changing crypto landscape, we decided that it was necessary to improve and even recreate the way liquidity is showcased in both a more accurate and more user-friendly way.

A few key reasons were: 

a)      We saw how exchange-reported volumes started to be an inferior metric to estimate the liquidity of market pairs due to errant exchanges inflating their volumes to gain visibility;

b)    We wanted to highlight the importance of understanding the need for liquidity in markets;

c)      And lastly, we wanted to report liquidity in a single number so that users could easily compare the liquidity of different market pairs without having to look at exchanges’  individual order books.

Our Initial Solution — Liquidity Metric

Liquidity is a difficult concept to grasp, as it involves understanding several concepts of how markets and order books work. Unlike volumes which can be tracked and reported in a singular number over a predefined time (e.g. one hour, 24-hour, weekly etc.), reporting “liquidity” is much more complicated.

There are a few ways that one can attempt to track liquidity: 

a)      Look at the absolute size of the order book (i.e. track how big the orders placed on the bids and asks are);

b)     Look at the spread of the order book (i.e. how far apart the best bid and ask orders  are);

c)      Track how much of a cryptoasset can be executed  at a predetermined percentage depth (e.g. at a 1% depth from the mid-price);

d)     Track how much slippage would be incurred at a predetermined order size (e.g., for a $10,000 order, what is the difference between the current price and the executed price) .

In our initial solution, we decided to take factors (a) and (b) as the main considerations for designing our Liquidity Metric. We wanted to track both the size of the bid- ask orders (how much money is involved) and the spread of the orders (the price difference between how much people are willing to pay vs. the selling price).

While doing so,  we assigned less weightage to orders that were very far away from the mid price (i.e. the average of the best bid and ask prices) because not all orders are alike (e.g. if everyone is asking for $8,000 per Bitcoin, no one cares about that one guy wanting to sell for $10,000!).In that way, Liquidity Metric was a unique metric that tried, for the first time, to quantify the liquidity available on the order book with a single number.

Feedback Received About the Initial Solution

The feedback was mostly positive, as the Liquidity Metric presented an alternative to simply using volumes to “estimate” the liquidity of markets. However, we also received constructive user feedback on some points could be improved on:

1.      Many users did not understand why the metric was in dollars ($);

2.      Some market pairs had large liquidity numbers, but the liquidity turned out to be “ineffective”. An example would be a liquid market like BTC having a very large order (say 50 BTC away at 0.5% depth off the mid price). This order made the market pair look very liquid, but in effect, because most of the trading happens at 0.01%-0.05% depth off the mid price, 0.5% depth was simply “irrelevant” liquidity;

3.      The range of numbers reflected on the Liquidity Metric were very large and it was hard for our users to get a good grasp of comparisons between markets.

To address all the issues above, we went back to the drawing board and redesigned the algorithm and presentation of the metric. We now call this — Liquidity Score.

The Improved Solution – Liquidity Score

As mentioned previously, there are many different ways of how a market’s liquidity could be reported. Instead of focusing on (a) the absolute size of the order book and; (b) the spread of the order book, we decided to shift our focus to (d) slippage —the biggest factor that most of our retail users should be concerned about.

Slippage refers to the difference in execution price due to the size of the order executed. A large amount of slippage means that the buy (or sell) order executed at a much higher (or lower) price than anticipated, due to the illiquidity of the market.

A liquid market minimizes slippage incurred by traders and this can equate to a significant amount of money sometimes. . 

With that in mind, the redesigned Liquidity Score will be all about showing markets that result in the least slippage so that our users can trade more cost-effectively. The amount of slippage incurred is dependent on two variables:

1.      The size of the (buy or sell) market order;

2.      The bids and asks that the market order fills against (the immediate order book depth).

The resulting slippage is the percentage difference between the final execution price and the mid-price.

While tracking slippage is nothing new, another problem arose – what order size should we be tracking? $1,000? $10,000? Or $1,000,000? This is a difficult question to answer. 

If we only tracked a single number, we expose our Liquidity Score to be easily “gamed,” as exchanges wanting a high score would simply put enough liquidity at a defined spread to achieve the score that they wanted. Next, utilizing only a single order size to test would simply alienate a portion of users. Traders come in all sizes, and a professional trader would want to know how much a $100,000 order slips in the market, as compared to an amateur, who wants to find out how much his $500 order slips.

To satisfy the above requirements, we decided to design our Liquidity Score to track the slippage of various order sizes over a large range of order quantities from $100 to $200,000. The resulting score will be a summation of how much slippage is incurred in every one of the different simulated orders we track.

The Liquidity Score is reflected as a number from 0 to 1,000:

–         A score of a perfect “1,000” means that the market has a very low slippage for orders up to $200,000 in size;

–         A score of “0” means that the order books have less than $100 in total value on either the buy or sell side. This implies a very illiquid market! Be very careful when trading on markets with low scores!

A score between 0 and 1,000 reflects the varying amount of slippage that the different orders incurred, and the score is incremental in nature. In other words, the lower the slippage, the higher the score given.  

In addition, we made sure to track slippage both ways, on the buy (bid) and sell (ask) orders, as markets need to have sufficient orders on both sides of the order book. This means that exchanges cannot post “lobbed-sided” order books to give users the impression of high liquidity.

For more details on how the algorithm behind the Liquidity Score works, read our methodology page.

In Summary

Understanding the importance of liquidity and the concept of slippage is essential in markets, especially in markets like cryptocurrency where illiquidity is prevalent. As cryptoassets have yet to reach mainstream adoption, markets are still relatively illiquid, as compared to traditional markets. Tokens with smaller market cap will reflect scores of less than a 100, and this will be indicative that traders will incur sizeable amounts of slippage or face an inability to execute due to insufficient liquidity.

By releasing Liquidity Score, we wish to enable all our users to quickly identify the best crypto exchanges markets to trade on with the smallest costs of execution.We will track these scores going forward, and the scores will be used as a reference for further improvements to other metrics that we show on CoinMarketCap.

The post An In-Depth Look at CoinMarketCap’s Newly Improved Liquidity Score for Finding Best Crypto Exchanges appeared first on CoinMarketCap Blog.