2020 Market Data Integrity Report

  • September 7, 2020

Highlights: 

 

  • Only 31% of the CMC top 25 is being washtraded compared to over 90% just 1 year ago, a 3x improvement with their new rankings system.

  • 8 major wash trading exchanges still found on the current CMC top 25 have found a way to beat the new ranking system by faking orderbook liquidity, using ghost trades, and purchasing web visits.

  • New individual pair rankings metrics for all data sites have vastly improved crypto data site accuracy, however many exchange market makers have already found ways to manipulate these new methods which we review in detail.

  • Nomics was found to have the highest quality data ratings out of the top data sites and CMC the lowest.

  • BTI Verified has partnered with Ciphertrace and Numio to offer exchange partners AML, KYC, and trade surveillance to improve transparency and market maturity in the space.
 
 

Overview

 

It’s been 2 years since BTI launched with our first report on the integrity of data in the crypto space. Our methods have improved drastically over this time and have inspired changes in many of the data aggregators in the space.

However, as data collection and ranking methods change, so do the methods of those trying to manipulate these new systems.


In this report, we will review the latest ways some exchanges and market makers are using to beat the
new ranking systems of the top data aggregator sites, and offer our suggestions for improvement as we all push for higher data integrity standards. 

 
These new adjusted ranking metrics of the top data sites are not giving traders direct information on accurate volumes, but instead ranking tokens by a group of metrics and letting traders and token projects make their own judgement calls.

 

This leaves a lot of room for exchange marketing teams to manipulate ranking systems and pass off lower ratings as due to other factors that aren’t related to number of traders or true volume.
 
Looking back at our April 2019 report we found 17 exchanges in the CMC top 25 which were engaged in heavy wash trading over 99% of their volume. Today, there are only 8 we found which is a vast improvement, but still leaving plenty of room for more.
 
While we found over 80% of the volume claimed on the CMC top 25 to be fake volume in 2018 and 2019, today we only found 34% to be washtraded.
 

The top 10 rankings of all top crypto data sites are looking much more accurate with only ZB.com exchange faking their way into CMC’s top 10.

Coinbase remains the largest exchange in the US and is consistently ranked in the top 3 across all data aggregators.

Bithumb has overtaken Upbit as the largest exchange in South Korea by both true volume and liquidity across all data sites. 

Liquid remains the largest exchange in Japan with Bitstamp staying strong as Europe’s largest exchange. BTC Turk is the largest exchange in Turkey across all data aggregators. 

The largest exchange in Central America by true volume is Bitso, with Mercado Bitcoin the largest exchange in South America.

All of these top exchanges in the world have accurate volumes which match their high unique visitor counts and deep order books which score high on the liquidity metrics new rankings systems are using. 

Although, many other exchanges have already begun manipulating these new methods and working their way higher in the rankings.

 

Unique Visitors

In our initial report in August of 2018 we used unique visitors as the key metric in determining which exchanges were faking volumes. Coinmarketcap and Coingecko have added this metric into their new exchange ranking methodologies. 

However, just as easily as exchanges could pay bots to wash trade their volumes, there are also ways to manipulate this metric. There are numerous sites promoting unique visitors throughout the internet with various ways of funneling real unique visitors to their client sites. 

Many of these tactics produce unique visitors by taking out websites which are similar in spelling to others and forwarding them instead to their customer websites. These are very low quality visitors which are not looking for the exchange websites.

Some claims of these sites selling unique visitors include:

“From as low as $1.5 per 1,000 visitors”

“Determine how many unique visitors you wish to receive”

 

These sites can be found with just a simple google search.

 

Liquidity

Another metric being used to determine the validity of a currency market is the liquidity of the order book. While this metric can be very valuable in assessing the real volume of a market pair, however it is another metric which can be manipulated.

In the past 2 years of research we’ve found many exchanges which fake their volumes also use “ghost orders” to make their pair markets look healthier and more liquid than they actually are. These are orders which show up on the order book, but are impossible to market sell or buy into. 

Any real order that comes through the exchange will immediately cause the ghost order to disappear from the book. This can be seen in many of the long candle wicks in many exchanges.

 

Manipulation

Two large exchanges with reported high liquidity and unique visitors were still penalized by Coingecko’s volume normalization algorithm. Coingecko reduced OKEx’s reported volume by 52% and Huobi’s volume by 84%. These exchanges also received a poor C rating from Nomics’ data transparency method.

As seen in the table below, using Unique Visitors and Liquidity metrics alone allow rankings to still be manipulated by more savvy exchanges. Many exchanges ranked in CMC’s top 50 that also have 100% confidence scores on individual pair volume accuracy, are poorly rated by Coingecko, Cryptocompare, and Nomics. 

Just a few exchanges slipped through the cracks of Cryptocompare’s rating system with B ratings, while a few suspect exchanges still received 8 and 9’s from Coingecko’s Trust system. Nomics has rated all exchanges above poorly with C or D scores. BTI algorithms also detect high volume manipulation in all exchanges listed in this table and we advise any token project to review this data before paying listing fees.

 

Coinmarketcap

CMC updated their exchange ranking in May of 2020 by adding a web traffic and a liquidity metric to their rankings page. They’ve also added a “Confidence” score to each pair of a token’s markets. 

As seen below, CMC’s top exchange rankings are looking much better than they were a year ago. 

However as discussed earlier, any exchange can purchase web traffic and use ghost orders to fool liquidity metrics, so there are a few exchanges still manipulating their way into CMC’s top 15, and many more faking their way into the top 50. More will likely follow suit unless methodologies are improved.

 

Exchange Rankings

BTI algorithms were used to analyze the current top 50 CMC ranked exchanges. While we found over 75% of the exchanges with over 90% wash trading in previous reports, only 32% were now found to be excessively wash trading their volumes. This is a big step in the right direction but still leaves room for improvement. 

CMC’s top 10 outside of ZB.com exchange (heavy wash trading detected) looks much different than it did just a few months ago with BTI Verified exchanges making up the other 9 spots. BTI still found 17 exchanges in the top 50 to be using high levels of wash trading.

Exchange Ranking Score: C+

Currency Markets

CMC uses a “Confidence Score” for each of their pair markets. They state that they derive this score from an algorithm using a combination of factors including liquidity, web traffic, order time, and sales.  However, due to the limitations of this method many markets show a 100% confidence score with very obviously manipulated charts and wash trading that can be seen with just the naked eye.

Below are many examples of the markets we found with clearly manipulated charts or orders hitting the trade history having never appeared on the order book that any novice trader can see. When we ran our algorithms on most of these pairs over 95% of the volume was wash traded. These exchanges also score poorly on Nomics, Coingecko, and Cryptocompare yet have 100% High confidence scores from CMC.

Currency Markets Score: C-

 


Coingecko 

The data team at Coingecko have updated the way they rank exchanges via a “Trust Score” addition to their site and a normalized volume metric. They launched their Trust Score in May of 2019, and it has gone through 2 upgrades including the latest edition released in July of 2020 which adds a cyber security score element provided by Hacken. 

The Trust Score is calculated via their own formula of exchange order book liquidity, web traffic visits, API standards, and cyber security rating. We found this formula to be more robust and effective than CMC’s, but still presents some of the same issues with exchanges able to fake traffic as well as liquidity.

 

Exchange Rankings

They offer a “Normalized” trading volume of an exchange similar to our initial report findings and methods of 2018 which adjusts volume based on unique visitors and liquidity. BTI moved into a more algorithmic based model on individual trading of pairs, rather than a full exchange overview due to the limitations of this method. 

We were still able to find 8 exchanges in the Coingecko top 50 which have a high number of wash traded markets.

Exchange Ranking Score: B


Currency Markets

Coingecko uses a Trust Score for each individual pair market on an exchange. Their algorithm puts many more warnings on the suspect exchanges’ pair markets. However, as it is solely based on liquidity, although more accurate than CMC’s algorithm, it still has its limitations with ghost orders.

Currency Markets Score: B


Cryptocompare

The Cryptocompare team released their initial Exchange Benchmark Report in June of 2019. This report provided grades for exchanges based on liquidity, security, data integrity, management team, AML/KYC, legal, and asset quality.  

Cryptocompare weighs the exchange score with many other variables which do not directly relate to data transparency and accuracy, but does give solid direction at least most of the time on manipulative exchanges.

 

Exchange Rankings

Cryptocompare’s benchmarking analysis of exchanges rules out many shady exchanges with ratings below a B. We found just 3 exchanges in their top 50 which have a B rating or higher which have sizable amounts of wash trading including Exmo, Bibox, and Bitmax. 

Exchange Ranking Score: B+

 

Currency Markets

Cryptocompare pair markets do not have individual ratings and instead just show the general overall exchange rating for every pair on the exchange. This is helpful, but does not give much guidance on individual pairs with regards to liquidity or accurate volume. 

Currency Markets Score: B


Nomics

Nomics released their Exchange Transparency Ratings in April of 2019. Their methodology focuses primarily on API data transparency standards. They found that historical API data is frequently lacking in exchanges with shady behaviors. 

An A rating on Nomics is pretty straightforward with no other factors or algorithms at play. An A rated exchange provides every trade on every currency pair, dating back to the first day of trading of these exchange trading pairs.

In our review, we found exchanges with a C rating and below tend to be the biggest wash trading exchanges manipulating their data.

It is interesting to note that both Huobi and OKEx are C rated exchanges on Nomics (a poor data transparency rating), yet CMC has partnered and accepted both of them into their Data Accountability and Transparency Alliance initiative.

 

Exchange Rankings

Nomics applies these transparency ratings to both their coin market pages and their coin pairs pages. No exchanges in the Nomics top 50 with a B rating or higher have high levels of wash trading over 50% of volume.

Exchange Ranking Score:  A-

 

Currency Markets

Nomics displays the exchange ranking next to each pair market. It does not offer liquidity metrics to distinguish between pairs of an exchange as the same exchange ranking is given to every pair. C rated exchanges are displayed at the bottom of pair market lists, clearly identifying exchanges with suspect market history. 

However, Nomics does not provide more detail on positive ratings for more liquid and clean markets on these exchanges with less transparent data.

Currency Markets Score:  A-

 

BTI Verified

After 2 years of collecting and analyzing trade data to determine exchange manipulation tactics we are using all of our data points and knowledge to offer exchanges access to our wash trade surveillance software for their internal teams.

The most accurate way to determine real volume requires knowledge of which accounts are making which trades. With this data we are able to determine if any accounts are not only trading back and forth with each other, but if they are involved in pump and dump schemes, spoofing attempts, and many other strategies to manipulate markets.

The BTI Verified software tackles many of the issues plaguing current compliance teams and trade surveillance systems including false positives and inefficient reporting – causing too many resources and employee time being spent on exchange market surveillance.
The software also integrates AML scanning addresses and transactions through Ciphertrace and user onboarding KYC through Numio.

Bithumb and Bitforex are initial exchanges using the BTI Verified platform to monitor the integrity of their markets and improve current systems.

More information can be found at: btiverified.com

Share This Post

Share on facebook
Share on google
Share on twitter
Share on linkedin
Share on pinterest
Share on print
Share on email