This post was last updated on August 9th, 2019 at 10:02 pm
“In God we trust; all others bring data.”― W. Edwards Deming
“If you have a quality dataset… I’ll consider it. But when it comes to opinions, I’m going with my own.”― Andy Grove (apocryphal)
In its short life, the cryptocurrency industry has moved from margins to mania; from something irrelevant to all but those trading illicit goods on the dark web to the hottest investment opportunity in recent history.
Today, we’re again in a phase shift in crypto, only this time, it is from diagnosis – understanding how mania seized the industry – to a new opportunity for mainstream transparency.
- The First Age of Crypto
(Margins: 2006 – 2015)
- The Second Age of Crypto
(Mania: 2016 – 2017)
- The Third Age of Crypto
(Diagnosis: 2018 – June 2019)
- The Fourth Age of Crypto (Transparency Goes Mainstream: June 2019 & Beyond)
- Transparent Market Data As A Growth Tactic For Exchanges
- What Do We Believe?
- From Centralized Transparency To Trustless Transparency
- We Believe That Toxic Exchange Behavior Is A Problem Enabled Almost Entirely By The Data Practices We Accept
- Conclusion: Our Beliefs & Our Commitment
The first age of crypto was the world of Satoshi Nakamoto, Hal Finney, email lists and Bitcoin Talk. It was full of miscreants and rebels who were both convinced that the financial system was failing and daring (or crazy) enough to do something about it. This world was built on the foundation of decades of experimentation and ethos development from cryptographers and Cypherpunks.
The First Age of Crypto
(Margins: 2006 – 2015)
The first age of crypto found roots initially in the shadow places, where life operates outside and around state actors and formal markets.
The very first infrastructure allowing people to actually use Bitcoin was born. And although it may have sometimes been used for less than reputable affairs, this very real infrastructure gave the network utility, liquidity, and an incentive to provide security.
As people came into this nascent financial ecosystem, what they found wasn’t the reductive “dark web” caricature presented in the media, but something that represented a potentially fundamental change in how money worked and how people exchanged value.
The Second Age of Crypto
(Mania: 2016 – 2017)
It wasn’t long before people started to wonder what else you could build on top of the new digital asset infrastructure kicked off by Bitcoin – and if even they themselves could launch their own crypto tokens? The first coin sales happened on the Bitcoin blockchain as early as 2013, but it wasn’t until Ethereum launched — and radically simplified how people could create tokens via the ERC-20 standard — that things really heated up.
Over the course of 2016 and culminating in the ICO boom of 2017 and early 2018, what had started with Bitcoin’s promise for a permissionless and capped supply global digital money turned into a rallying cry to tokenize everything.
Dissatisfied not only with finance and the banking system, but an increasingly monopolistic technology field as well, entrepreneurs and developers flocked to the space. They had an idea that tokens could realign the incentives of networks to allow new forms of true peer-to-peer engagement without rent-seeking middlemen.
It just so happened that those same tokens that would be used to exchange value within the networks they would build were also the most powerful and friction-reducing tool for fundraising the world had ever seen. Tokens unlocked capital from anyone, anywhere, in any amount. Within months, token sales were massively outstripping traditional seed stage venture capital financing.
Of course, money on that scale rapidly attracted scammers and con artists alongside the good actors. When the ensuing maelstrom crashed, retail investors around the world were left holding largely illiquid assets often 95% or more off their all time highs, often with functionally dead communities and teams who had moved on to other things.
The Third Age of Crypto
(Diagnosis: 2018 – June 2019)
As the markets cooled off, many in the crypto industry sought to understand what had happened to create such a bubble. Was it simply a classic tale of the human capacity for greed on the one hand and the desire to get rich quick on the other? And even if that was the case, what enabled it to get so far?
For many, an important causal factor was the pernicious incentive alignment between tokens, exchanges, and price aggregators to make things look as bullish as possible.
If tokens were doing well, project founders (and existing bag holders) saw their net worth rise, exchanges saw higher volume (and could consequently command higher listing fees), and aggregators saw more traffic, enabling them to charge more for ads … often from the same exchanges and crypto projects whose fake volumes they were publishing.
Indeed, these incentives were so powerful that in some cases they seemed to have trumped ethical consideration. The last few years saw rampant wash trading among exchanges, liquidity providers, token projects and investors. Additionally, a new kind of spam arrived on the scene: exchange spam.
CoinMarketCap & The Rise of Exchange Spam
While the web is largely indexed and organized by search engines like Google, the crypto ecosystem (of token projects, exchanges, and related metadata) is organized by aggregators like Nomics and CoinMarketCap.
As opposed to crawling the web, as Google does, these aggregators ingest structured exchange data via APIs. Unfortunately, data ingested from these APIs are not created equal. On one end of the transparency spectrum, there is historical raw trade data – the firehose of granular data of the complete record of activity. On the other, there is low-granularity ticker data with limited history. Historical raw trade data makes malfeasance extremely difficult to hide, while ticker data – you guessed it – makes it easy.
For the majority of CoinMarketCap’s history, exchanges have only needed to provide non-historical ticker data in order to get indexed by the aggregator. Given that for some time, CoinMarketCap was the only game in town, the unfortunate legacy of these non-rigorous data requirements is that the vast majority of exchanges (i.e. over 75%) only provide non-historical ticker data via their public market data APIs (i.e. the bare minimum to get listed). Put another way, most crypto exchanges only provide low quality, non-transparent, non-historical market data. And its this kind of data that is the easiest to manipulate, and therefore most frequently used to spam aggregators and fake volume.
Historically, these aggregators have ranked exchanges and token projects by simplistic & imperfect indicators like market cap (for cryptoassets) and trade volume (for crypto exchanges). The simplicity of these ranking methodologies made spamming sites like CoinMarketCap relatively easy.
Unfortunately for consumers, these ranking sites have historically lacked incentives to clamp down on this activity. In fact, it was quite the opposite: these sites often had offending exchanges and token projects as some of their largest customers.
Make no mistake about it, this is big business. One source who ran ads on CoinMarketCap in December of 2017 reported that a month-long placement at the top of CoinMarketCap.com might cost over $1M USD. Indeed, some of the most offensive, toxic, and fraudulent crypto projects and exchanges have been among the largest customers of CoinMarketCap’s ad products, often purchasing their most expensive ad blocks and packages.
Exhibit #1: BitConnect, Laser Online, and Bitpetite’s Premium Ad Placement At The Top of CoinMarketCap’s Home Page
Above you will observe BitConnect, Laster Online, and Bitpetite ads all running on the home page of CoinMarketCap.com. All three were high-yield investment program (HYIP) ponzi schemes & exit scams (with BitConnect promising 1% daily compound interest, Laster Online promising 12% daily profit, and Bepetite promising 4.5% daily). As the most trafficked crypto site at the time according to Alexa, CoinMarketCap helped these (obvious) scams reach a global audience by accepting them as advertisers.
As the most notable of the three, Bitconnect was, in part, a high-yield investment program (i.e. Ponzi scheme) that pulled off one of the most iconic exit scams in crypto history. In January of 2018, a U.S. District Court granted a restraining order that froze Bitconnect’s assets and “to disclose cryptocurrency wallet and trading account addresses, as well as the identities of anyone to whom Bitconnect has sent digital currencies.” The head of Bitconnect India, Divyesh Darji, was also arrested in 2018 and at the time of writing is wanted for actions related to another alleged cryptocurrency scam.
Exhibit #2: FCoin’s May of 2019 Premier Ad Placement On CoinMarketCap
CoinMarketCap’s willingness to work with (and promote) toxic members of the crypto community, especially when they are willing to purchase expensive ad slots, is not limited to the crypto craze of 2017. Indeed, CoinMarketCap has promoted wash traders as late as March of 2019.
According to page 8 of Bitwise’s May of 2019 report to the SEC . . .
The CoinMarketCap data gets even more surprising when you dig into any particular exchange. For instance, the largest reported exchange in April, Fcoin, with $1.7 billion in daily trading volume, has a limited public profile. It has been mentioned just four times on Bloomberg (all of which are in the context of fake volume), has just 4,781 followers on twitter, and is the 56,539th largest website in the world, according to Amazon’s Alexa.com. By comparison, the largest exchange that the original Bitwise Study showed had real volume – the 15th ranked exchange, Binance–reported just $218 million in daily trading volume (less than 1/7th of Fcoin), but has been mentioned on Bloomberg 6,830 times, has 342k people following its CEO on Twitter, and is ranked as the 971st largest website in the world by Alexa.com. This doesn’t make sense.
Elsewhere in their whitepaper (page 71), Bitwise states that . . .
“[FCoin was] the number one exchange by reported volume in April … with $1.7 billion in reported volume. It was also the number one exchange by reported volume in March, with $802 million in average daily volume. But in February, it was the 55th largest reported exchange, with just $12 million in reported volume. The crypto market moves fast, but rising from $12 million in average daily volume to $1.7 Billion average daily volume is hard to believe.”
An Unholy Alliance
The unholy alliance between crypto projects, exchanges, and aggregators like CoinMarketCap that began in 2016 and 2017, came under scrutiny in 2019 with the release of Bitwise Investment’s presentation and whitepaper to the SEC.
Since then, the public has become much more aware of a range of activities that have driven the growth of crypto’s seedy underbelly.
These include, but are not limited to:
- Exchanges engaging in wash trading to increase volumes (and, as a result, to cheat their way to the top of ranking sites like CoinMarketcap).
- Exchanges engaging in ticker stuffing to fake increase volumes (see the FCoin example above).
- Aggregators accepting large fees in the form of ad payments (to look the other way).
- Aggregators allowing token projects (usually who have purchased lots of ads) to influence the listed “circulating supply” for an asset, thereby inflating an asset’s market cap and ranking.
- Token projects paying exchanges to facilitate wash trading (i.e. volume as a service)
- Token projects paying liquidity providers for exchange volume as a service (or price manipulation)
In all of the above, investors are getting screwed.
(Note: The list above is far from complete. If you have additional examples of toxic activity, please share in the comments).
Again, the issues with CoinMarketCap’s promotion of toxic actors do not seem to have ended after…
- Bitwise Investment’s March of 2019 presentation to the SEC
- Bitwise’s May of 2019 whitepaper to the SEC, or
- CoinMarketCap’s promises to do better.
Indeed as of today’s date of publication (August 5th, 2019), there is an ad running on CoinMarketCap touting 20% monthly returns, as well as CoinTiger in the #1 spot and CoinBene in the 6th ranking spot.
Let me put this into perspective: in June of 2019, Cointiger had just 782K visits according to SimilarWeb.com vs. Binance (ranked just below CoinTiger in Figure 3 above) at 44.55M visitors. Furthermore, the visitors to CoinTiger stayed an average of 23 seconds vs. an average of 9 min. on Binance; this suggests that the Binance visitors were trading and that the CoinTiger visitors might have been paid traffic or bots. Additionally, the bounce rate for CoinTiger (i.e. 96%) vs. 37% for Binance suggests that the vast majority of CoinTiger visitors were not trading.
Finally, in Figure 3, CoinBene is ranked in the 6th spot above exchanges like CoinBase. For reference, Bitwise dedicated 5 entire slides to Coinbene’s toxic practices and fake volume March 2019 report to the SEC (see slides 29 – 34).
The Fourth Age of Crypto (Transparency Goes Mainstream: June 2019 & Beyond)
The first half of 2019 saw an aggressive calling out of rampant wash trading, ticker stuffing, exchange spamming, and fake volumes on the part of exchanges in an attempt to boost rankings on large crypto aggregators. Indeed, Bitwise Investments’ research (presented to the SEC) showed that nearly 95% of BTC trade volume is faked. Unsurprisingly, this evoked a healthy response from a number of upstart data aggregators who responded by — among other things — creating new exchange ranking criteria and algorithms.
This shift in tone signals the end of the second (mania) age of crypto and the beginning of the third – the mainstream era.
Yet as we enter this new era, it is desperately important that we not only embrace and embody transparency, but that we do so in a way that stays true to the fundamental roots of the industry as a whole: trustlessness, leaderlessness, and the belief in everyone’s capacity as self-contained and sovereign individuals.
Luckily, we’re seeing this kind of thinking emerge more and more.
Transparent Market Data As A Growth Tactic For Exchanges
Market forces are demanding that exchanges adopt increasingly transparent data distribution practices. In this context, the availability of highly granular market data, with full history, is becoming a marketing & distribution opportunity for exchanges. At the same time, not having this data is becoming a sales liability.
Distribution of Market Data as a Marketing & Distribution Strategy for Crypto Exchanges
The opportunity for permissionless innovation has fostered an environment that’s birthed hundreds of cryptocurrency exchanges. Nearly every day a new crypto exchange is born (of course, whether that exchange is successful is another story). Using toolkits from projects like 0x, an experienced team can launch a basic crypto-to-crypto exchange in a few days.
In such an environment, exchange market data becomes an important marketing asset, and the distribution of that data by aggregators becomes an important marketing channel.
Exchanges’ products are their trading pairs. They make money (primarily) from fees earned when traders execute trades across these pairs. Given how highly competitive the crypto exchange business is, it behooves exchanges to distribute info about their product offerings — i.e. their trading pairs — as broadly as possible so they can get listed on as many aggregators, news sites, registries, and market data APIs as possible (imagine the growth prospects of an exchange not listed on the usual aggregators). Because cryptoassets can only be listed on aggregators, news sites, etc. after these services are receiving market data for trading pairs involving a project’s token, crypto projects are incentivized to get listed by exchanges with wide market data distribution.
(Side note: Crypto projects are constantly approaching us, asking us to add their asset to Nomics’ list of cryptocurrencies. Our answer to them is always that we can’t list an asset directly; instead, we need to add an exchange that can provide market data about their asset so we can price it. If such an exchange with even a bare-bones API doesn’t exist, we can’t list the asset).
In contrast to this highly-competitive crypto exchange market, equities exchanges like Nasdaq and the New York Stock Exchange have a de facto monopoly over trading in the United States, and therefore have a monopoly over related market data (which is partly why ICE, the company that owns the New York Stock Exchange, makes more money from data fees than trading fees).
Poor Market Data Availability As A Sales Liability For Exchanges
Not only is the distribution of highly granular market data (with full-history) a marketing opportunity, but the unwillingness to provide this kind of data is also a liability for exchanges. Our experience serving customers in this space (via our market data API) has reinforced one thing: institutions (with very few exceptions) just don’t work with exchanges that don’t provide full trade-level history.
This preference for highly transparent exchanges has only been amplified in the first half of 2019, following Bitwise Investment’s presentation and report to the SEC, in which they claim that (1) only 10 of the exchanges they examined had “actual [BTC] volume,” and (2) that 95% of reported BTC exchange volume is fake.
Our examination of the Bitwise report revealed that eight of the 10 cryptocurrency exchanges identified as good actors by Bitwise, provide historical trade level data (i.e. the most granular and audible form of pricing data available) about exchange activity. That is, the one thing we found in common among the good actors identified by Bitwise, is they were very transparent about trading activity. In contrast, we found that — of the 16 exchanges explicitly called out by Bitwise as bad actors — all but two provide limited trading history and virtually no granularity around trading activity.
The good actors from Bitwise’s report had highly transparent data practices, and the bad actors were largely not transparent.
What Do We Believe?
We believe that investors, regulators, developers, users and everyone else has become smarter and more sophisticated about digital assets. We believe that, more than ever, the good actors are crowding out the space of the bad. But in this shift, we also believe the industry must transition from centralized transparency to trustless transparency.
From Centralized Transparency To Trustless Transparency
Centralized transparency refers to the idea of trusting a transparency agency or transparency specialist, rather than having truly democratized access to the source data upon which transparency is founded. Historically speaking, this form of transparency has led all too often to fake transparency. In many respects, centralized transparency is worse than no transparency at all, as it provides the illusion of safety.
- Fake transparency gave us ratings agencies — i.e. Standard & Poor’s & Moody’s — that provided investors with false confidence in the risky mortgage-backed securities that precipitated the housing crisis of 2007 – 2008 (the worst economic disaster since the Great Depression of 1929). These inflated ratings helped the financial system take on far more risk than it could safely handle.
- Fake transparency gave us the EDGAR disclosure system, which is so jam-packed with often thousands of pages (and sometimes hundreds of thousands of pages) of non-machine readable docs about each company that important facts become unidentifiable without a sizeable resource investment.
- Fake transparency gave us the financial data economy, where companies like Intercontinental Exchange (ICE), which runs the New York Stock Exchange, provide wealthy traders and institutions with privileged, and advanced, access to data in a manner that allows them to front-run the market and drive up the cost of trades for everyone else. Indeed, these companies make the bulk of their profits not by fees collected by directly serving customers, but instead by selling data that allow a select few to profit from their average user.
(Note: The video below is NSFW.)
The illusion of transparency can be used to cover all manner of sin.
In this spirit, we assert that . . .
- Crypto doesn’t need a Standard & Poor’s or Moody’s
- Crypto doesn’t need an ICE Data Services (the service that provides privileged access to New York Stock Exchange Data)
- Crypto doesn’t need a bloated EDGAR disclosures database
- Transparency doesn’t just mean being transparent with a ratings agency or curator, who blesses an entity or financial product for all of us.
- Trustless transparency requires that disclosing entities provide primary source data, available to all, directly from their web properties
- It’s unethical to charge exchanges or cryptoassets for ratings (see the video below)
Crypto needs primary source data in its purest form, available to all without privileged backdoors that grant enhanced granularity or lower-latency access. This data can be used to provide incredible clarity and confidence when it’s available to all, at scale, and analyzed by open-source methodologies that can be criticized by the public.
As we recognize the way data is used and manipulated in the crypto industry, our approach can’t be to simply create new layers of abstraction, or new centralized ratings agencies or curators that bless projects “good” or “bad” and require us to simply trust them.
Instead, we must think critically about:
- The types of data exchanges make (and do not make) available
- The conclusions that are drawn from that data
- The methodologies used for drawing conclusions (which should be open sourced)
The whole point of this industry is “don’t trust, verify.” What makes that possible is, simply put, better data, exposed for all, with all the tools to allow anyone and everyone with an interest to come to their own conclusions and make their own decisions.
We Believe That Toxic Exchange Behavior Is A Problem Enabled Almost Entirely By The Data Practices We Accept
Just like an IRS audit, the more data history and granularity provided by an exchange engaging in nefarious activities, the more likely they are to be caught. On the other hand, upstanding exchange operators have every incentive to provide high-granularity data (with as much history as possible), as this kind of transparency attracts market makers, generates the broader discovery of an exchange’s markets and trading pairs and engenders trust among institutional traders, investors, and regulators.
To us then, the answer to exchanges behaving badly are incentives that demand access to the complete data for anyone to audit, and market-led accountability for those who don’t.
This is why we’re building Nomics as the crypto data company. We believe our job is to aggregate all market data – not some of it. Not just the data from “good” projects or exchanges; all of it. We believe that when you empower markets with data and the tools to use it, market forces will drive out bad actors.
In data veritas.
Conclusion: Our Beliefs & Our Commitment
Nomics is building the transparent data infrastructure for the mainstream generation of crypto. As we endeavor on this mission, here are five principles that guide us:
- Transparency must be trustless – Transparency can’t come exclusively from third party transparency providers that bless exchanges, projects, and financial institutions for all of us (although they can be a valuable part of the ecosystem). Transparency must be rooted in open data and methodologies.
- Cryptocentric outlook on data – Too much of today’s data is presented in comparison to fiat, and in a manner that accepts USD as the de facto quote currency. But as we strive towards a world where cryptoassets overtake fiat as the world’s unit of account, we must increasingly provide crypto-centric views of the financial world. That is why every page on Nomics.com allows the user to select almost any cryptoasset as the quote currency. The whole point is to be able to effortlessly understand how digital assets compare to one another.
- Agnosticism about tokens and projects but not about data transparency – We don’t believe it is the role of data aggregators to take positions about the quality of projects and tokens; instead, it’s their role to aggregate as much data as possible. In so doing, however, it is the role of data aggregators to take an aggressive stance about the quality of data transparency from exchanges and token projects. Providing subpar data is the easiest way for exchanges to hide bad behavior, and we must have market forces demanding the best data possible: historic, trade level data.
- No payments in exchange for listings, ratings, or badges – We will never charge an exchange, cryptoasset project, or anyone else for a rating, index inclusion, registry inclusion, badge or any other seal of approval on Nomics.com. Not even to cover our own expenses.
- We won’t censor crypto projects or exchanges – It is our job as a data aggregator to not only provide as much data as possible, but to fight for more data from the source. Given that, we make it a point not to hide data from unsavory sources (although we will exclude low-transparency data from our aggregated prices and indicators). Instead, we give you the tools to contextualize data and make your own decisions about how to treat it.
We are embarking on a new, fourth, mainstream era of crypto that demands better data practices. And we’re doing our part to help that become a reality. Our goal is to provide crypto markets with the most comprehensive data set possible and the most powerful tools with which to work with that data.