The Numbers Behind 0x
|Joel John||May 8|
This is part of a series I will be doing on different ledgers in collaboration with Nyctale.io. They are currently pioneering an in-depth study on on-chain behavioral economics and have offered me data-sets as a part of it.
If you are an analyst, with a record of publishing about the industry often, do reach out to me at email@example.com - I will be happy to plug you to the team.
What is 0x
Ox is a protocol layer that makes it possible for individuals to settle trades against one another without the need of a centralised custodian. In a traditional environment, settlement of a trade at scale requires two parties to send the assets they are swapping to a central party (eg: NYSE, Bitfinex, Bitmex etc) that will then match orders on basis of supply and demand. This moves the risk of custody to the exchange and gives the exchange itself an outlook on who is willing to pay how much for an asset or liquidity. The book Flash Boys by Michael Lewis gives an interesting look at the extent to which funds may go for the slightest edge against other peers in the market. The biggest challenge in creating decentralised exchanges is the maintenance of an orderbook on-chain. Decentralised asset marketplaces have existed since 2015 in Nxt and Bitshares but they have not been able to see scale due to the fact that orderbooks being on-chain can bloat up a network and slow it down completely.
0x finds a work-around to this by relying on what they call relayers. Relayers - as the name implies do the job of taking bid and ask data from individuals and sending it across a network. In the event that an order is matched, the settlement itself happens onchain. Each bid and ask is cryptographically signed and can in-fact be sent through multiple mediums including a snail mail if needed. What this means is, the Ethereum network will not be storing information of bids and asks that are not filled. This in turn translates to lower expenses on gas costs & the network remaining functional for the most of it. It also brings modularity to how order-books work. Individual DApps can have their own relayers looking at what a community internally is willing to pay for an asset. However, said relayers can communicate with multiple other DApps to see if there is liquidity in the network of relays. For more on understanding how relays and networked liquidity works - I suggest exploring the 0x wiki. Some quick stats before we dive into how staking works in 0x. According to 0xtracker.com, there have been over 10300 traders providing some $340 million in liquidity in 170,000 traders on 0x's most used relay (Radar-Relay) so far. Quite impressive I'd say.
Staking and Governance In 0x
All markets require liquidity and at the crux of what drives interest in 0x is a system that rewards individuals for staking towards the same. Market-makers on the network are rewarded in ETH for filling orders. The proportion of the fee they receive is reliant on the amount of 0x that is staked towards the market-maker. What this basically means is - a market-maker with a lower amount in liquidity to provide could realistically earn more than someone else bringing in more liquidity in dollar terms or trade count so long as there are individuals with ZRX staking towards the market maker with lower amounts to offer. ZRX staking drives an efficient market as every stake is a vote on how efficient a market-maker is in the protocol. In addition - the project has also released a series of incentives and on-boarding guides towards making it easier for institutional automakers
It also means ZRX token-holders are placing a bet on the market-makers they stake with because if they act adverserially the value of the ZRX token they have staked could drop drastically. The democratisation of ZRX staking has started with services like Coinbase rolling out support for it. According to 0xtracker.com, some ~260 ETH has so far been distributed through the network. The ZRX wiki on staking is an interesting resource for anyone looking to dig further into how it works.
On-Chain Activity in 0x
One way to get an estimate of the chain's health is to take a look at the number of individuals engaging with the wallet. For the purpose of this, we take a look at the total number of wallets holding the token - and the ones that are actively sending or receiving tokens. Instances where there are more tokens receiving the token than sending usually indicate the network is being used to hold value increasingly. Receiving wallets increase in count when individuals move tokens to cold storage addresses or batch send tokens to multiple addresses. In the case of 0x, there is a market difference between the count of receiving address and the ones sending it. We will shed light on this shortly, but for starters - we take a look at how total wallet count (existing), new and active wallets differ.
Wallet behavior on 0x indicate a healthy addition of new users & retention of existing ones
There are a total of some 400,000 wallets today that have engaged with the 0x token in some form. Of these, in an average month about 5k wallets are active. The ratio of new wallets to active wallets is about 1:2. For every new user the ledger adds, there are two existing users that are actively using the ledger. These statistics offer some unique insights about the ledger itself. While open to interpretation, this is how we see it
The high number of wallets that have engaged with holding a 0x token in some form represent the fact that the token has been used beyond its utility, primarily for speculation or holding over the course of time.
The 5k active wallets per month represent a high number of active users within the network. This number is a variable depending on volatility in the market given the role the project plays in the DEX space. Volatility also translates to a rise in new wallets as more individuals engage with the project to acquire erc-20 tokens faster than they otherwise could from a traditional exchange. The token has the advantage of using market conditions for self-marketing
The balance between new wallets and existing ones show that the project is able to attract new users while being able to retain existing ones. We often notice that projects with a high number of new users manage to do so because of bounty programmes. 0x on the other hand has a steady balance between new comers and active wallets.
The problem with this data and the one given below is that it is likely that the vast majority of the users mentioned here may likely be using the asset only for staking or trading instead of directly using the underlying utility
Transaction Activity For 0x Tokens
The average weekly transaction on the 0x token is at about ~5000 transactions with the regional peak being at slightly close to 11,000. There is an inclination towards receiving wallets being markedly higher than the number of wallets sending the tokens. In total, over 900,000 transactions have occurred on the ledger with about 360,000 of them occurring the past year alone. Q1 2020 has already witnessed ~100,000 transactions. If those trends hold, we will see the year, end with a 10-15% increase in the total transaction count on the 0x ledger leaving the total at a high of 400,000. Almost a 300% rise in the total transaction count in 2017, the year the 0x ICO itself happened. This matters because where transaction counts may have fueled in 2017 as a result of the ICO boom and speculation, the current transaction peaks are a result of the increasing reliance of the De-Fi ecosystem on 0x. Transaction counts themselves do not speak much. In order to take a deeper look, we will have to study the volume moved and the average size of a transaction on the 0x ledger.
When put in light of the number of unique transactions that occur on the chain, some ~6000 to 10,000 tokens are transferred in each mean transaction. At 0x's current price of ~$0.2 that comes to around ~$1200, indicating that the chain has a relatively high concentration of whales and speculators engaging with the token. On studying different user-profiles within the ledger, it became quite clear to us that whales indeed hodl a large part of the network. Given the depth of that data, I have not added it to this publication. However, there are some signs about what is going on when we look at trade volumes and frequency
In order to do this, we need to explore the number of trades that occur on 0x and the volume that comes with it
Trade Count & Volume Recover ..
Close to 343 million in volume has been generated from 0x markets with some 620k trades occuring since late 2019. The network was able to generate some ~260 ETH in volume over the past quarter. Primarily driving on the volatility of March 12th and growing adoption of DeFi . Based on the above data, we were able to compute the average DAI traded per week on the network. It has risen from a mere ~90 dollars to over ~1800 dollars during the events of March 12th. As of writing this piece, that figure continues to stay at a 10x high of ~917 dollars per week. Another indication of the fact that volatility does the job of marketing 0x fairly well. The reasoning for this becomes quite clear when we look at the number of trades occurring on the platform. January 2020 was the first month of substantial volatility when (i) DeFi was established as a theme and (ii) price momentum was upwards. Seemingly, individuals have used 0x to trade quite frequently during that phase.
But User Count May Not Be Growing (Yet)
Naturally, in order to have a holistic look at what is going on in terms of activity - we will need to look at what is occurring with the number of traders that are actively trading and the average trade counts. This is where the numbers get a little interesting. In terms of the size of the average trade - we have seen a marked democratisation. The figure has dropped from a peak of $3000 to mere $1000 for the month I am writing this. In fact prior to the recent volatile months the average size of a trade on 0x was $200 - indicating that it is not just whales using the project but also retail folks sent in through DeFi products.
However - based on judging the number of traders on the platform, I am inclined to believe the active user's growth rate may not be as high as anticipated. If anything - it may have flatlined. One way I see this data being "dirty" is in terms of individuals directly engaging through relayer contracts instead of using 0x itself. Products like Nuo have seen an explosion in user-count and routinely wire orders through 0x. As such, this data in particular can't be relied. What I do find interesting is that the ~300 or so wallets that engage with 0x on a daily basis seem to be incredibly sticky. The future of the project more or less relies on (i) bringing in more DApps and DeFi products that can leverage the base-layer to accelerate internal economies and (ii) on-boarding more traditional marketmakers to give the kind of liquidity a traditional exchange can.
0x has clearly come a long way from the days of the ICO hype in 2017 and if anything has delivered the product they set out to. The boom in DeFi may just not have happened if the money APIs that enable the seamlessly easy transfer and conversion of value between projects and assets today was not set in place. It is also one of the few projects where the numbers display a visible recovery from the hype cycles of 2018. The question now is (i) whether the underlying fee structure can evolve substantially enough for the governance and staking mechanisms at play to be profitable and (ii) how much of 0x's market-share in the Dex space will be eaten up by competitors.
I will be diving into some of that in the next piece covering Kyber and shed light on what 0x v3 means for the project and the market.
Have a pleasant weekend.
Data sources - Nyctale, Nansen and Santiment
Thanks to Alex Svanevik for helping me think this through
At the time of writing this, I hold nothing in 0x tokens.
My pet dog says, I should mention this is not investment advice
I believe you should not speculate on tokens and as such - kindly don't take this as a buy / sell recommendation. The purpose of this piece is to explore the economics at play behind 0x's operations
Nyctale.io is pioneering some cutting edge work in behavioral economics. I have not shared all of their data-sets in this piece. However, if you are an analyst - please drop me a message. I would be happy to intro
Nycale is a portfolio company at Outlier Ventures. My current employer.
If you find any of the numbers / calculations being wrong - do drop me a DM. Happy to pay a $10 bounty for corrections
0xtracker is an incredible public good. Keep an eye on the project if you are 0x enthusiast.