Our Network: Issue #23

Coverage on BAT, LINK, Nexus Mutual, Opyn, and PoolTogether.

Click here to join our telegram chat. 

Welcome to Issue #23 of Our Network, the newsletter about on-chain analytics that reaches more than 2000 crypto investors every week.

This week our contributors cover the following projects:

  • PoolTogether

  • Nexus Mutual

  • Chainlink

  • Opyn

  • Basic Attention Token 🆕


Network Coverage

🟢 PoolTogether

Contributor: Leighton Cusack, CEO of PoolTogether

  • PoolTogether is a protocol for no loss savings games on Ethereum. The two primary growth metrics are unique wallets with deposits and total deposited funds. Unique users have grown from 423 on January 3rd 2020, to 8,296 on May 28th. Monthly user growth is shown in the image below.

  • In contrast to the consistent user growth, total funds deposited has fluctuated up and down. Early in the year, the pool was dominated by $100,000+ deposits, this has been replaced by many more deposits from smaller holders.

  • The prizes the protocol awards are derived from the interest earned on all the deposited dollars, due to major decreases in interest rates (especially after the market crash in mid-March) total prizes awarded per month have also fluctuated. The following chart shows total prizes disbursed so far in 2020.

  • PoolTogether released the "pods" functionality enabling users to link their tickets together and automatically split any prize. The pods have grown to be the single largest ticket holders making them a unique investment vehicle. Currently, the Dai Pod has 82,060 tickets deposited in it, equating to a 1 in 6.43 chance of winning each week. Since the Dai pool has a total of 893,895 each time the Dai pod wins ticket holders are effectively earning 10x the interest they would have earned over the same period of time.

  • The Argent Wallet recently launched a direct integration into the PoolTogether protocol. This provides a good data point in how improving ease of use can expand protocol usage. Net new users increased 240% in the 10 days after the launch of the Argent integration compared to the 14 days prior to the integration launch. This shows significant untapped demand once onboarding is improved.

🟢 Nexus Mutual

Contributor: Richard Chen, Partner at 1confirmation

  • Two large covers of 1000 ETH (~$220k) each were taken out recently for DXdao and ParaSwap. You can keep track of all new covers with this Twitter bot.

  • Below are active cover amounts (in USD) on each project. Flexa ($550k), dYdX ($455k), Uniswap ($421k), and Compound ($405k) have the most money insured.

  • Active cover amount increased by 12% from $2.8M to $3.13M over the last month. There is still a lot of cover capacity available as currently the mutual can support up to $18.3M in total covers. The upcoming partial claims launch will expand coverage to not only smart contract bugs but also oracle risk.

  • Since inception 2032 NXM (worth $7169) in staking rewards has been paid out to NXM stakers. The upcoming pooled staking launch will increase staking rewards by 2.5x, encouraging more people to stake and thus decreasing the cost of covers.

🟢 Chainlink

Contributors: ChainLinkGod and CryptoSponge

  • This may not be news to you, but Ethereum’s blockspace is not free. Because of this, tracking the total percentage of blockspace (gas) being consumed by a protocol shows just how much stress (demand) an application is putting on Ethereum’s bandwidth. A higher level of gas consumption is bullish, but at the cost of making Ethereum transactions for everybody else.

    The chart below shows what percentage of gas available each day is being consumed by Chainlink oracle networks. At mainnet launch, with only a single price feed (ETH/USD) and three nodes, Chainlink was consuming 0.33% of Ethereum’s daily bandwidth available. Over the course of 2019 this rose to a peak of 3.5% when there was two price feeds (ETH/USD and BTC/USD) each with 21 oracle nodes causing a 10x in bandwidth consumption.

    There was peak up to 6% consumption during Black Thursday, but since then gas consumption has stayed nearly completely rock solid at 1% despite the fact that there are now 30+ price feeds supported by 30+ oracle nodes. One potential reason for this decrease and stabilization in gas consumption could be attributed to Chainlink networks moving to a deviation based schedule instead of just a heartbeat schedule thus matching the demand rhythms of Ethereum itself (as volatility goes up more liquidations and arbitrage happens), but if you have any other theories please let us know!

  • A key metric to visualizing the sustainability and profitability of running an active Chainlink node is overlapping the daily LINK rewards in USD to how much ETH in USD the data requesters and node operators spend daily on transaction fees (gas costs). To create a sustainable oracle network, rewards paid to nodes for external data needs to be higher than the costs spent by those nodes in delivering that data on-chain. The requesters total costs are LINK paid to nodes + ETH paid for data request transactions.

    There was a very clear and obvious spike that occurred on the 12th and 13th of March, due to Black Thursday where Ethereum’s transaction fees skyrocketed. Since Chainlink oracle networks now update on a deviation basis, the more volatility there is, the more updates that need to be triggered, leading to such a large spike. On these particular days the USD spent on transaction gas costs are far higher than the rewards, but luckily these situations are an anomaly and do not affect the overall sustainability for running a node.

    Lately average Ethereum gas fees have been far higher than normal digging into the profitability of running a Chainlink node. Most of these costs are borne by the requesters (currently the CL team) but a non-insignificant portion is also being paid by the nodes for their response transactions. If this continues into the future, LINK rewards to nodes may be bumped up, but with technical improvements to decreases gas consumption this will largely become a non-issue.

  • Each Price Feed built using the Chainlink framework pays a different amount of LINK rewards to oracle nodes depending on the security needs of each particular network (more security needed = more LINK needs to be paid to nodes). This is best represented by the below chart which compares the amount of total job runs (data requests to each node) per price feed to the amount of LINK rewards (provides economic security) per price feed.

    What this shows is that while most networks have a similar amount of job runs per day, the ETH/USD (light blue) and BTC/USD (light green) networks are by far the highest paying price feeds in the Chainlink ecosystem. This intuitively makes sense as these particular feeds are not only high in demand by DeFi applications, but they contain a higher level of decentralization than most networks (21 nodes as opposed to 7-9) thus requiring more payments per update.

    It’s also not too much of a surprise that the ETH/USD price feed is currently the most secure and in-demand Chainlink network on the Ethereum blockchain.

  • Zooming out and looking at the distribution of LINK payments paid to each active Chainlink node since mainnet launch in May 2019 gives some interesting information on the evolution of the Chainlink network. The first observation is that the network has been continually decentralizing over time. First only beginning with three nodes (Chainlink team, LinkPool, and Fiews), the network has since grown to over 30 active nodes, a 10x increase in decentralization.

    The overall percentage of LINK payments paid to the Chainlink core team’s node has also drastically lowered over time. Initially beginning at 33.33% of the total daily payments, it has since lowered to only 1.34%. Additionally payments have become more variable in the last third of the chart due to the transition to a deviation based schedule. Additionally notice how each node is paid differently, this is due to the fact that not every Chainlink node is not equal in quality and thus do not serve the same amount of oracle networks.

Dig into the charts from today’s post here: Source 1, Source 2

🟢 Opyn

Contributor: Zubin Koticha, CEO of Opyn

  • Opyn provides DeFi risk management built using protective put option tokens called oTokens. Opyn provides protection on Compound deposits and ETH protection with ETH protective put options (oETH). At time of writing, there is currently an all time high of $2.17mm of collateral protecting ETH and Compound deposits (source).

  • Opyn has 5x’d total notional volume in the past month with $36mm in notional volume since launch, $25.45mm in notional volume this past month and over $6.5mm in notional volume this past week (source). Notional = oETH * ETH price at time of trade. This is how notional is usually calculated for options in traditional finance (option * price of asset at time of trade).

  • Daily notional volume has doubled over the past month with $1.08mm seven day moving average for daily notional volume and $848k thirty day moving average for daily notional volume. Additionally, in the past month, Opyn hit a new daily volume record with $2.55mm in notional volume on May 10th (source).

  • In the last month the number of unique Ethereum addresses interacting with oTokens has doubled with, 752 unique Ethereum addresses having interacting with oTokens, protecting themselves against DeFi risks (source).

🟢 Basic Attention Token

Contributor: Blaise Cavalli, CEO and co-founder at Nyctale

  • BAT network has an impressive long-term investor community, with more than 160k wallets exhibiting holder behavior (balance superior to 10 tokens, weekly ‘balance evolution’ < 25%, weekly ratio ‘exchange volume / initial balance’ < 25%). This indicator is:

    • Up from 90k wallets 12 months ago (75% annual growth rate from May 2019 to May 2020)

    • Up from 54k wallets 24 months ago (66% growth rate from May 2018 to May 2019)

    BAT tokens are also owned by different sized crypto-investors. On the graphic below, the legend splits holders in the following categories: 10 tokens < Micro < 100 tokens < Little < 1.000 tokens < Medium < 10.000 tokens < Big. As you can see, most investors own less than 1k tokens.

  • Dynamic wallets’ behaviors (labeled as ‘investors’ for a weekly balance evolution > 25%, and as ‘speculators’ for a weekly ratio ‘exchange volume / initial balance’ > 25%) are following a periodic pattern representing monthly advertisement payments.

    • Currently around 15k dynamic wallets on a weekly basis, up from 2k in June 2018 (650% growth rate in 24 months)

    • Clear sign of adoption and network usage beyond trading around exchange platforms (of which wallets have been excluded here)

  • The wealth distribution within Brave network has evolved with the different market cycles. The graphic below shows the percentage of wallets within the community owning 90% of all tokens in circulation (out of exchange platform’s wallets).

    • During the 2017 bull market, the wealth concentration has significantly decreased, reaching almost 25% for wallets owning 90% of tokens in circulation.

    • The 2018-19 bear market brought a break in this trend with a consolidation of wealth on a smaller part of the community, reaching the 10% level in late 2019.

    • Wealth concentration is now back to a decentralized trend with around 15% wallets holding 90% of tokens in circulation out of exchange platforms’ wallets.

  • Out of any behavioral labels, more than 700k wallets have interacted with BAT tokens since its inception. Even though it appears as one of the bigger communities around a utility token, it represents only a portion of Brave’s browser active users, estimated at 13.8M at the end of April.

    • Around 550k of these are holding less than 10 tokens, within which Brave users who hold tokens’ compensations after having viewed advertisements.

    • Lately, more than 50k wallets have been active monthly, up from 7k in May 2018 (+600% growth in 24 months).

    • This contrasts with the 1.7M Brave’s users that are said to have an active wallet.

    Information note on Brave: Brave is a privacy browser. They do not want users to be identified by on-chain transactions. In browser, wallets and transactions happen off chain. We are not able to see these transactions in our network analysis. Brave integrates with Uphold as their exchange partner. If a user withdraws or transfers their BAT from Uphold, an on-chain transaction will be recorded.

  • Monthly on-chain volumes are back around the 1B tokens level lately, with a transaction’s mean value around 10k tokens. With more than 100k transactions performed each month recently, Brave network is on its way to achieve a new activity ATH by the next few months.

Data and insights provided by Nyctale (nyctale.io).


Subscribe to Our Network for on-chain metrics every week:

About the editor: Spencer Noon leads investments for DTC Capital, a fundamentals-focused crypto fund. He actively tweets about on-chain metrics.

Our Network: Issue #22

Coverage on WBTC, Uniswap, Kyber, dYdX, and 0x.

Click here to join the Our Network telegram chat. 

Welcome to Issue #22 of Our Network, the newsletter about on-chain analytics that reaches almost 2000 crypto investors every week.

This week our contributors are back covering DeFi:

  • 0x

  • dYdX

  • Kyber

  • Uniswap

  • WBTC

I am an active DeFi investor and user myself, and something I have been thinking a lot about lately is just how quick this ecosystem is maturing. It’s a lot faster than people realize. A year ago, I would have agreed with many of the skeptics who said this is a space for hobbyists and tinkerers. But fast-forward to today and the signs that we are crossing that chasm to mainstream adoption are becoming evident:

And the list goes on…

As one of my favorite DeFi investors put it to me recently: “The reason there is close to $1 billion in DeFi [and billions in stablecoins] is not because users are playing around. It’s because they are making serious money.”

I couldn’t agree more.

-Spencer


Network Coverage

🟢 WBTC

Contributor: Alex Svanevik, co-founder of Nansen and D5

  • Earlier this week, Nexo took the final step and locked up their remaining 997.4 WBTC (~$9.2m USD) in Maker, as expected. And just yesterday, Nexo minted another 1.5k(!) WBTC - beating their last record. This latest mint ultimately made it into Maker as well. The result is a massive spike on the "WBTC in Maker" chart:

  • These transactions place Maker firmly at the top of the list of WBTC-holding addresses. The Maker WBTC contract now has a balance of 2,626 Wrapped Bitcoins, almost 14x higher than #2 on the list (Set Protocol). As you can see in the table below, WBTC is predominantly held by DeFi-related contracts.

  • In fact, 68% of the total WBTC supply is now locked in Maker! The chart below shows the total WBTC supply over time - currently at 3,851 WBTC:

  • Finally, the DAI minted against this WBTC has maxed out the debt ceiling for WBTC. At the time of writing, it's sitting at ~95%.

    Sources used in this post: Nansen, Dai Stats

🟢 dYdX

Contributor: Brock Elmore, Co-Founder of Topo Finance

  • dYdX recently launched their BTC perpetual contracts. These contracts operate on the Ethereum blockchain and this system is the first non-custodial way to get exposure to BTC price movements in the Ethereum DeFi ecosystem. It was in private beta prior to launching publicly on May 13th. The protocol offers up to 10x leverage. Since launch, the protocol has seen good volumes, with daily volume exceeding $12M in a single day of trading. That pushed dYdX to the top of DEX volume.

  • Average trade size is always a good metric for an exchange. dYdX shines here with liquidity from their off-chain orderbook/on-chain settlement system. Post-launch, they saw average trade sizes of up to $7k.

  • Trades per day naturally follows volatility, but dYdX has seen on average around 400 trades per day since launch.

  • Perpetual contracts use a funding rate to help balance longs/shorts to minimize risks in case of a large market move. This works similarly to BitMex, except interest is paid every second (but does not compound), unlike BitMex where they sweep funds earned every 8 hours. This rate is quoted at an 8-hour return percentage. During high volatility, the book was unbalanced (predominantly long) so you could earn 0.2% in 8 hours if you were willing to go against the market. One twitter user suggested short 10x on dYdX, long 10x on BitMex to capture the funding rate arbitrage (ignoring liquidation risk here).

🟢 Kyber Network

Contributor: Deniz Omer, Head of Ecosystem Growth at Kyber

  • While Automated Market Makers are a great hands-off approach to supplying liquidity, they are not capital efficient when it comes to volatile assets (ie all non-stablecoins). Professional market makers have an advantage in this space as they can deploy their own unique algorithms with customized trading strategies to provide far more dynamic pricing and this reflects in the data as professional market makers with their more competitive pricing facilitate 63.5% of volume passing through Kyber.

  • The march of the stablecoins continues and they now make up a whopping 63.7% of all token volumes on Kyber. We see this increase in volume in non-speculative assets as a clear sign of Ethereum’s increasing adoption as a platform for decentralized finance and part of a narrative of maturing lending, borrowing, and margin trading products. A year ago stablecoin dominance on Kyber stood at roughly 40% with SAI making up most of that volume.

  • Kyber Network continues to attract a diverse range of dapps for their liquidity needs and in April, 71 different dapps and services pulled liquidity from Kyber. After KyberSwap’s web/iOS/Android offerings, DeFi dapps like 1inch.exchange and DeFi Saver and popular wallets including MyEtherWallet and Trust Wallet made up the largest consumers of Kyber’s liquidity last month.

  • 1inch.exchange increased its dominance of Kyber's DeFi volumes over the last month although it must be noted that this chart excludes volumes from half a dozen other DeFi integrations (including Set Protocol) because they do not have a registered wallet address to receive protocol fees which is the primary way we calculate DeFi volumes. For the first time in over a year Fulcrum was absent from the DeFi data as the team works on an updated protocol for their relaunch.

🟢 Uniswap

Contributor: Caleb Sheridan, co-founder of Blocklytics

  • Yesterday, Uniswap v2 launched! Congratulations to the Uniswap team! One more reason to celebrate the launch is the opportunity to see Uniswap (the platform) go through an upgrade process which Uniswap (the team) cannot force on users. There is no off switch or upgrade lever -- Uniswap v1 will run as long as there are traders and liquidity providers. So far, Uniswap v2 has attracted $11m in liquidity and $1m of daily trading volume (source).

  • The team's launch and migration efforts have migrated over 20% of liquidity so far. For reference, Black Thursday saw the platform's liquidity drop 13% (in ETH terms) as LPs liquidated their positions.

  • So far, the biggest Uniswap v2 markets are simple migrations from v1:

    1. WETH-REP ($5M)

    2. WETH-DAI ($1.1M)

    3. WETH-USDC ($660K)

    Uniswap v2 also allows for more exotic token to token pairs. Notably the much anticipated stablecoin markets have started to build liquidity but are far from generating meaningful volumes:

    1. USDC-DAI ($110K)

    2. USDC-USDT ($40K)

    3. cUSDC-cDAI ($23K)

  • Since our last update in early April, Uniswap v1 has seen the following developments:

    • HEX became the deepest market

    • DAI liquidity ~doubled

    • sETH liquidity ~halved (likely due to new incentivization mechanisms launching)

    • REP received a $5M liquidity injection (which was subsequently migrated to v2)

    • UMA grew a $2.5M market after a very controversial IUO (Initial Uniswap Offering)

  • Finally, you cannot get this track on Spotify. Grammy-award-winning artist RAC released $TAPE as a limited edition good on Uniswap. As you can expect from similar limited-edition physical goods, the token's price performed extremely well increasing from $20 to a high of ~$930. RAC revealed that he has "actually been a liquidity provider for a while" (source).

🟢 0x

Contributor: Alex Kroeger, Data Scientist at 0x

  • In an effort to make metrics about 0x more broadly accessible, editable, and auditable by the DeFi community, the 0x core team has begun developing visualizations in Dune analytics. Dune analytics is a tool for querying a rich set of Ethereum blockchain data. The data in this post will highlight items from the 0x Staking dashboard on Dune. Check it out, fork the queries, and stay tuned for more!

  • To recap previous posts, market makers using 0x receive rewards in proportion to the fees generated from fills of their orders and the amount of ZRX staked against their addresses. If the market maker does not own sufficient ZRX to maximize their payouts, they can form pools to accrue stake from other ZRX holders in return for sharing a portion of their rewards. The proportion of rewards that are shared with pool members is set by the market maker and can be adjusted upwards.

  • In setting the percent of rewards to share, the market maker faces a maximization problem. A larger percentage of rewards shared decreases the market maker’s share of the eventual reward pot, but it also can make their pool more attractive to potential stakers, and more ZRX staked means the reward pot will be bigger. The latter effect depends on how attractive the pool is relative to other pools, so we should expect to see pools compete on rewards shared. Indeed, we’ve seen many pools increase the proportion of rewards shared to pool members, as shown in the chart below. Of the top 5 pools, Volleyfire has set this value most aggressively at 33%.

  • These changes have affected the way stake is allocated across pools. Dust Pool’s increase of rewards shared from 5 to 10% on March 24 helped solidify their position as the top staked pool at the time. Likewise, Prycto’s increase in rewards shared from 5 to 12% on April 22 preceded with their re-taking of the top spot.


Subscribe to Our Network for on-chain analytics every week:

About the editor: Spencer Noon leads investments for DTC Capital, a fundamentals-focused crypto fund. He actively tweets about on-chain metrics.

Our Network: Issue #21

Coverage on the Bitcoin Halving and L1 Networks.

Interested in discussing on-chain analytics? Click here to join the Our Network telegram chat. 

Welcome to Issue #21 of Our Network, the newsletter about on-chain analytics that reaches almost 2000 crypto investors every week.

This week our contributors cover L1 networks, and while we’re on the subject, I wanted to flag a video from a recent panel I moderated at the ReadyLayerOne conference about valuing L1 tokens:

The discussion ended up being highly engaging, informative, and fun—I highly recommend checking it out. The investor community has not reached consensus quite yet on how to value L1 networks, but it has made significant progress over the past few years.

Now let’s dive into this week’s coverage!


Network Coverage

This week our contributors cover the following networks:

  • Bitcoin

  • Cosmos

  • Decred

  • Ethereum

  • Tezos

🟢 Bitcoin

Contributor: Nate Maddrey, Research Analyst at Coin Metrics

  • Bitcoin experienced its third halving on May 11th, with block rewards dropping from 12.5 to 6.25. This marks a shift into a new era for Bitcoin, in more ways than one. Miner revenue, which is composed of block rewards and transaction fees, is critical to the overall security of the Bitcoin network. The more revenue, the larger the incentive for miners to try to win a block and claim the prize. This competition between miners ultimately keeps the network secure.

  • Up to this point, a large majority of Bitcoin miner revenue has been generated from block rewards. Over the last five years, only 4.38% of miner revenue has come from transaction fees, on average.

  • The recent halving decreased Bitcoin’s annualized issuance percentage to about 1.5%. In the long-term, lower block rewards will be a good thing for Bitcoin, as it means less supply dilution. But it also means less miner revenue, at least in the short-term.

  • Over the long run the Bitcoin network will need to become more reliant on transaction fees to make up for that lost revenue.Interestingly, Bitcoin fees have been surging in the few days after the halving. Daily transaction fees hit $1.05M on May 13th, the highest daily total since July, 2019. A sustained surge in transaction fees would be good for miner revenue, and an overall positive sign for the long-term health of the network.

🟢 Cosmos

Contributor: Chjango Unchained, Director of Community at Cosmos

  • From 1 to 4 teams: What was once a single-celled 'organism', the Cosmos core development team has divided into multiple 'organisms'. In other words, the Tendermint team multiplied last March into these distinct teams with distinct responsibilities:

    1. Tendermint Inc is responsible for the Cosmos SDK and ecosystem development of Cosmos.

    2. Interchain Foundation manages grants to teams building infrastructure towards the Cosmos vision.

    3. Interchain GmbH maintains Tendermint Core and leads the development of Interblockchain Communication (IBC), the flagship chain-agnostic protocol for communicating value and data across blockchains.

    4. Iqlusion is responsible for running Game of Zones.

    I know—it's a lot to take in. In short? This is unprecedented in the industry. The implication for Cosmos? This move arguably decentralizes the core development team by having multiple entities working on the same project, but without a core team intact, it inevitably increases coordination friction. But isn't enabling social coordination what blockchains are all about?

  • Game of Zones was initiated in the first week of May. There were 183 registrants with ~130 who successfully spun up testnet chains to make IBC connections at the beginning of the games. Teams and connections visualizer can be seen below (source). However, after a software upgrade, a fatal bug crashed the incentivized testnet game. A postmortem of the event can be read here. The next phase of Game of Zones is restarting May 18th after the bug fix is merged into the software.

  • The peg zone, or bridge space is a-booming! ChainSafe announced yesterday that it's launching ChainBridge, which will function as one of the several 2-way peg zones between not only Cosmos and Ethereum but for many other related blockchains. This peg zone model builds from the initial Peggy architecture that we laid out for Cosmos in early 2018. Many dev shop projects have since picked up the torch on building these chain bridges. ChainSafe's ChainBridge is one of many Cosmos <> Ethereum 1.0 bridges that will coming down the pipeline soon. Althea is launching their version of the peg zone, or 'Peggy', and they are "optimistic" about deploying one in the near future (full story here).

    And that's not all there is for the bridging economy. Swish Labs is coming out with their own version of Peggy, giving this cross-chain network at least 3 bridges to start with by the end of 2020. If one were to analogize these peg zones to toll roads, then this multi-bridge system gives the user many different IBC options and fee models.

  • Cosmos governance is on its 25th proposal—and a rather significant one, at that. Prop 25 seeks community approval for CosmWasm to be integrated into the Cosmos Hub. If passed, it will enable faster feature integration for the Cosmos Hub post-IBC launch and will introduce a constrained form of smart contracting on the Cosmos Hub. Voter turnout is so far at 32% with a 99% Yes vote (source).

🟢 Decred

Contributor: Checkmate, Decred contractor

  • The Decred blockchain has a consistent baseload of demand for block-space, a result of the PoS ticket system and, more recently, on-chain CoinJoin privacy transactions. As such, the Realised Price metric differs in interpretation to Bitcoin. A strong conviction Decred holder actually has a regular and frequent on-chain signature moving DCR as opposed to the equivalent of long periods of dormancy for Bitcoin.

    The Decred Realised Price tends to follow the spot price quite closely, however lags behind the day-to-day fluctuations in off-chain price on exchanges. The chart below presents an experimental metric that takes the 28-day gradient of the Market Cap and Realised Cap, and produces an oscillator from their difference (purple). This tool distills times where off-chain price momentum bias flips before the on-chain response is observed. This is due to DCR being bound in tickets and held in cold storage wallets which take time to transact. Where the oscillator crosses the zero level, it often precedes a shift in price momentum in the direction of the flip.

  • As noted, Decred has a consistent transaction demand which also shows up as reliable NVT and RVT signals. These metrics take the ratio between network valuation (market cap or realised cap) and the adjusted daily transaction value flowing through the chain, all denominated in USD. The chart below presents the NVT and RVT both in 28-day and 90-day moving average format with sound agreement in trend and magnitude between all.

    During periods of bullish sentiment, we can observe low NVT|RVT ratios indicating that the chain is settling a substantial value relative to its network valuation, and vice-versa indicates bearish sentiment. Of particular interest is the period of strong demand for on-chain settlement since Aug 2019 at which point the CoinJoin privacy mix server came live. This provides valuable feedback for the community and developers regarding actual demand for the mixing service, and also gives miners a new basis for future fee market expectations.

  • Digging into transaction demand further, the area chart below shows the cumulative DCR settled on-chain through protocol history, divided into regular transactions (orange), ticket purchases (green) and CoinJoin mixes (red). The line charts to the right axis presents the daily transaction volume in DCR for ticket purchases and CoinJoins.

    The gradient of the area plot has steepened since the privacy mix service went live, confirming increased demand for block-space. There has been a steady uptick in DCR flowing through the anonymity set with around 110k DCR mined in CoinJoin transactions per day. This represents around 0.96% of the total circulating DCR supply in CoinJoins, and is substantial when compared to the 192k DCR that are mined into tickets daily (1.67% of circ. supply).

  • The Decred Treasury underpins the self-sovereign development of the protocol, and its accumulated value is subject to the market's pricing of DCR. To date, the treasury has spent a total of $7.625M bootstrapping the network from genesis to now when pricing each outgoing transaction on the day of the spend. This represents around one third of the incoming DCR so far and 16% of the total DCR inflows that will occur via the block subsidy ending in year 2140.

    Based on a current DCR coin price of $14/DCR, the Treasury is capitalized with enough USD value to build another Decred (assuming $7.625M build cost) and can repeat that metric for each $12 uplift in DCR price given the current Treasury balance of 636.3k DCR.

  • Finally, an interesting metric to gauge stakeholder governance power is to look at how much Treasury value is governed by each ticket in the PoS pool. The chart below presents the Treasury balance divided by the count of tickets in the pool (red), showing that each ticket commands decision making power of around 15.5 DCR. If we divided this by the purchase price of a ticket denominated in DCR (blue), governance power typically represents around 11% of the ticket value. Given tickets vote on average every 28 days, this means governance power on an annualised basis is equivalent to 143% of a typical ticket in value.

🟢 Ethereum

Contributor: Santiment

  • The Daily Active Addresses metric for Ethereum has improved steadily these past six months. After bottoming out in late January, it surged forward throughout ETH’s and many other altcoins’ great February respective runs. After Black Thursday, the DAA really exploded, and this in part has contributed to the enormous rebound from $85 over these past two months. Since this time, we have twice seen daily active addresses top out right on 361,000 on different days, both corresponding with price tops. We will see if its next attempt at this level can push past this fascinating barrier (source).

Image 2020-05-15 at 10.22.37 AM
  • Substantial amount of tokens were moved at ETH bottoms. In several cases over the past couple of months, the ‘Token Age Consumed’ metric spiked whenever bottoms formed on Ethereum. This metric indicates the amount of tokens changing addresses on a certain date, multiplied by the time since they last moved. Spikes on the graph signal a large amount of previously dormant tokens on the move. So with this in mind, it was clear that old-time ETH holders have been increasingly active. Many dormant tokens ended up making big moves these past few months, particularly around Black Thursday (where a single day had 413.6M ETH consumed). In general, this is a primary indicator to indicate a price direction shift is imminent, and we can see that spikes have correlated with local bottoms heading all the way back to last November (source).

  • ETH miners have switched back to accumulation mode. After ETH miners’ balances bottomed out on Feb. 19th, just as the 6-month ETH price high was taking place, the balances very quickly increased. Heading through Black Thursday in mid-March and all the way through April 12th, miners added about 50,000 ETH to their wallets in this one-month timespan. Once again, as they dumped their balances over the week following, ETH’s price hit a local top. But since May 9th, we’re starting to see an accumulation trend again beginning to form (source).

  • As we have noted in previous studies, the marketcap of DAI and Ethereum’s price tend to correlate in fascinating ways. Prior to ETH’s great price surge from late January to mid-February, we noted a steady increase in DAI’s marketcap. And again, even before its crash along with the rest of the crypto markets on Black Thursday, ETH’s price was foreshadowed by a slight dip forming in DAI’s marketcap on the days prior. DAI’s marketcap levels are now staying fairly level at a range of 108M to 111M (source).

Click here to read Part 2.

Our Network: Issue #21 (Part 2)

Continued from Part 1.

🟢 Tezos

Contributor: Alexander Eichhorn, Founder at Blockwatch Data

  • Staking: Tezos' network wide staking ratio reached a new ATH of 80.11% in mid May while staking yield fell to its all-time-low of 0.94% with an inflation rate sitting at 5% currently. This number reflects the true past inflation, i.e. total amount of coins generated over the past 365 days (39,77M tez) vs. the total supply that existed a year ago. Absolute inflation in Tezos is almost constant, so inflation rate slowly decreases over time. However, since the amount of coins generated per block is dynamic to discourage attacks the future inflation rate may fluctuate slightly.

  • Custodians: The trend towards custodial staking in Tezos is still unbroken, yet has slowed down significantly in Q2 so far. Coinbase gained another +13% month-over-month (+8.2M tez) and is now controlling 11.2% of the Tezos network consensus. The top 5 exchanges combined hold 18% of total supply in staking alone and more in non-staking wallets.

  • Growth: Tezos saw a growth of 35.5k new funded accounts (+8%) in April. The highest growth rate was in the balance range between 10 .. 100 tez (+3.7k, +30%), followed by 100 .. 1k tez (+3k, +21.5%). Notably, accounts holding 100k .. 1M tez jumped by +46 (+7.7%) to 636, indicating that many small investors and potentially few whales entered the ecosystem. The top 1k accounts still hold 63% of total supply, down -1% since last month. 7% of total supply (59M tez) was on the move last month while 78% of supply hasn't moved for more than 3 months now.

  • Loyalty: Tezos is all about delegated proof of stake and its community is very close knit so we were interested in visualizing the behavior of delegators over time. For brevity we only focus on two aspects, the number of new delegators per month (black bars) and the average time those delegators who switch baker or sell their stake actually keep their initial baker (blue bars). Each month represents an individual cohort of delegators. We see that a genesis wave of staking occurred during the first 4 months, then a second wave followed with the price surge in Spring 2019 and a third much larger wave followed in Nov 2019 when custodial staking opened. Long-term hodlers who dominate Tezos are very loyal to their bakers: out of the 35.5k currently active delegators 27.5k (77.5%) have never left. Over time only half of all delegators became disloyal at all and only 14% have switched baker multiple times. Delegators who leave eventually stick on average of one year, but with new cohorts this number is rapidly decreasing.

  • Adoption: After the Babylon testnet shut down on Apr 8, development activity moved over to Carthagenet. There has been less growth in developer activity in terms of deployed contracts and calls, but the rising gas usage suggests, the contracts that are developed are getting more sophisticated. Diversity in contract interfaces is also increasing. Highest development activity happens around oracles, tokens and games. Smart contract activity on mainnet is improving in terms of traffic and gas, but nowhere close to other chains yet. However, high value assets are getting deployed. One example is tzBTC, a wrapped Bitcoin on Tezos with currently 358 Bitcoin issued.


Subscribe to Our Network for on-chain metrics every week:

About the editor: Spencer Noon leads investments for DTC Capital, a fundamentals-focused crypto fund. He actively tweets about on-chain metrics.

Our Network: Issue #20

Updates on Aave, Balancer, Compound, and Set Protocol.

Interested in discussing on-chain metrics? Click here to join the Our Network telegram chat. 

Welcome to Issue #20 of Our Network, the newsletter about on-chain metrics and crypto analytics that reaches almost 2000 investors every week.


Network Coverage

This week our contributor analysts cover DeFi:

  1. Aave

  2. Balancer 🆕

  3. Compound

  4. Set Protocol

🟢 Aave

Contributor: Isa Kivlighan, Aave team

  • Aave has experienced remarkable growth since its mainnet launch in January. This dashboard shows Aave’s current market size ($63.5M), total amount borrowed ($11.3m), and TVL ($52.2m) of the protocol. TVL ranks #4 overall among DeFi projects on DeFi Pulse.

  • For a more granular picture of historical activity, Aave Watch provides a breakdown of deposits, borrows, repayments, and flash loans since Aave’s mainnet launch:

Image 2020-05-08 at 12.10.57 PM
  • One metric that Aave tracks is 3rd party referrals driving users to the protocol. Here we can see a breakdown of cumulative referral activity for deposits and borrows:

Image 2020-05-08 at 10.49.50 AM
  • This Dune Analytics dashboard created by Matteo Leibowitz shows the amount of interest paid per day broken down into each asset on Aave:

  • Here we can see the number of unique lenders per day since Aave Protocol went live. The daily all-time high occurred on March 13th with 219 unique lenders, but as we can see, that milestone is close to becoming eclipsed (source - dashboard created by Matteo Leibowitz).

  • Lastly, let’s take a look at protocol fees. The graph below shows the daily net origination fee per token. An origination of 0.25% of the loan amount is taken from every borrow transaction. Out of that 0.25% origination fee, 80% is taken to buy and burn LEND, decreasing the monetary supply of the LEND token (source - dashboard created by Matteo Leibowitz). Note: This chart does not take into account the Flash Loan fee of 0.09% .

  • Aave Watch also has a breakdown of the fees by asset which shows how much has been collected, how much is yet to be collected, and how much comes from Flash Loans. A 0.09% fee is collected from every Flash Loan. 70% of this fee goes to reward depositors, and the remaining 30% is split 80/20, with the 20% from that going to 3rd party integrators/referrals (source).

    As you can see, the majority of flash loans occur using ETH, DAI, and USDC, while USDC and sUSD have collected the largest share of fees on Aave.

  • Finally, data from our token burn page shows how much LEND is ready to be burned, how much has already been burned, and how much comes from origination fees and Flash Loan fees. As expected, the total LEND burned over time is steadily increasing.

🟢 Balancer

Contributor: Fernando Martinelli, Co-founder & CEO at Balancer Labs

  • Launched on the last day of Q1, Balancer has already surpassed $2.5M in liquidity provisioned to its pools. Balancer is a generalized AMM (Automated Market Maker) that enables a wide variety of configurations: up to 8 tokens per pool, distributed in any proportions (aka weights) of the pool’s total value, and any custom trading fee. The configurations of the pools created so far reveal users have been extensively experimenting with the protocol’s flexibility. As can be observed below, there’s great pool diversity (source).

Image 2020-05-08 at 10.20.02 AM
  • Inspecting the aggregate liquidity per asset across all pools, we see below that Balancer has attracted its first MKR whales (source - query created by Matteo Leibowitz). A plausible hypothesis is that pools with uneven weights (as is the case with the “75% MKR & 25% ETH” pool) can maximize the ability of opinionated portfolios in providing liquidity. These pools are compelling enough (desired exposure is maintained; lower ETH requirement; less impermanent loss) that previously dormant portfolios may finally decide to join the AMM arena. As a point of reference, within just one month since launch the 2.77k MKR on Balancer pools already account for 41% of the 6.69k MKR yup currently on Uniswap. Further analyses and simulations around uneven pools are explored here.

  • Another observation from the graph above is that tokenized BTC on Ethereum has shown a meaningful presence with some of its available flavors. This is mainly due to PieDAO’s BTC++, a Balancer pool comprised of BTC-pegged tokens that is managed by a DAO and showcases the Balancer protocol as a powerful building block in DeFi. BTC++ holders benefit from a proactive management of the pool, diversified risk and the revenue generated from trading fees between the underlying assets. The supply of BTC++ is around $450k and has been growing rapidly since a recent code audit by Quantstamp has been concluded. Other future Pies (PieDAO’s tokenized portfolios) such as USD++ are also expected to leverage Balancer, attracting further liquidity into the protocol (source 1, 2).

  • The use of AMMs for initial token listing was recently pioneered by UMA. While Uniswap as the chosen platform felt familiar to most Ethereum users, Balancer’s LBPs (Liquidity Bootstrapping Pools) could have arguably achieved a broader token distribution under a more inclusive and organized process.

    An LBP could start at 80/20 weights (80% project_token & 20% ETH) and gradually turn into a 50/50 pool over the course of a week (or any other desired period). With such parameters, the token ends up 4x cheaper (a 75% discount) than its initial price, so the team can set a very high starting price (no FOMO) and let the token slowly devalue, similarly to a reverse auction. Participants may buy the token at any point in time (whenever they individually find the price attractive), effectively counteracting the downtrend and hence contributing to a much smoother price discovery process. Whales are still welcome, but not at the expense of excluding retail users. More about LBP mechanics and possible configurations are explored in this blog post.

🟢 Compound

Contributor: Lucas Campbell, Growth & Research at DeFi Rate

  • In the past 6 weeks, Compound has experienced over $140M in cumulative deposit volume - averaging roughly $20M per week. The most notable contributors to these volumes are DAI, ETH, and USDC, which account for approximately 98% of all deposits to the money markets protocol. Leading deposit volumes is Dai, aggregating ~$56M in total deposits in the past six weeks. This is followed by ETH and USDC which represent for $42.8M and $38.2M in cumulative volumes, respectively. The next closest asset is WBTC which totaled for $1.3M. Lastly, USDT was added as a supported asset in late April. In its first two weeks, the fiat-backed stablecoin has accumulated nearly $600,000 in total deposits.

  • Looking at borrowing activity, Compound aggregated around $36M in total volume over the past six weeks, averaging ~$5.2M per week. Borrowing volumes are dominated by Dai as the crypto-native stablecoin has accounted for ~80% of all borrowing activity in the past six weeks. The next closest contributor is USDC which accumulated over $5M in total volume or around 13% of cumulative borrows. ETH is the only other asset on the protocol aggregating seven figures in borrowing activity, reaching $1.1M in volume since late March. WBTC was close behind with $818K while all other assets (BAT, REP, ZRX) totaled for ~$121K in volume. Lastly, USDT has averaged $72.6K in borrowing activity in its first two weeks on the protocol - making it the 5th most actively borrowed asset on Compound.

  • Compound users have earned ~$61,000 in interest in the past 30 days, distributing around ~$2,000 per day to depositors. Despite the fact that interest rates are at historic lows - 1.16% APY on USDC and 0.77% APY on DAI - USDC and DAI still account for 95% of all interest accrued on Compound this month. The addition of USDT should also serve well in the coming future as Tether currently has the highest APY relative to the other stablecoins, offering 2.75% APY on deposits. What’s interesting to note is that USDC has "flippened" DAI in terms of interest accrued per day for the second time this month as USDC is now distributing ~$840 per day compared to DAI's ~$350.

  • Last week we saw the first two on-chain governance proposals go live on Compound. While COMP tokens are still restricted for public access, there are currently ~36 delegates with voting power. The first proposal was the more controversial of the two with voters ultimately electing to support USDT on the protocol. With 24 delegates weighing-in on the proposal, 90.3% voted “For” while 9.7% voted “Against”. Notable voters on the “Against” side include prominent community members Ryan Sean Adams, Ric Burton, Zerion, and DeFi Rate.

  • The second proposal was submitted by Dharma - a DeFi application that relies on Compound - to change the interest rate model in light of the current landscape with the Stability Fee and Dai Savings Rate (DSR) at 0%.

    This proposal was less controversial than USDT as 97.29% voted “For” while 2.71% voted “Against”. IDEO CoLab Ventures was the only major delegate to vote “Against” which is cited to be due to Dharma's submission process and not the substance of the proposal itself. Expanding on this, Dharma went ahead and submitted the proposal on-chain without any discussions or signals from the community beforehand. This is generally not in line with best practices for decentralized governance and IDEO elected to vote “No” in order to highlight the importance of setting standards for long-term sustainable governance.

🟢 Set Protocol

Contributor: Anthony Sassano, Product Marketing at Set Protocol

  • This new graph on our Scout dashboard shows the overall sentiment of both Robo Sets and Social Trading Sets on the TokenSets platform. You can see that during March, the majority of the Sets were bearish (holding USDC and cUSDC) whereas during April, they started flipping bullish and quickly reallocated. Today, the majority of the collateral in the Set Protocol Vault is WETH, WBTC and LINK which points to an extremely bullish sentiment for those 3 assets.

  • In early April, we relaunched the majority of the Social Trading Sets because we added the ability for traders to set performance fees and streaming fees. Since then, Set Protocol's annualized earnings has exploded with it now sitting at $43,000 after just 5 weeks worth of performance and streaming fee accrual for traders. We expect this growth to continue as traders build out more performance and get additional inflow into their Sets.

  • Set Social Trader Aaron Kruger recently started building out TokenViz - a website that visualizes various metrics using Set Protocol/TokenSets data. One of the sections of the website allows you to measure each Set against all possible metrics and plots this data in a top 10 list. Below, I've plotted each Set by USD market cap and you can clearly see that there is a Pareto distribution forming among Sets based on performance. The top 3 Sets listed are the best performing Robo Sets against ETH over the last 3 months with the RSI Set returning 66.2%, the 20 Day MA Set returning 50.7%, and the 26 Day EMA Set returning 44.4%.

    TokenViz has also already become TokenSets' #4 source of referral traffic over its short life of around 2 weeks!

  • Due to the out-performance of many of the Sets over the last few weeks as well as overall price appreciation, the Set Protocol Vault has grown from $4.3 million on April 1st to $17.5m at the time of writing which places Set Protocol at #8 on DeFi Pulse. As for where this value is allocated: $3m is currently positioned in the Social Sets with $14.5m positioned in the Robo Sets. We expect this to become more equal over time as the Social Trading Sets build up more of a performance history.


Subscribe to Our Network for on-chain metrics every week:

About the editor: Spencer Noon leads investments for DTC Capital, a fundamentals-focused crypto fund. He actively tweets about on-chain metrics.

Loading more posts…