Band Protocol: A Dual Token Model for Decentralized Oracles
An analysis and valuation on the recent Binance Launchpad project
When it comes to blockchain products reaching a mainstream audience, oracles are commonly mentioned as one of the pivotal solutions necessary for adoption. For those unfamiliar, an oracle is a third-party information source that supplies data to blockchains that can trigger predefined actions in smart contracts.
To provide some context, oracles are a tool used to allow smart contracts to read and incorporate a variety of real-world data such as sports scores, temperatures, winning lottery numbers, flight delays or stock price(s). In doing so, smart contracts can execute specific tasks such as paying out a given address $X amount of DAI if the Patriots win the Super Bowl.
With this in mind, it’s important to recognize that oracles allow blockchain systems to utilize data that were previously incompatible with smart contracts. This means that the breadth of use-cases that automated systems can accommodate is drastically increased.
For this week’s analysis, we’ll be taking a look at Band Protocol, an Ethereum-based solution that incentivizes the creation and flow of valuable information through decentralized oracles.
Seeing as oracle providers must execute code in a trustless manner, there’s often a degree of centralization required to ensure external data points are valid. As a result, many decentralized systems suffer from delays in throughput or increased transaction costs in order to validate and verify external data before pushing it to the contract requesting the information.
Band Protocol looks to bridge the gap between decentralized applications and real-world data by ensuring that data is both valid and trustworthy through economic incentives.
In comparison to an oracle project like ChainLink, Band’s functional product provides a strong foundation for us to make logical assumptions about the token’s future. As it currently stands today, the LINK token economy is theoretical, meaning that we’ve chosen to forego covering their interpretation of a decentralized oracle economy until tangible utility is actually live.
Token Economy Overview
Band uses a dual-token economy to segment and incentivize unique data sets. This approach ensures that providers of any given set are compensated properly in the event that their information becomes more valuable than others. The token economy can be broken down into two unique assets: dataset tokens and the native BAND protocol token.
Datasets are split into multiple Dataset Governance Groups, each of which utilize a unique token to stake, curate and govern information through mechanics like a Token-Curated Registry or Token-Curated DataSource.
Without diving too deep, Token-Curated Registries (better known at TCRs) allow for different addresses to be whitelisted to participate in the respective ecosystem. In the context of Band, this means that a given address can be verified as being trustworthy or marked as having the ability to vote on any governance issues that arise (such as a dispute in the finality of a specific data point).
In short, dataset token holders can become a "data provider" who is responsible for curating high-quality data that is fed to smart contracts, purchase data from a dataset or vote for data providers they trust to contribute valuable (and valid) information.
Why Curate or Participate?
Every time data is sold to a smart contract, rewards are distributed to providers pro-rata, with subsequent voters receiving a fraction of the reward. In short, data providers stand to earn an income by contributing valuable data whereas participants stand to earn from voting for that provider. Furthermore, competition from multiple datasets in similar verticals further drives the quality and demand for any given dataset.
While the data governance tokens are unique, they are all secured by Band’s native protocol token, BAND, through a bonding curve.
BAND provides liquidity to data governance groups by serving as a primary value capturing mechanism in a few different ways. First and foremost, BAND is used as collateral to issue dataset tokens. As such, BAND token value is directly correlated to the value of the protocol’s datasets.
As with many other strong token economies, BAND will be used to govern future protocol upgrades. In short, community members must acquire BAND to signal their opinion on changes to the protocol as a whole.
Next, BAND is used to control dataset quality through a curated dataset registry. Similar to how an address can be whitelisted on a given dataset, Band Protocol maintains an active list of all verified datasets. In order for a dataset to join the registry, they must stake BAND as a proxy.
Lastly, Band Protocol has mentioned that they will eventually migrate to BandChain, in which BAND will be used as a staking mechanism to secure the network via a Delegated Proof of Stake (DPos) consensus mechanism. Similarly, BAND will be used as the underlying currency for all that transactions that happen on the network.
Band Protocol relies on query fees to cover the cost of data providers and incentivize honest data curation. Whenever a smart contract issues a data query function call, it must attach a fee denoted in a blockchain’s native currency (ETH in the case of Ethereum). Query fees are split among providers and stakers based on a fee schedule determined by a group’s Governance Parameters.
Band Protocol utilizes Uniswap to instantly convert the accepted currency (ETH) to BAND, which then gets converted to the dataset token through a purchase on the bonding curve.
By doing this, regardless of what currency a dApp pays in, data providers and token stakers still receive revenue in their unique dataset token. Taking this a step further, this means that as the system acquires more query fees, more BAND will be locked in dataset bonding curves.
This past September, Band Protocol was announced as one of the premier projects to undergo a Binance Launchpad sale, raising $5.85M in exchange for roughly 12% of the total supply. At the time of sale, the base rate for BAND was set at $0.473, signaling a diluted market cap of around $50M.
Shortly after the fundraise, Band Protocol went live on Ethereum main-net, meaning that the system is up and running today. BAND is currently listed on Binance, Bilaxy, and Dcoin with the dataset for Financial Data Token operational here.
According to CoinMarketCap, there’s currently 15,662,706 BAND currently in circulation. With a large majority of these tokens stemming from the Binance Launchpad Sale, there’s still an additional 84,337,229 BAND left to enter into circulation over the next few years. Here’s a look at how the supply will be diluted in the coming years:
Graph via Binance Research
According to Etherscan, only 344 unique addresses currently hold BAND with the majority of the tokens still being held in the Band multi-sig wallet. While this seems rather small, the amount of individual holders is well above the listed addresses. During the Binance Launchpad sale, a total of 16,321 out of 35,579 participants won the lottery draw while the rest of the 19,258 participants received tokens from the 631,800 BAND airdrop (~32.80 per participant).
Chart via Etherscan
As we mentioned above, query fees serve as Band’s primary revenue stream. Whenever there’s a data query function call, a small fee is attached. With this, query fees get split among the dataset’s data providers and token stakers based on certain parameters dictated by the protocol’s governance.
As noted above, while users pay the query fees in ether, the protocol leverages Uniswap to automatically swap the ETH for BAND, deposits that BAND into the bonding curve, and distributes the respective dataset tokens to the stakers. This allows us to assume that all query fees are converted to BAND.
As it currently stands, the initial cost per data query is set at 0.002 ETH ($0.40 at $200/ETH). The first dataset, Financial DataSet Token ($XFN) is live on Ethereum main-net. In addition, dApps like BitSwing have generated $11,500 USD equivalent in query fees as of September 30th. That being said, having launched less than a month ago, we’re still in the very early stages for BAND.
In this section, we’ll explore potential valuations for BAND by using a standard DCF model and the expected “cash-flows” from query fees. This model is similar to the ones found in our earlier Token Tuesday’s like REN and SNX. The equation for calculating staking revenue can be seen as follows:
Equation via Medium
While this formula calculates the per staker revenue, we’re looking for total network revenues for stakers.
Basic assumptions in this equation include:
Query Fee = 0.002 ETH
Circulating Token Supply: Based on Release Schedule
There are two main variables in the ecosystem economics that we can make basic assumptions on to evaluate the token price.
Assumption 1: Number of Queries
This was the hardest aspect surrounding the valuation. The number of queries could vary greatly in the coming years, especially depending on the adoption of DeFi and other applications that rely on oracles to operate. With that being said, we’re extremely bullish on the proliferation of DeFi and similar products that need oracles in the next few years. Ethereum-based DeFi applications are at the very beginning of the growth curve in terms of total value locked. With this, the number of queries per day will see parabolic growth as DeFi continues to grow with it. As such, we estimated that as the Band Protocol continues to release new datasets and more applications rely on BAND for data queries, we could see upwards of 500,000 queries on a 24-hour basis with further growth over the next 5 years.
Assumption 2: Number of Providers
Band is a decentralized oracle service and relies on data providers to update the relevant data stream. As Band continues to expand into other dataset pools, the number of data providers will gradually increase to further decentralize the protocol. As it stands, there are currently 7 data providers in the XFN dataset with more coming in the future.
In summary, with these two main assumptions in place, we can calculate 24-hour and annual staker revenue with respect to the circulating supply. With a fully diluted token supply, BAND stakers could expect up to 0.0006 ETH per BAND token in annual revenues (denominated in dataset tokens) in the coming years.
The last aspect of our model is to assume a discount rate on the token. In typical venture capital and early-stage investing, a discount rate between 30-50% is the norm for many startups. As such, we assigned BAND with a moderate discount rate of 40%. The math for calculating present value can generally be seen as
PV = (Year 1 Cash Flow /1+r)^1+ (Year 2 Cash Flow /1+r)^2…+(Year n Cash Flow / 1+r)^n
To calculate the terminal value (TV) and factor in the number of fees paid after the five year period, we used the Gordon Growth Method and took the sustained growth rate of 2% which is standard for most mature companies.
By plugging in our numbers, we calculate that the fair value of BAND lies around 0.0027 and 0.0039 ETH per BAND token. At a current rate of $175/ETH, this represents a price between $0.48 and $0.69 per BAND and a fully diluted market cap of $48.25M USD. As a comparison, the leading decentralized oracle protocol, ChainLink, currently trades at $2.6 billion with a fully diluted market cap.
With BAND currently trading at $0.36 and a market cap of $5.6M, the fair value price represents between a +33.3% and +91.6% upside on the token depending on the dilution of the supply.
It is important to note that this number represents what BAND should be trading at right now, and could be seen as a potential upside in the short-to-medium term. Any fundamental changes could change the valuation model and doesn’t take into account the “FOMO” effect that crypto is notorious for.
For a full breakdown of our assumptions, feel free to review our spreadsheet here.
As we mentioned towards the beginning of the article, Band faces an uphill battle due to the brand that ChainLink has developed in the past year. With ChainLink announcing support from Google, it’s clear that the decentralized oracle market is shaping up to be an extremely competitive one to say the least. Beyond ChainLink, there are a number of other decentralized oracle providers, all with a similar mission to drive value through the curation of high-quality data.
With this in mind, the value of BAND token is entirely reliant on the growth of Band Protocol as a market leader for oracle information. While Band has displayed an exceptionally clean UI/UX supplemented by a robust token economy, it’s all for naught if there is no demand.
Finally, as it stands today, there are few dApps that need oracle information to scale. While this is highly likely to change in the coming years, it’s worth noting that until a killer dApps emerges that relies on oracle data to succeed, Band’s potential market is limited.
In summary, we believe that Band Protocol has displayed a number of strong signs for future growth. In their main-net announcement article, CEO Soravis Srinawakoon mentioned that Band will be looking to explore DeFi partnerships, signaling that their mind is in the right place regarding engagement from the Ethereum community at large.
As blockchain applications continue to evolve, there’s no doubt that oracles will see increased demand as smart contracts become more complex. We at Fitzner Blockchain are excited to watch how this market unfolds and to see whether or not a robust token economy such as that of Band Protocol will be able to capture value as it was designed.
*DISCLOSURE: The managing partners do not endorse or recommend any investment action in BAND. This document should not be regarded as investment advice, offering documents, or as a recommendation regarding a course of action. The managing partners of Fitzner Blockchain own BAND. These views are those solely of the managing partners of Fitzner Blockchain Consulting and do not represent the views of the Band Protocol team.
Token Economics: https://bandprotocol.com/features/dual-token
Binance Report: https://info.binance.com/en/research/BAND-2019-09-06.html