NFT
NFT Theory, Practices and Valuation
Date of Publication:
1st Edition Apr 2022Jan 2023 Updated
Athor:
Dr Liang Han (ACA)
Key Guidance
Preface: This article might be a bit wordy and some parts of the analysis might be harder to understand, but I hope you enjoy it.
Useful Websites
Not recommended to buy from OpenSea as fees charged are high, but we may use it to list.
Exchanges:
CEX: https://looksrare.org/explore****DEX: https://blur.io/
Analytics and Monitoring:
**Dune:**https://dune.com/sealaunch/blur****
Lend and Borrow:******Ethereum: **https://www.benddao.xyz/
Yield Farming:******Best liquidity: **https://nftx.io/
** **
Table of Content
Chapter 1: Overview of the Universe of NFT
Market update and sentiment
Simplistic NFT Market forecast
Is the odds (NFT issuance) against you?
How to obtain an NFT and its risks
Chapter 2: Understand the value of the NFT
The architect of systemic predictive model
What are the main groups of interests?
What could be an appropriate valuation model？
The Case for Licensed Commercial Use Rights
A bit about Tokenomics: Bonding curve
The importance of creating “Events”: The case study of moonbirds
The economics behind Staking and burn
Was it actually about Membership or utilities
The smart use of Breeding: The case study of Nike
Statistical evaluation of the film star model** **
Chapter 3: Defi Solutions on NFT
Technical background of yield farming
Workshop: recalculate APR for NFT
Workshop: using twitter or facebook data to predict valuation
Rarity: one attribute of impressive or a proxy of impressive
** **
Chapter 1: Overview of the Universe of NFT
1.1 Market update and sentiment
As at 05 May 2022, the Blue Chip Index (NFTgo) appears to be at a reasonably good momentum. Recently price index moves were also negatively correlated with mainstream crypto currencies.
If we take another overview on market cap, it’s not hard to find out that the $19.52Bn total market cap from NFT alone is about 1.1% compared with $1.7Tn on the whole of the Crypto capitalization.
Historically, it took about 2 years for Ethereum to double its market cap ratio and currently stands at 1/3 of BTC’s. A straightforward mind calculation will probably give us 6% out of the total market cap in 2 years. One can probably use it as a rough benchmark as a starting point for any further analysis on individual NFTs, e.g. one way to think about is whether the floor price of one NFT collection can climb quicker than 3 times in 2 years as new issuance will also bump up the market stats?
1.2 The Odds is against you?
In terms of category, PFP still has the largest community by far.
The top 10 NFT collections account for 35% of the market. It appears to be a sign of a young market that is not yet saturated.
According to IntoTheBlock/Twitter, there are around 80,000 collections in total, the average gain is 65%*$20,000 Mil /80,000 = 0.1625 Mil, with a 2/3 chance to “become” average or better. So the expected gain is about $100K. Sequence release.
1.3 How to obtain a NFT?
New NFT collection is usually offered at a pre-mint price before the general public if you have made a contribution towards the community and are in the whitelist. One may also mint at the public release day. Buyer will be responsible for the gas fees of the public mint.
One could also pick up any NFT from a centralized exchange platform or Defi protocol. Opensea, Rarible and Looksrare are among the largest service providers presenting 90% of the transaction volume. There is 1 Defi LP (nftx.io) with sizable TVL and allows 1/10 of the normal size to participate at the costs of 30~50% on bid-ask spread.
There is always a risk of market manipulation when trading on the secondary market. NonFungible Corporation, established in 2018, has been tracking such illicit trade behaviour on the public Ethereum Blockchain and has identified many simple and advanced patterns. The motifs for Wash Trading have evolved with the development of the NFT ecosystem. Table 1 summarises all the motifs of wash-trade patterns. The main signature of identifying patterns are of two kinds - abnormal transaction activity and recurring transaction activity.
Wash Trading is necessarily done on purpose. Manipulating a market involves generating a large volume of transactions, which incurs costs and can be very expensive. A Market in full development and underpinned by such economic issues is likely to attract speculation and market bad behaviour. The trader practising such behaviour may have different motivations but the greater the lure of profit, the more the techniques will get evolved. Four key reasons that motivates:
-Market Making - Artificially inflates the price of an asset or asset type. -Rate Making- Artificially increases the value of Artist to sell future assets at a higher price. -Incentivizing - Some platforms provide rewards for each trade, and the rewards are substantially more than wash-trading cost. -Project Promotion- Trading volume improves ranking of a project to attract organic investors.
Chapter 2: Understand the value of the NFT
Before we tap into the value of the NFT. One perhaps should have a clearer view of the parties of interests, because their behaviour and view could be in conflict. Having “a” model to represent all groups might not be realistic. A good model should have a way to break down high level information into smaller pieces and try to relate to these parties, in order to find persistence.
2.1 What are the main groups of interests?
**Crypto group.**The group of people that has been actively engaged in the previous cycle. Many of them have been actively providing services to the community. They will keep looking for new opportunities and among them, some of the “whales” are also in their interests to maintain a reasonable level of liquidity for the crypto community.
** **
Artists. Their views are important because not only is it perhaps the best judgement on NFT’s artistic value, but also represents an important contribution to innovation and publicity. A significant number of notable sellers on NFT are from the fashion industry, including some bigger names such as Gucci. It is not impossible that some sort of “pattern” merged from a token eventually led to the discovery of fashion, which then created a ripple effect on the NFT’s space. One could argue this has already been what we are now.
**Walls street people.**They are relatively new to the new community. Their motivation of get into the NFT space perhaps can be summarized into 2 forms:
An alternative asset category , i.e. negative correlation. Because their perspective could be more long term oriented, they will ideally be rewarded a fixed yield as a source of passive income.
A centralized hub for financial service providers: lend or deposit NFT assets for leverage or liquidity.
**X-group (Teenagers).**The X-group who perhaps have a more natural instinct to the digital community. The NFT equivalent of a Twitter blue check can and deliver credible authenticity, thanks to that NFT deed. This is the metaverse vision of interoperability that could help make digital belongings feel similar to physical belongings. Digital belongings exist on the internet, but there aren’t that many types of them. On Fortnite you can acquire guns and outfits. On Reddit you gain badges. Point is: There’s a lot of stuff on the internet, but there isn’t much stuff that’s yours.
On their path to discover their personality and the view of the world, they might be interested in obtaining a digital identity and/or digital belongings. They view is quite distinct from previous two group, because:
They will live longer so that the importance of their view grows day by day.
They provide a consumer perspective so that they are the actual clients.
Collector. Christie’s and Sotheby had already put Crytopunks’ into a group of 6 as one lot for auction. They are good at improving the media exposure for their clients’ collection. Their next move is perhaps to provide a special auction just about NFTs. They have done it for Chinese blue and white, Japanese Art and Antiques or somebody important that has the provenance of their lifetime collection.
2.2 Was it actually about Membership or Utilities?
If we start from the simplest form, The logic of buying a ticket is its entry to an experience and to collect one ticket after expiry is its connection of memory. With all else equal, the value of a single entry should be lower compared with a multiple entry. So what about the value of a multiple entry ticket vs a single entry ticket suddenly becoming a multiple? This depends on a number of endogenous factors, and one of them is about risk preference.
Since IreneDAO’s success, many projects started emphasising the narrative of owning NFTs for membership utility.
Alpha Shark is an NFT project that enables owners to gain access to an elite Chinese Discord group filled with data scientists, blockchain analysts, macro trends and NFT experts.
Tiger Being is a private travel club membership NFT project on Solana, where holders get exclusive access to hotel / airline / private yacht offers via an “NFT pass”
2.3 The Case for Licensed Commercial Use Rights (LCUR)
Bored Ape Yacht Club launched in April there were two characteristics that set it apart, flat pricing at 0.08 ETH without a FOMO ramp and licensing of commercial rights.
ApeFather.eth highlighted how their ape is licensed in an upcoming comic book being produced by Myth Division, an Intellectual Property development studio founded in 2017.
There are also some insights shared from Vivek, an individual who focuses and holds rare tokens across multiple projects, for his view on projects licensing commercial rights to token holders.
Jan 23 Updated : New projects such as the**goblintown **have proved to be successful by giving away LCUR for anyone, so essentially Common Collective (CC). So having LCUR doesn’t mean it will generate value and not having LCUR doesn’t mean it will have zero value. It will also be about which way can bring in more commercial exposure.
2.4 The “film stars” Model
How can we measure its tangible value from LCUR? Creating a brand around these rare tokens is a new concept that has limited use testing, and applying textbook tangible valuation techniques is difficult.
Listed blow a few examples that one may refer to structure a way to analyse the future value of a token NFT if it has some sort of branding value:
The initial assumption is that a token with character brand recognition will have future utilisation across multiple mediums (books, comics, animation, etc.).
We can reasonably view the token as a next generation “actor/actress”. This example from metaMori.eth shows how a BAYC token can be rigged for broadcasting.
Per Wikipedia, the highest paid film actors earn $20M to $30M/year. It is not an apple to apple comparison to assume an individual NFT will generate that revenue today, nevertheless it does provide a good base for a range of tangible valuations.
The above chart highlights how a branded token could be valued in the future based on the following assumptions:
Revenue base estimated at the median of top Hollywood actors/actresses of $25M.
A discount of that revenue base beginning at 99% and scaling up to present different scenarios.
An income multiplier between 1x and 5x, to highlight how a potential buyout of the intellectual property may be valued based on projected revenue.
An understanding of this is considered “back of the envelope” calculation, and a real valuation would be far more detailed.
Based on the above, even the most conservative scenario (99% discount, 1x multiplier) would indicate a value of $250k, which is nearly double the value of a current floor ape (39.9 ETH, $138k). All other scenarios further highlight the upside of a strong brand.
2.5 A bit about Tokenomics: Bonding curve
Bonding curves are essentially a mechanism that allows the continual liquidity of a token, with the price changing depending on how much “activity” is conducted. Most use cases are for buying and selling into the curve, e.g. the more people that buy tokens, the higher the price goes up, the less people that buy tokens, the lower the price goes. The concept is closer to “network effect” in the context of traditional economics.
Traditional bonding curve graphs might look like this (simplistically, currentPrice = tokenSupply²):
** **
** **
It might be hard to understand at 1st glance for anyone from the traditional economics background, because we know the price is an inverse function of the total supply. The model represents conceptually when there is a new NFT project hitting the market. To have the price taking off as described by the model, a series of unexpected events will need to be in the pipeline to create more use cases and to make the network larger.
** **
** **
** **
2.5 The importance of creating “Events”: The case study of moonbirds
On March 19th, Moonbirds’ made the official project announcement. On March 23rd, Moonbird retweeted an affiliated tweet from Kevin Rose, founder of the PROOF Collective community, indicating there is a connection.
On March 25th, a Twitter Spaces meeting was held and key information about the Moonbirds project was announced: the mint date was finalized to be on April 16, Dutch auction will last for six hours on the day of minting (starting at 2.5E), and Moonbirds holders will unlock:
access to the PROOF community internal channel
eligibility for subsequent airdrops in the PROOF community
eligibility to recreate Moonbirds IP
early bird qualification and discount in the PROOF community automatic approval for each PROOF Collective holder to qualify for two free mints of Moonbirds
Now Moonbirds' price has also skyrocketed from its mint price of 2.5E all the way to nearly 20E. Realising some traders might quickly flip their NFTs to book the net, a tactical floor price is set.
Nesting is a process where the moonbirds are locked without ever leaving the user’s wallet. Nesting Moonbirds will begin to accumulate additional benefits and unlock rewards. When a moonbird is unnested its owner will be not eligible for any future drops.
Some say that it was floor price manipulation by telling people that they would lose Nesting. The truth is that by letting the holders know in some way that as soon as they flip their NFT for quick profits automatically, you will be unnested. This is to force early investors to hold their NFT to prevent quick flippers.
2.6 Statistical evaluation of the ‘film star’ model
As the data for all the worldwide film stars is very hard to obtain and verify. We have used data from previous analysis on NBA clubs instead. Conceptually, a NBA club is equivalent to a NFT collection. The purpose of exercise is to verify if there is a persistent relationship between social media exposure and the value of the NBA club. If this is the case, one could possibly provide a reasonable estimate of any NBA club (NFT collection) based on real time data from social network platforms. The underlying assumption is that the social media exposure is a proxy of the number of the total fans.
2.7 ‘Breeding’ your shoes: The case study of Cryptokick (Nike)
In December 2019, Nike patented a system and method for providing cryptographically secured digital assets, as the response to its counterfeiting problems and a move to the crypto space. It describes a system for minting, exchanging, and intermingling cryptographic digital assets in the form of digital shoes, which can each be linked to a real-world physical shoe.
Each digital token would be “unlocked” with the purchase of a corresponding physical shoe by linking a 10-digit shoe identification code with the owner identification code. The system apparently aims to provide a way to ensure the authenticity of the goods.
Nike also suggests that token creation may be linked to shoe sales, which would also allow for the verification of the scarcity of the shoes in circulation. When the footwear is sold, ownership of the digital token can be passed on alongside the physical product. The digital token is primarily an avatar of the physical shoe hosted on a blockchain whose cryptographic keys are stored in locker app of the shoe owner.
The patent also borrows ideas from popular blockchain collectible cats game CryptoKitties. On top of the standard blockchain functionalities the buyer is enabled to "breed" the digital shoe with another digital shoe to create "shoe offspring" and, based on rules of acceptable shoe manufacturability, have the newly bred shoe offspring custom made as a new, tangible pair of shoes. It is also possible to mash-up multiple CryptoKick objects to create an offspring CollaboKick, which can likewise be traded, sold, or embodied in a physical shoe.
2.7 Rarity: one attribute of impressive or a proxy of impressive?
In the analysis above, one could infer that the rarity of NFT is a matter of concern because:
The collector group had already developed a taste of looking for the rarest.
In absence of more reliable data, one may use rarity as a proxy or a condition for impressing future investors.
There are a few different ways to combine trait rarity to return a single score representing each NFT. The method went on calculating the rarity of each individual trait, and then multiplying all the scores together sounds more statistically reasonable.
To walk through an example on Koda #5803, by collecting data and we have:
· Head: Aqua Elf (86 / 9632) = 0.89%
· Eyes: Pink Staredown (34 / 9632) = 0.35%
· Core: Tranquil Water (104 / 9632) = 1.08%
· Clothing: Grimace Armor (33 / 9632) = 0.34%
· Weapon: Bramble Thumper (19 / 9632) = 0.20%
Second, calculate rarity by combining the scores, so that Rarity = 0.0089 * 0.0035 * 0.0108 * 0.0034 * 0.0020.
Chapter 3: Current innovation and engineering
# TODOhttps://medium.com/@Austerity_Sucks/a-simple-valuation-model-of-lifinity-flare-nfts-e6249900c787
#TODOhttps://fanadise.medium.com/honey-tokenomics-28c49ebf3924
3.1 The economics behind staking and burning
The reward of staking is usually from generating an extra block which is linked to the number of transactions in the block. Some of the token makers are willing to pay a very high staking rewards, one might be worried about the source of the rewards and its persistence. A simple analysis is done as the following:
| Scenario 1 | Scenario 2 | Scenario 3 | |||
Initial Value |
| 50,000 |
| 50,000 |
| 50,000 |
Volume |
| 150000 |
| 150000 |
| 750000 |
|
|
|
|
|
|
|
Staking | 10% | -1500 | 20% | -3000 | 8% | -1200 |
LP | 30% | -4500 | 60% | -9000 | 12% | -1800 |
|
|
|
|
|
|
|
Gas fees | 0.40% | 600 | 0.25% | 375 | 0.10% | 750 |
LP fees | 0.30% | 450 | 0.25% | 375 | 0.12% | 900 |
|
|
|
|
|
|
|
Inflation | 10% | -5000 | 10% | -5000 | 10% | -5000 |
Demand increase | 20% | 10000 | 30% | 15000 | 25% | 12500 |
|
|
|
|
|
|
|
End Value |
| 50,050 |
| 48,750 |
| 56,150 |
For some NFT collections, the community will provide rewards for anyone who has burnt their NFT. There is a clear gain on all other NFT holders if anybody burns a NFT. Only holders with a significant number of NFTs are likely to be benefiting from burning NFTs personally.
** **
3.2 Technical background of yield farming
There is a growing tendency to provide yield on NFT, as the community has realized that a large amount of assets in the form of NFT investments are sitting still, without an actual use case.
Source: https://mutantcats.io/
NFTI has recently published a NFT index that is implicitly** ****connected to the floor price** of their respective NFT collection. The numbers of NFTs in vault is shown in the table below, as at 9th of Mar 2022 :
This is achieved by utilizing the NFTX’s xTokens. Here these steps are shown below by using CryptoPunks as an example:
Convert a CryptoPunk NFT into a PUNK ERC-20 token by depositing the NFT into an NFTX vault. This PUNK token, called a vToken, represents a 1:1 claim on a random CryptoPunk in the vault.
Convert the vToken PUNK into the xToken xPUNK by participating in NFTX Inventory Staking. Staking vTokens is reversible and when unstaking at least as many vTokens as used for staking are returned.
xToken holders are rewarded a share of the protocol fees that are generated from NFTX vault activity (minting, redeeming, swapping). Therefore, by effectively holding xTokens, NFTI holders receive this extra yield as the amount of vTokens one xToken represents grows.
For NFT collections that are currently part of the index, the NFTX vaults hold a large set of NFTs. This can serve as a proxy for current maximum availability of xTokens but of course, if there is more demand,more tokens can be minted at any time. People will accessibility can always watch out if a sensible amount of xToken has been minted. However, there is no audit trails.
With some background information, it is not difficult to see that by adding a small percentage of any NFT collection one can theoretically set up a chosen floor price at any level. However, to provide a more convincing solution the pool should be large enough to accommodate 50x-30x average daily trading volume. It would means the daily maximum IL loss is limited to a market move that is based on 2-3% of the pool, i.e. f(1/50*Z), where Z=X*Y. Compared with exchange trading mechanisms, this way implicitly assumes there is an exponential function on price level stepwise. For example, the buyer who bought 10 NFT at once is at a pre-defined exponential function compared with the buyer who bought 1 NFT.
3.3 Understand the fees from LP on NFT
To understand the entire process of yield farming, it is perhaps ideal to start with recalculating APR. A typical NFT LP selection page is shown below:
(Source: https://nftx.io/rewards/pools/)
**Fees:**There are three revenue streams when staking on NFTX. Two streams as part of the liquidity providing, and one stream on the inventory sided staking.
Sushi Fees (0.25% fees paid to liquidity providers)
Vault Fees (80% paid to liquidity providers)
Vault Fees (20% paid to inventory providers)
When vaults are first created they default to the following Vault Fees:
· 10% Mint/Sell
· 10% Redeem/Buy
· 5% Random Redeem
· 5% Random Swap
· 10% Targeted Swap
**Changes of Fees :**While these fees are set by default, they can be changed if:
If a user has more than 50% of the liquidity pool
or the NFTX Forum which requires at least 80% votes in favour of the changes.
3.4 Back of Envelope calculation on APR
On-chain data can be retrieved from dune which provides pseudo SQL driven data analytics online.
By using the data from dune alone, one can calculate a simple estimated APR for liquidity provider as the following:
| Fees Actual | Fees Annualised | Weight | Est. Avg fees | Est APR |
2021 Dec - 2022 May | 7.15 | 17.16 | 5/15 (2/15) | 5.72 (2.288) |
|
2021 Jul - 2022 May | 121*8% = 9.68 | 9.68/15*12= 7.75 | 10/15 (13/15) | 5.167 (6.717) |
|
Avg Fees |
|
|
| 10.886(9) |
|
Fees Return |
|
|
|
| 64% (54%) |
Impairment Loss |
|
|
|
| -21% (-18%) |
Est. Transaction Fees |
|
|
|
| -30% (-50%) |
Net Return |
|
|
|
| +13% (-22%) |
APR might initially look misleading, but one might also compare with the buy and hold experience with 2.5% US treasury. Does the buyer actually get 2.5% in the end?
Last updated
