Let’s dive right in. 🌊

What does liquidity really mean?

Let’s say you’re trying to sell a 3-bedroom house that has a 10-car garage. How much extra value does the massive 10-car garage add to that house? The answer will differ depending on who you ask. To most purchasers, it would add very little, as they only need a small garage if at all. However, to an active car collector, having the extra garage space could mean a lot!

That said, the number of active car-collecting house shoppers is much smaller than the total number of house shoppers out there, so it would take longer to find a buyer that would be willing to pay the premium due to the 10-car garage. Sure, the house can be sold quickly to any purchaser, but to receive the premium (or the highest value anyone would be willing to pay), you will have to sacrifice time on the market to find the buyer who values the garage.

The value of an NFT depends on the amount of time you are willing to wait for the right purchaser, which is why it is incomplete to consider valuation without considering liquidity, and vice-versa. There is a trade-off between how long it may take to sell, and the total price you will receive.

The challenge with liquidity

Just as a house with a 10-car garage might receive an extra 50% in sale price if you are willing to wait for the right buyer, to make an informed decision, you’d need to know how long that may take. You can imagine how important this is in markets, especially given how fast markets move and prices change. Is it worth waiting for the right buyer? Is there more upside selling this NFT immediately and jumping into a different collection? These are all important questions, but until now, the answers haven’t been immediately clear.

Most people try to get a sense of liquidity by eying the transaction history of the NFT collection being evaluated. Collections with hundreds of transactions per day can be assumed to be more liquid (ie, you don’t need to wait as long to find a buyer who’s willing to pay you the value of your NFT) than a collection with a few transactions per month.

To drill down into the liquidity level of a specific NFT, advanced traders may analyze the transaction history of NFTs with a specific trait shared by the NFT. For instance, a solid gold BAYC is much more valuable than a BAYC at the floor, but it will also be less liquid if listed at its fair market value.

How traders tackle liquidity today

There are a few simple technical metrics one can also use to measure the overall liquidity of a collection or trait within a collection. One that we’ve identified is the floor : ceiling ratio (which we’re also referring to as the liquidity ratio), which is calculated by dividing the floor listing price of the collection by the bid ceiling (or highest collection-wide offer). For example, if the floor of BAYC is 77.4 ETH and the bid ceiling of BAYC is 72.5 WETH, the liquidity ratio is 1.07. Likewise, if the floor of the Solid Gold Fur trait in BAYC is 750 ETH and the bid ceiling for this trait is 200 ETH, the liquidity ratio of the Solid Gold Fur sub-collection is 3.75. The lower the liquidity ratio is and closer to 1 it approaches, the more liquid the collection or sub-collection is. This signal helps one understand the level of demand for a collection relative to the level of market supply or willingness to sell.

Below is a non-exhaustive exhaustive list of other signals that help to indicate liquidity:

• Price: we can reasonably think higher priced NFTs may take longer to find a buyer.

• Collection: some collections are currently more popular than others.

• Sales volume: amount of sales in the last week, day, hour, minute.

• Bid volume: amount of bids in the last week, day, hour, minute.

• Rarity: are there many NFTs similar to this one?

• Listings: are there numerous competing NFTs currently for sale? What percent of the collection is for sale?

• Bids: are there a lot of traders indicating they are looking to buy?

• Discount: the premium or discount of the NFT to its current valuation.

• Average floor sale time: the time delta between when a floor listing is posted and then sold.

• Listing sales volume vs. offer sales volume: are most people accepting offers or purchasing spot?

• Depth of the floor: what does the structure of floor listings look like? How would the floor change if some number of floor NFTs were purchased?

• Depth of the bid ceiling: what does the structure of collection bids look like? How would the bid ceiling change if some number of collection bids were taken?

However, these simple metrics do not paint a complete picture and while they are helpful, they are far from the optimal solution. So how can we go deeper? To do this, we need to understand the properties of non-fungible assets.

In fungible markets, liquidity is usually a comparative measure: for example, a stock traded on an exchange is more liquid than a stock only traded over the counter (OTC). Additionally, analysis is much simpler because assets are completely fungible (one share of SPY is indistinguishable from another, and one 1 ETH = 1 ETH). In these markets, valuation is also much simpler: the current best estimate of value is the current price, so a liquidity analysis primarily looks to show how much of the asset you can buy or sell without moving the price of the asset.

Non-fungible markets are different: every single asset is unique. The liquidity functions much more like liquidity for a house. In order to determine liquidity for a specific NFT, one must assess the specific properties of this NFT, its market behavior, and its transaction history. Doing this for each listing in a collection is quite challenging and time-consuming for human traders, and in a market where timing is important, this can lead to subpar returns. This is why we’ve built a robust, machine learning-based liquidity analysis model: to give you superpowers for analyzing liquidity.

How Atlas helps traders navigate liquidity

We have developed three distinct approaches which leverage machine learning to analyze the specific liquidity of each individual NFT in a collection in real-time. From here, you are able to dig deeper into the NFTs you are interested in, which is much more efficient than trying to assess the liquidity of all listings yourself.

Each of these approaches incorporate the same data we use to assess valuation for an NFT. More information on this can be found in our Valuation Analysis article.

1. Valuation + liquidity signals → sale time: Look at every sale that has happened, create a valuation for when the listing was added, and then add signals that you think best categorize how liquid that asset might be. Then calculate the time it took between the listing and the sale. Build a model that predicts the time it will take given the listing’s premium or discount and all other signals.

   This is a very direct approach that gets right to the heart of the problem, but has some of its own potential drawbacks. The main concern with an approach like this is that we condition our sample data on an item being sold. There is a really interesting statistical phenomenon called adverse selection and this happens when your data sampling technique is not dictated by yourself, but by others. In this case our data sampling technique is dictated by the market. That condition (sample WHEN a sale happens) leaves the sampling out of our hands, and may over sample with conditions we are not controlling for.

2. Correcting for adverse selection: Same approach as number one, but change the sampling structure to include sampling from every single listing that has happened (in an attempt to address conditional bias on sampling only sales). For unsold listings change the time to a defaulted maximum.

   This approach also has some drawbacks. It introduces a lot more data, and a lot of it is superfluous and unnecessary. Choosing how to properly address canceled listings, and repriced listings, introduces its own bias into the data. Although this addresses some of the conditional sampling bias of the first approach, the error estimate increases wildly and introduces your own sampling biases in how a sparse dataset is cleaned.

3. Learning listing-to-sale time buckets: Sample from every listing we see, and create defined time periods (such as 10 seconds, 1 minute, 10 minutes, 1 hour, 1 day, etc). Then, determine if the listing sold within that time period. Model the likelihood of sale given a valuation, for each time period and then output the liquidity curve as the connection of prices that have a 90%+ probability of selling within that time frame.

   The drawbacks here are that the predetermined time periods can be clunky and untenable across many many different collections. Modeling and calculation time explodes with creating a unique training set for each time period which may not be suitable for real-time analysis. Additionally, data is incredibly sparse for sales compared to listings. Traditional techniques to deal with this include upsampling or oversampling which may bring in the initial conditional bias you were trying to remove.

While each of these approaches have their own drawbacks, we’ve found that together, they form the basis of a powerful system for analyzing the level of liquidity for NFTs in a collection in real-time.

Using liquidity analysis to level up your trading

Trading NFTs is ultimately about inventory management. After deciding to make a bet on a collection, having control over your entry point and exit point is paramount to the returns you’re able to realize.

Before buying an NFT, it’s a good practice to evaluate the liquidity of your own portfolio as well as the liquidity of the NFT. If you’re putting a substantial amount of your liquidity into an NFT that is less liquid, will you be able to react quickly enough if a better opportunity comes along? What’s the probability of such an opportunity coming along? These are questions that you must answer for yourself, but having done so, a solid grasp of liquidity of the NFTs you are considering and the ones you already hold is crucial for managing your NFT portfolio in an optimal way.

Likewise, before selling an NFT, consider how long you’re willing to wait for it to sit on the market. Do you feel good about the market holding steady or even increasing? Or would you prefer to exit quickly? After determining your desired timeline, you can input this into Atlas’ liquidity analysis tool to find the optimal listing price for the NFT. This way, you’ll be able to manage tighter, more process-driven strategies for inventory management of your NFT portfolio, and ultimately, your end returns.

Remember, executing trades on Atlas means you’ll be saving on gas with the most gas-efficient aggregator around, and you’ll also be earning points toward the eventual airdrop!

Join our waitlist (and earn points doing so):

https://atlas.xyz/signup

Jump into our Discord:

https://discord.gg/atlasserver

Connect with us on Twitter:

https://twitter.com/tlas_xyz

About Atlas

Atlas is a revolutionary trading and financial products platform for NFTs, powered by cutting-edge machine learning. The team is deeply technical with a strong track record of building great products, having founded or been on the founding team of an NFT unicorn, a DeFi unicorn, and a highly successful quantitative trading firm. We’ve been building NFT marketplaces since 2018, quant and AI trading strategies since 2010, and have scaled multiple products from 0 to millions. Join Atlas today and be a part of the future of NFT trading.

Uncovering hidden gems in the NFT market can feel like searching for a needle in a haystack. The process of evaluating and comparing the value of each NFT to its listing price can be time-consuming and often by the time you've found a good deal, it's already been snatched up. But what if you could have an AI-powered sidekick that could help you identify the most valuable NFTs in a collection with just a few clicks, improving your cost basis and boosting your overall returns?

This is where Atlas comes in. We leverage cutting-edge machine learning techniques to deliver real-time, best-in-class valuations for NFTs, so instead of haphazardly purchasing the lowest-priced NFTs in a collection, you can use a sophisticated system that helps you make informed decisions and maximize your returns:

1. Search for the collection you're interested in buying into

2. The collection will automatically be sorted by the best value, but you can also sort by lowest price for comparison

3. Use filters for traits or price ranges, take a closer look at a few of the top NFTs, and purchase or sweep with confidence.

Under the Hood of Atlas’ AI Engine

Thousands of data points per item. Thousands of items per collection. Thousands of collections.

Building a robust, real-time machine learning pipeline for fast inference using billions of data points is not easy. And that’s why we’re excited to share our learnings with you. ⚡

We calculate a valuation distribution for each individual NFT in each collection we support. None of those valuations would be useful if they aren’t reacting real-time to how the markets are changing! We knew the strict importance of both breadth of coverage and speed of valuations criteria when designing a useful machine learning system and were up to the challenge. To accomplish this, we've segmented our learning system into two parts: a static metadata system and a real-time inference system.

Traits and other static metadata

The static metadata system precomputes all of the unchanging (or at least changing less frequently than our model needs to update) characteristics of an NFT, things such as all of the trait combinations, the number of traits an NFT has, how many are gold traits, what other collections it is similar to, how long the collection has been around, what the historical volume for the collection is…really anything that doesn’t change often, but contributes to the value of an NFT.

This static system then allows us to do deep inference on all that metadata. Think of this as doing all of the ‘heavy lifting’ of comparing all of those different combinations. We use neural networks to create embeddings of this metadata for each NFT so that they can be easily processed downstream in real-time as new order events come in. One technique we use to embed these features is Deep Hyperedges. With this method, we represent an NFT collection as a hypergraph. A hypergraph can be thought of as a big Venn diagram: points in the Venn diagram represent NFTs and circles represent traits. So, if there’s a circle for “gold fur” and another circle for “coin eyes”, an NFT with both of these traits would live in the intersection.

This method and way of representing an NFT collection allows us to create a very powerful relational model and get a large amount of implied information about an NFT’s value given the market behavior of other NFTs in the collection.

Real-time market dynamics

But like we said earlier, if the valuations aren’t in realtime, what’s the point!? Enter the real-time inference system. We feed the output of the heavy-lifting static system into the real-time system and then add our real-time signals. This takes into account all of the NFT market structures that are in flux. Things like the current floor, the current collection-wide bid ceiling, the current NFT listings, and which sales just happened seconds ago are important events we stream through our system to perform real-time inference. Those real-time market factors don’t have as many combinations, and can be distilled much more numerically than all of the static parts, so tracking them is (relatively!) less computationally intense for us to process.

NFT-specific market sentiment

Each individual NFT’s orderbook and transaction history gives us insight into the sentiment, market performance, and demand relative to other NFTs in the collection. By analyzing the historical and current offers and listings for the NFTs individually in the collection, we can get a clear sense of what the open market is willing to do at a baseline. For instance, if an NFT has an offer for 5 ETH and is listed at 7 ETH, we know the value of the NFT is between 5 and 7 ETH. This information is used to calculate various technical indicators (such as the floor:ceiling ratio) that help us understand liquidity and market sentiment. Our sequence models process this with a bit more complexity and comprehension than a simple signal could.

Above is a visualization of the desirability rank of bored apes over time based purely on their offers and listings (the orderbook) using a ranking algorithm we created. This serves to “sanity check” the valuations we provide.

Other market factors

A considerable amount of signal comes from Twitter for many collections. We leverage natural language models for processing related tweets, as well as graph neural networks to analyze the follower graph surrounding a collection. By analyzing social media data, we can determine the NFT collection’s reach and popularity, which can have a significant impact on its market cap. Similarly, we use the performance of other assets such as crypto and equities to gain insight into market conditions and how they may affect the collection’s value. We incorporate these other market factors into our dynamic real-time models as features to give traders not just a sense of how the NFTs are valued relative to the NFT market, but also relative to the overall market.

How Should I Trust Atlas’ Valuations?

We’re glad you asked. We’ve asked the same questions of other NFT valuation services in the market, and that’s why we’ve built a robust system of interpretability tools to give you the information you need to leverage Atlas’ AI in your own workflow as a trusted companion. Stay tuned for our next articles on Liquidity Analysis and Interpretability which will dive into details about how you can use Atlas effectively and confidently in your trading workflow.

https://mirror.xyz/0x2d62841615EAE1541554B3Ba1be1b59bD0BFdbf9/4tpJqKVJGSjtYZXAv9DCm2W6nm89JwR_0UT_2v6DD7k

Remember, executing trades on Atlas means you’ll be saving on gas with the most gas-efficient aggregator around, and you’ll also be earning points toward the eventual airdrop!

Join our waitlist (and earn points doing so):

https://atlas.xyz/signup

Jump into our Discord:

https://discord.gg/atlasserver

Connect with us on Twitter:

https://twitter.com/tlas_xyz

About Atlas

Atlas is a revolutionary trading and financial products platform for NFTs, powered by cutting-edge machine learning. The team is deeply technical with a strong track record of building great products, having founded or been on the founding team of an NFT unicorn, a DeFi unicorn, and a highly successful quantitative trading firm. We’ve been building NFT marketplaces since 2018, quant and AI trading strategies since 2010, and have scaled multiple products from 0 to millions. Join Atlas today and be a part of the future of NFT trading.

We are thrilled to announce the launch of Atlas.xyz, a revolutionary, AI-powered trading and financial products platform that will give NFT traders a significant advantage in the market.

Atlas is more than just another NFT trading platform: it's a financial OS for NFTs, powered by cutting-edge machine learning. Our advanced algorithms enable us to conduct real-time, in-depth pricing and liquidity analysis for NFTs, providing the foundation for a range of innovative financial products and services we’ll be rolling out.

These financial products will give you greater control and flexibility over your NFT portfolio, allowing you to manage risk and optimize returns in a way that was never before possible.

The Token

At the heart of Atlas's financial OS for NFTs is our native token, which plays a key role in powering our financial products.

Atlas's token will be used to govern and provide liquidity for these innovative financial products. By staking the token, users will be able to participate in the growth of our financial products, and will be rewarded for their contribution.

Users earn the Atlas token by providing liquidity to our financial products as well as through trading volume and the airdrop.

One financial product we will launch is Buy Now, Pay Later: users who stake the token will provide the initial liquidity needed for others to purchase NFTs at market by only putting down a down-payment and paying off the rest with interest. Our pricing and liquidity oracle and other safety mechanisms will determine the liquidation threshold in real-time. Stakers earn a return on their contribution, and the buyers are able to access the financial products they need to grow and manage their NFT portfolio.

We will be launching a whole suite of financial products like these through the Atlas platform over time. More details will be announced soon 🤫

The Airdrop

The airdrop for Atlas's native token is an exciting opportunity for users to get their hands on some of the first tokens to be released on our platform. We will be airdropping our initial distribution tokens based on our point system, which rewards users for their trading activity on Atlas. The more points you have, the larger your token airdrop will be.

Point values for each activity can be viewed on our launch announcement article. Note that point values for all Twitter and Discord activity will be verified during various snapshots up until a future date after the window for activity-earned points has closed, so make sure your actions remain active throughout this time.

No product was ever possible without the support of its earliest believers. We are grateful for all of you who believe in the potential this has, and we want to ensure that your support is recognized.

The Point System

For a limited time, there are a few other easy ways you can score some early points, such as joining the waitlist, checking out our article(s), joining our Discord, following us on Twitter, liking and retweeting our Genesis Tweet, and inviting others to join the platform.

Points will be used to award tokens in the airdrop, as well as access to the platform and early beta features.

Invite a friend and you both get 250 points! If a friend accepts your invitation to join our waitlist, you will each get 250 points.

1. Joining the waitlist (+250): Join the waitlist to earn these points. ETH address and a Twitter handle will be required, and an email address is highly encouraged.

2. Checking out our Mirror blog (+50): After joining the waitlist, click the link in the Points page to earn these points. Consider subscribing for some good alpha!

3. Joining our Discord (+200): Users who are in our Discord will be awarded points after the snapshot. We’ll use Collab.Land to verify.

4. Following us on Twitter (+200): Users who follow us on Twitter will be awarded points after the snapshot.

5. Liking our Genesis Tweet (+100): Users who like our Genesis Tweet will be awarded points after the snapshot.

6. Retweeting our Genesis Tweet (+300): Users who retweet our Genesis Tweet will be awarded points after the snapshot.

   📣 Important Note 📣 we will take a snapshot at a future date to award all Discord and Twitter points to users. Make sure your actions are active throughout the waitlist period so you don’t miss it.

7. Invitations (+250 or more per acceptance): You will earn 250 points per user who accepts your invite. You will also earn points based on the activity of those you invite.

Well, what are you waiting for? Those points aren’t going to earn themselves!

Join our waitlist (and earn points):

https://atlas.xyz/signup

Jump into our Discord:

https://discord.gg/atlasserver

Connect with us on Twitter:

https://twitter.com/tlas_xyz

About Atlas

Atlas is a revolutionary trading and financial products platform for NFTs, powered by cutting-edge machine learning. The team is deeply technical with a strong track record of building great products, having founded or been on the founding team of an NFT unicorn, a DeFi unicorn, and a highly successful quantitative trading firm. We’ve been building NFT marketplaces since 2018, quant and AI trading strategies since 2010, and have scaled multiple products from 0 to millions. Join Atlas today and be a part of the future of NFTs.

You + Atlas = Unstoppable.

• The best way to buy: Most people buy into a collection through the floor. Titans use Atlas to buy the most discounted NFT in the collection.

  We use machine learning to price NFTs in real-time with institutional-grade accuracy, as well as meta-analysis so you can DYOR. Learn more in our Valuation Analysis documentation.

• The best way to sell: selling has been more of an art than it is a science…until now. Our liquidity analysis gives you the best way to price an NFT if you want to sell it in a given time window.

We’ve pioneered new machine learning methods to calculate this liquidity analysis. To our knowledge, no comparable product exists for this anywhere in the market today. Learn more in our Liquidity Analysis documentation.

• The best way to analyze individual NFTs and collections: take “smart money” to a whole new level with our machine learning-powered suite of analysis tools. Atlas helps you to validate your insights and do better research with ease.

We’ve built the most gas-efficient aggregator contract around so you can save more on gas with Atlas than anywhere else. We will also be supporting programmatic + automated order execution soon. We’re also very mobile-friendly - check it out!

Earn points for the airdrop!

Atlas maintains a point system based on your trading activity on Atlas. Points will be used as a distribution mechanism for our token airdrop.

For a limited time, there are a few other ways you can score some early points:

1. Joining the waitlist (+25): Join the waitlist to earn these points. ETH address and a Twitter handle will be required, and an email address is highly encouraged.

2. Checking out this article (+5): After joining the waitlist, click the link in the Points page to this article to earn these points.

3. Joining our Discord (+20): Users who are in our Discord will be awarded points after the snapshot. We’ll use Collab.Land to verify.

4. Following us on Twitter (+20): Users who follow us on Twitter will be awarded points after the snapshot.

5. Liking our Genesis Tweet (+10): Users who like our Genesis Tweet will be awarded points after the snapshot.

6. Retweeting our Genesis Tweet (+30): Users who retweet our Genesis Tweet will be awarded points after the snapshot.

   📣 Important Note 📣 we will take a snapshot at a future date to award all Discord and Twitter points to users. Make sure your actions are active throughout the waitlist period so you don’t miss it.

7. Invitations (+30 or more per acceptance): You will earn 30 points per user who accepts your invite. You will also earn points based on the activity of those you invite.

There’s also an easter egg hidden somewhere in the Points page; if you find it, you will be rewarded. Learn more about the innovative utility of our points and token below.

https://mirror.xyz/0x2d62841615EAE1541554B3Ba1be1b59bD0BFdbf9/zexq8wIbGoTUvMHgYrBAd_vxyVfbrPL4YNvqFL-az0I

Join our waitlist (and earn points):

https://www.atlas.xyz/signup

Jump into our Discord:

https://discord.gg/atlasserver

Connect with us on Twitter:

https://twitter.com/tlas_xyz

About Atlas

Atlas is a revolutionary trading and financial products platform for NFTs, powered by cutting-edge machine learning. The team is deeply technical with a strong track record of building great products, having founded or been on the founding team of an NFT unicorn, a DeFi unicorn, and a highly successful quantitative trading firm. We’ve been building NFT marketplaces since 2018, quant and AI trading strategies since 2010, and have scaled multiple products from 0 to millions. Join Atlas today and be a part of the future of NFTs.

🤫 Sneak peek at our upcoming UI 🤫