.png)
Introduction
The price variance in an NFT collection can be vast – up to a 100x difference between the floor price and grail price. This is because the price of an NFT depends on myriad factors, including their traits and sales history. This complexity causes uncertainty for investors – particularly in lieu of an industry-standard valuation model – making it difficult to make quick investment decisions with high confidence.
To combat this uncertainty, Nansen has developed a new NFT Price Estimates machine learning model backed by our industry-leading on-chain data. This battle-tested Machine Learning (ML) model powers our brand new NFT Sniper dashboard, which unlocks new insights and enables market participants to make rapid decisions with high confidence across 880 of the biggest NFT collections (and growing each day).
TLDR:
- Nansen has improved its machine learning model to produce its Price Estimates v2.0 model, a tried and tested model estimating NFT prices based on sale history and NFT traits.
- Among the many features powered by Nansen’s Price Estimates v2.0 model, the NFT Sniper is one of them.
- The Price Estimates v2.0 model has helped make meaningful changes to features such as NFT God Mode, Item Profiler and Rarity Profiler.
- Users can ‘snipe’ for undervalued NFTs using Nansen’s NFT Sniper promptly while also scouting for potential NFT trades that match their investment budget or intended trade margin.
NFT Sniping Fueled By ML-Derived Price Estimates
Of the many Nansen dashboard features powered by the Price Estimates ML model, Nansen has introduced a new NFT Sniper dashboard as part of our NFT Analytics initiative. The NFT Sniper dashboard can be located as a tab within the Nansen NFT Paradise.
The NFT Sniper dashboard showcases the listings with the most undervalued listings in each collection. In addition, NFT Snipers can organized the undervalued listings stable by the 1) Estimated Price Difference (%), 2) Estimated Price Difference (ETH), 3) Listed Price (ETH), 4) When It Was Listed, and 5)The Collection’s Volume In The Last 24 Hours.
NFT Sniper
A user will typically sort the NFT Sniper dashboard based on the Estimated Price Difference (ETH) to scout for the most undervalued listings (denominated in ETH). Users who want to find a trade or investment better suited for their budget can filter the table by the percentage difference. On the other hand, more active users with capital can look at the most recent listing; a default listing - to take immediate action when snipping NFTs. Floor and Volume columns are intended for due diligence purchase. For example, when users spot an undervalued NFT, they can verify the collection’s trading volume to gather a more holistic picture of the market’s sentiment towards that collection.
Additionally, users can filter by NFT-500 or the Blue Chip-10 collections. These collections have been included in the NFT indexes, meaning they have met a minimum liquidity requirement (you can read more about the NFT indexes here).
Undervalued listings zoomed in
For example, at the time of writing, I filtered the undervalued listings by Listing Time (the default listing). Several MAYC were detailed on the Undervalued Leaderboard, such as MAYC #17952, #8616 and #5435. Looking at MAYC #8616, for example, the NFT was undervalued by approximately 12% (2.32 ETH). It was listed for 19.67 ETH with an estimated price of approximately 22 ETH. Upon identifying a specific NFT, users can click directly on the NFT ID to navigate the associated NFT Item Profiler.
NFT Item Profiler: MAYC #8616
The NFT Item Profiler, powered by the new Price Estimates machine learning model, details the median estimated price of the traits associated with the specific NFTs while providing a line graph comparing the NFT’s historical price estimates and floor prices. For example, in the case of MAYC #8616, due to its specific trait types, we have seen its estimated price outpacing the growth of the collection’s floor price.
NFT God Mode
Users can use the NFT Price Estimates table to compare the value of a specific NFT relative to the other pieces in the collection. For example, MAYC #23876 is the most undervalued price in the collection, followed by MAYC #28910. A similar table can also be found in NFT God Mode for the NFT collections.
NFT God Mode Users can also explore the collection’s top trait types by estimated values. An exciting feature of Nansen’s NFT God Mode is showing the distribution of NFTs within the collection as a function of their estimated prices. For MAYC, most price estimates were between 17 to 19 ETH, and the top trait types by estimated value are 1) Mega Trippe, 2) Mega Death Bot, and 3) Mega DMT.
Nansen’s Edge
When NFTs’ traits are revealed, the price estimates model uses the traits of the NFT to compute a feature vector (’unique’ fingerprint) for each token in the collection. This means that the Price Estimates v2.0 model can value any NFT within the collection as of ‘now,’ especially ones with no prior sales history.
The Price Estimates v2.0 model:
- estimates any NFT as long as we have traits and have (sufficient) sales
- estimates are real-time, and the first estimates and estimates of new collections are available within 1-3 hours of (1) being satisfied.
Nansen’s Price Estimates model is expected to improve with the coverage for NFT collections over time. Among the many features powered by Nansen’s Price Estimates v2.0 model, the NFT Sniper is one of them.
Why not try it yourself?
Nansen’s continuous innovation means that you can now Snipe NFTs with confidence. Our price estimates powered by machine learning are helpful when you plan to buy or sell an NFT. Click here to try Nansen’s Price Estimates v2.0 model. As NFT natives, we recognized a need for swift action when sniping undervalued NFTs. This is why we have designed the NFT Sniper dashboard to provide users with a more seamless experience when they conduct their due diligence and make the purchase - all within a few clicks away.
.png)