The Concept of NFT Pricing

NFT assets are different from fungible tokens like ETH, which always have a real-time transaction price that can be directly used for their pricing in financial services such as lending and mortgage. Each NFT based on the ERC-721 protocol is unique, and they have a separate price. This non-fungible feature leads to the problem of much lower liquidity than fungible assets.

In the traditional way, the NFT price is estimated by the floor price or the last sale price.

If using the floor price, the trait premiums are ignored, which harms grail NFTs(those with high rarity rank).
If using the last sale price, the price is prone to be outdated due to the NFT's low liquidity problem.

Both of pricing methods can not provide a reasonable price, significantly limiting the liquidity of NFTs and hindering the development of NFTfi. To accurately evaluate NFTs, we need a better method to capture the trait premiums, which should be reflected in the final valuations

How We Implements NFT Pricing

NFT pricing is similar to real estate pricing. An NFT collection can be regarded as an apartment. The traits of an NFT correspond to the features of a house. Just as a BAYC NFT possesses traits like "Trippy Fur" and "Party Hat," similar to how a house has different attributes such as size, number of floors, and orientation.

NFT pricing is similar to real estate pricing

The value of a house can be perceived as twofold:

the inherent value of the apartment, and
the value derived from various features, such as different sizes, rooms, direction and floors.

Similarly, the value of an NFT can also be seen in two parts:

the value of the collection as a whole, and
the premium associated with its traits.

A good way to anchor collection value is floor price. It is real-time and relatively reasonable, and difficult to be manipulated by individuals. And trait premiums can be learned by machine learning from all the historical sales(exclude wash trades). Thus, it's obvious that we can evaluate the NFT value using floor price(collection value) plus the trait premium value.

A NFT pricing consists of collection value and its trait premiums

By leveraging machine learning techniques, the pricing of NFTs achieves a transformation of liquidity from the collection level to the individual NFT level. The crux of this transformation lies in capturing the mapping between premiums and traits. Through the analysis of extensive data, machine learning algorithms can discern patterns between different traits and their associated premiums within the NFT market. This signifies that for NFTs possessing specific attributes, their pricing relative to the entire collection can be determined based on predictions from machine learning models.

Consequently, the liquidity that was once confined to the collection level can now be extended to each individual NFT. This machine learning-driven approach not only aids in a more accurate valuation of each NFT but also provides participants in the NFT market with increased opportunities for trading and investment, thereby fostering further growth and maturation of the NFT market.

collection-level liquidity is transformed to NFT-level liquidity by capturing trait premiums with machine learning

GoPricing Framework

The GoPricing Framework is based on collection floor price. Floor price is used as a pivot for the basic value of a collection. We aggregated the collection floor price from main-stream marketplaces, and calculate a volume-weighted base value.

The relationship between sale price and traits is learned by machine learning models from all the historical sales(excludewash trades). This floor price-based framework is suitable for any regression model. Normally, gradient-boosting models perform well in small data scenarios of the NFT market.

Compared with other NFT pricing models not using floor prices, GoPricing is always real-time because the floor price is real-time. This is especially useful in markets with extreme price changes.

Learn More

Algorithm Principle: Digging into the model, algorithm and advantages of GoPricing.
Evaluation Method: How to evaluate the performance of model.
Performance Monitor: How we monitor the perforamnce of model, like accuracy, stability and Interpretability.
Use Cases: Typical use cases of NFT pricing feature.
API Reference: Integrating NFT valuation feature to your Dapps using APIs.

Our premium appraisal model has already been published in Standford Blockchain Review.

Have a try！

You can now try GoPricing for free! Follow the steps below to register:

Register on NFTGo Developers
Obtain your private key through Dashboard
Access our pricing interface services through your private key