Background

NFT price oracle offers real-time pricing information for NFTs and is considered the foundational building block for NFT financial primitives (lending, indices, derivatives…etc.)

However, NFTs are illiquid and unique by nature (1 NFT ≠ 1 NFT)… which makes pricing/ valuing NFTs very difficult. Protocols like Upshot, NFTBank, and Abacus can appraise individual NFTs based on rarity and historical sales. This pricing information serves the P2P loans market well, but it doesn’t help with collection-wide valuations.

In the current NFT speculation environment, there’s one valuation metric that the majority have adopted - the floor price. The floor price is simply the lowest seller asking price for an NFT in a given collection across marketplaces. The floor price has become the status quo in pricing the market cap of NFT collections.

For the use cases of NFT financialization, there are quite a few problems with using marketplace floor price as an oracle:

Introducing nftperp’s True Floor Price

We define our True Floor Price as a tamper-proof NFT price oracle with robust statistics and backed by sales between real buyers and sellers in a given NFT collection.

The overview of our True Floor Price computation method is outlined as follows:

  1. Collecting and parsing on-chain/ off-chain NFT transaction events on top NFT marketplaces.
  2. Determine data eligibility based on transaction event type, token IDs, and wash trade detection.
  3. Filtering extreme outliers and probable outliers using statistical methodologies and volatility scoring.
  4. TWAP True Floor Price

<aside> 💡 Our True Floor Price methodology is inspired by DropsDAO’s NFT oracle model. You’ll see some similarities but we have made adjustments to fit our protocol’s risk profile.

</aside>

Data Pipeline

We query, filter and parse sales event data from OpenSea, LooksRare, Sudoswap, Blur and X2Y2 API.

Data Eligibility