Dark
Light
Banner

Trait Checker

Profile Pic

What is Rarity?

For NFT collections with different traits, the Raity score is a way of ranking the NFTs in a single collection to show how rare each one is - the rarer the combination of traits is, the more souhgt after the NFT is, which increases is value to collectors.

What Rarity Ranking Algorithmns do we use?

We provide two different rarity ranking algorithms, each is provided with and without normalization:

Statistical Rarity

Statistical Rarity multiplies together the rarity vale of each trait in an NFT to get a percentage rarity value for each NFT.

Point based Rarity

Points based rarity adds together the different rarity values of each trait in an NFT to give a score value for each NFT.

Official Rarity

There are advantages and disadvantages of each method and we know that each project has it's own preference for calculating rarity.
We're not aware of an official rarity algorith for Sorcies but the most commonly used type is "Normalized Statistical Rarity". If you are involved in Sorcies and would like to confirm your Official Rarity, please DM us from your official Twitter account.

What is Normalization?

Normalization allows us to weight the scores depending on how many traits exist in a category.

Example: A collection of 8000 NFTs has a Face Color category with 4 face colours and a Hair Color category with 20 hair colours.

Within the faces, there is an even split of 2000 NFTs in each color. In the Hair color, Black hair appears 2000 times in the collection and the other colors each show 300 times.

Without normalizatiom, the score for getting black hair in your NFT would be the same as for getting red skin: Normalization adjusts this to allow for the fact that within the hair category the chance of getting black hair is a lot higher than the chance of getting red skin is in the skin category.

Customized Rariry Scoring

Occasionally a project will have it's own strategy for rarity ranking - in these situations we will provide an additional sorting type based on their specific rankings.