2021-11-19
NFT: Standard Ranking vs Statistical Ranking
The Non-fungible Token (NFT) market is booming as celebrities and big companies are jumping into the scene. A piece of NFT can be sold as high as millions of dollars. The idea of owning a digital asset that lives on the blockchain has caught the interest of many.
Rarity
In a collection, there are thousands of images that are minted to be sold, but there are only a few that was sold with the highest value. For example, the famous NFT collection in Ethereum blockchain; CryptoPunks. There are about 10,0000 punks available, but the price for each punk can go as high $ 530 million, while some are only in the 5 digit area.
NFTs are differentiated using their uniqueness, or in NFT term, rarity. The rarer the item, the higher the price of the NFT would be. Although rarity is not the only thing that a collector should look at when purchasing NFT, it is still considered as one of the important factors.
So how do we determine the overall rarity of an NFT?
Determining rarity
Determining rarity can be considered as one of the problems that collectors face before buying their NFT. Different collectors will value different things. In CrepeToast NFT, we help to ease your NFT search by determining the rarity of NFTs listed by using 2 types of ranking, which are:
1. Standard ranking
For standard ranking, we use a method that calculates the rarity score of an NFT by using this formula:
The total Rarity Score for an NFT is the sum of the Rarity Score of all of its trait values. This calculation gives good result and it is used widely on NFT marketplace.
To give an example on how the calculation were made, we will use a mock up data for 2 pieces of NFT in the same collection.
| NFT \ Traits | Headband | Albino | Glasses |
| 1 | 0.3 | 0.7 | 0.1 |
| 2 | 0.2 | 0.6 | 0.6 |
To convert these data to rarity score, we will use the formula shown above and calculate the total sum of each trait.
| NFT \ Traits | Headband | Albino | Glasses | Total Score |
| 1 | 3.33 | 1.43 | 10 | 14.76 |
| 2 | 5 | 1.67 | 1.67 | 8.34 |
In standard ranking, the higher the rarity score of an NFT, the rarer it would be. Therefore in this case, NFT 1 is rarer than NFT 2.
2. Statistical Ranking
As for statistical ranking, it is obtained by multiplying all of its traits rarities together. The calculation is simple and straightforward. Formula for statistical ranking is as shown below:
For statistical ranking, we will also use another set of mock up data as example.
| NFT \ Traits | Mohawk | Light Skin | Beard | Total Score |
| 1 | 0.2 | 0.6 | 0.6 | 0.072 |
| 2 | 0.3 | 0.7 | 0.1 | 0.021 |
For statistical ranking, the lower the score, the rarer the NFT would be. Hence for this example, NFT 2 is rarer than NFT 1.
Standard vs Statistical Ranking
Based on the examples shown above, there are several differences that can be found in both of the ranking, which are:
| Standard | Statistical | |
| Formula |
The sum of Rarity Score for all traits value
|
Multiply all traits together |
| Calculation Process | Need to know all traits collection to calculate | Easier to calculate |
| Rarity | Higher score is rarer | Lower score is rarer |
| Focus | Emphasize to single rare traits and includes overall trait rarities | Emphasize the overall level of rarities of all traits |
As shown in the table above, there are pros and cons for both standard and statistical ranking. Therefore, collectors need to analyze the rarity of the NFT which they are interested in themselves before making a purchase.
Reference
1. Rarity.tools