Hacked together scripts to rank nft items relative to their collection. There are a couple different strategies implemented:
-
STATISTICAL: Calculates the distribution of different traits for the collection, then calculates the probability that a specific nft was generated with the traits it has. Lower probability is considered more rare.
-
RARITY: This is based off the blog post found here: https://raritytools.medium.com/ranking-rarity-understanding-rarity-calculation-methods-86ceaeb9b98c
The example given in the article is implemented here in
MockStrategyto show it gets the same results suggested in the article.
Currently, the app pulls collection data from https://blockfrost.io/. To run the app, you'll need to get an api key and add it to environment.js.
To run on a specific project, you'll need the projects "name" provided in that projects strategy. You can pass the name with the parameter --nft. Ex:
npm run -- --nft=space-budz
By default, this will run both ranking strategies. To run a specific strategy, you can use the parameter --strat. Ex:
npm run -- --nft=space-budz --strat=rarity
npm run -- --nft=space-budz --strat=statistical
On the first run, collection data will be pulled from blockfrost, but will then be cached in .json on your machine. To re-pull (maybe more assets were minted), you'll need to delete the cached file in cache/.
I'm not a mathematician, just a coder and wanted a way I could rank projects I was getting into. If you're making purchasing decisions off my calculations, shame on you, and no coming after me if you lose money. As they say, I'm not a financial advisor.