NFT industry is going to the moon nowadays! We hear many people talking about NFT’s future and about how much money they made selling NFTs. And thanks to blockchain technology, you don’t need any paperwork or a lot of money to sell your NFTs, too.

Building an NFT Collection is not that hard. It takes three steps: designing, marketing, and minting. It is better if you build up a team of 3 people and split the tasks. But you can do all of them by yourself. You just have to watch some YouTube videos for each step.

I have built my NFT collection by myself. I did that because I want to understand and try the complete process. While I’m doing that, I wrote every new lesson I had learned, so I can benefit from them in my future NFT projects.

Unlike math, we all know that measuring an artwork is subjective. However, sometimes artists take advantage of math too in order to create more elegant art. In the Mona Lisa, Leonardo da Vinci used the Golden ratio to draw the body and facial proportions.

Therefore, I share with you 4 calculations that will help you improve Your NFT art:

1. The number of NFTs in a collection

We see hundreds of NFT Collections show up every day. Most of them have around 10,000 unique NFTs. Of course, they aren’t totally unique, because all these NFTs share a main character in common, but they differ in some details.

Take the famous Crypto punks collection as an example. The main character has specific layers: hair, eyes, mouth, etc. So, Every NFT in the collection comprises the same layers but in different variations.

• the hair layer has variations of wild hair, a cap, or no hair.
• the mouth layer has variations of cigarette, medical mask, or smile.

So How many NFTs you can generate using these variations? Well, using The Combinations:

The number of NFTs= ∏ (The number of each layer’s variations)

This equation means that if you want to get the number of NFTs in the collection, take the number of variations of each layer and multiply them together.

For example, if you have a hair layer with 5 variations, eyes layer with 6 variations, and mouth layer with 10 variations, then you can generate 5*6*10=300 unique NFTs.

Hint: to make 10,000 NFTs, you just need 4 layers with 10 variations each!

2. Layers and traits rarities

Rarity factors are the most powerful things you can add to your NFT collection. The rarest NFTs in your collection will be the most valuable ones in the marketplace.

To make some NFTs rarer than others, you need to set rules for their variations (sometimes called traits). You can call these rules a trait’s rarity. And combining rare traits will generate a rare NFT (I will talk about how to score your NFT in the next section.)

these rules may look like these:

• 50% of NFTs have black hair.
• 30% of NFTs have brown hair.
• 20% of NFTs have white hair.

This will make any NFT that has white hair more valuable than the rest. The following equation can calculate a trait’s rarity

Trait’s rarity = (The number of NFTs with that trait / Total Number of NFTs) * 100

But sometimes you may need some NFTs without a certain layer (no hair, for example). well, this can be specified with the layer’s rarity. A 100% layer’s rarity means any NFT in the collection has one of the layer‘s traits. And 0% layer’s rarity means that this layer won’t appear in the entire collection.

Layer’s rarity = (The number of NFTs with that layer / Total number of NFTs) * 100

Then your new trait’s rarity can be calculated with this:

Trait’s rarity = Trait’s rarity * (Layer’s rarity / 100)

Make sure to set up a list with all your traits’ rarities. This will help you whether you hire a programmer to generate the NFTs or do it yourself.

3. Rarity Score

As I mentioned early, NFT with the rarest trait is the rarest. But this is only true if we have one layer of traits. We need a more complex way to determine NFT’s rarity that has many layers of traits. And here it comes the rarity score methods.

A rarity score is a number that measures how rare an NFT is within a collection. Often, the NFT with the highest score is the rarest one.

There are many methods to calculate the rarity score for an NFT. The most common way, which rarity.tools applies it, is using trait score:

Trait’s score = 100 / Trait’s rarity

And then comes the rarity score for an NFT:

NFT’s rarity score = ∑ Trait score of that NFT

That means the rarity score for an NFT is equal to the sum of all its trait scores. Rarity.tools has a great article about this method and why it is the most effective way to determine the NFT’s rarity.

4. Rarity Distribution

This is more related to geometry than real calculation. But it is a very important way to analyze your overall collection rarity.

You can illustrate your rarity distribution by drawing a graph. The x-axis of the graph represents your NFTs (Assets), whereas the y-axis of the graph represents their corresponding rarity scores.

This is just like the Exponential growth graph. And this is the type of distribution you want to achieve with your rarity settings. The vast majority of your NFTs are common in the collection, and a very few NFTs are very rare.

These are the most math-related problems I have faced with my NFT project. It’s really fun to apply some math to art. Though, it took me a lot of time and effort to configure these calculations manually. So, it is better to use software like NFT-generator. It can generate your 10,000 NFTs with an accurate calculation of layers and traits rarities. And It also has great analytics tools for rarity scores and rarity distribution graphs.