Trait Rarity

Trait rarity is the statistical measure of how frequently a specific visual attribute — such as a background color, eye shape, hat type, body color, accessory, or special effect — appears within a generative NFT collection, expressed as a percentage of total supply (e.g., “Gold Background appears on 2% of tokens”), used as the primary input for calculating rarity scores and ranking NFTs within their collection from most common to most rare, and applied by collectors, traders, and rarity platforms to assign price premiums to statistically uncommon tokens under the widely-held belief that scarcity within the collection’s own trait space creates relative collectible value analogous to rare trading card variants or limited print runs. Understanding trait rarity requires recognizing that numerical rarity (low frequency) and market-perceived rarity (desirability) often diverge: a mathematically rare trait combination may not actually command a premium if collectors don’t find it aesthetically appealing, while a more common but visually striking trait may trade above its statistical rarity would suggest.


How Traits Are Structured in Generative Collections

Trait Categories (Typical PFP Collection)

“`

Token #1234 traits:

Background: Blue (12% of collection)

Fur: Brown (17% of collection)

Eyes: Laser Eyes (3% of collection)

Mouth: Grin (15% of collection)

Hat: Seaman’s Hat (4% of collection)

Clothes: Striped Tee (8% of collection)

Earring: None (60% of collection)

“`

Each trait option has a defined rarity weight set by the artist/algorithm.

Special Traits

  • 1/1 Traits: Traits that appear on exactly 1 token; typically command extreme premiums
  • Rainbow/Holographic: Special visual effects; rare by design

Rarity Tiers (Common Terminology)

Tier Occurrence Description
Common >10% Most tokens share this; negligible premium
Uncommon 5–10% Slight premium over floor
Rare 1–5% Meaningful price premium (2–10× floor)
Epic/Ultra Rare 0.1–1% Significant premium (10–50× floor)
Legendary/1/1 <0.1% / unique Maximum premium; often 100× floor or more

These tiers are informal; different projects and rarity tools use different cutoffs.


Trait Rarity vs. Overall Rarity

An individual trait’s rarity ≠ the token’s overall rarity. A token with multiple rare traits compounds its rarity:

“`

Token A:

Gold Background: 1.5% (rare)

Alien Fur: 0.5% (ultra rare)

Laser Eyes: 1.2% (rare)

→ Very rare overall (multiple rare trait combination)

Token B:

Purple Background: 0.8% (rare)

Brown Fur: 18% (common)

Normal Eyes: 22% (common)

→ Less rare overall despite one rare trait

“`

Rarity calculators combine all trait frequencies to produce an overall score.


Trait Frequency in Metadata

Trait data is stored in NFT metadata JSON files:

“`json

{

“name”: “Ape #5241”,

“attributes”: [

{“trait_type”: “Background”, “value”: “Gold”},

{“trait_type”: “Fur”, “value”: “Alien”},

{“trait_type”: “Eyes”, “value”: “Laser Eyes”},

{“trait_type”: “Mouth”, “value”: “Grin”},

{“trait_type”: “Clothes”, “value”: “Tuxedo”}

]

}

“`

Rarity tools read all token metadata → compute frequency of each trait value → assign scores.


Trait Desirability vs. Statistical Rarity

Two tokens can have the same statistical rarity but very different market values:

“`

Token X: Statistical rarity rank #50

Traits: “Beige Background” (0.5%), “Messy Hair” (0.8%)

Market reception: underwhelming visuals

Secondary price: 2× floor

Token Y: Statistical rarity rank #500 (less rare on paper)

Traits: “Gold Background” (2%), “Alien” (0.5%) — visually striking

Market reception: iconic look, heavily tweeted

Secondary price: 8× floor

“`

Key insight: Trait desirability (how cool/iconic collectors find the combination) often matters more than pure statistics.


Famous Rare Traits

BAYC

  • Laser Eyes: Multiple types; eye traits generally command premium
  • Crowns/Kings: Rare hat; iconic look

CryptoPunks

  • Ape Punk: 24 total
  • Zombie Punk: 88 total
  • Beanie hat: 44 total

Azuki


History

  • 2017: CryptoPunks establishes type rarity (9 Aliens, 24 Apes, 88 Zombies vs. 9,869 Humans) as the first NFT rarity framework
  • 2021: Generative PFP collections formalize multi-trait rarity systems; Rarity.Tools and Rarity Sniper launch to calculate and display rarity rankings
  • 2021 Q4: Trait rarity becomes primary price driver in secondary market; “floor sweeping” vs. “rarity sniping” emerges as two buyer strategies
  • 2022: Statistical rarity vs. cultural desirability tension becomes widely discussed; community debates which matters more
  • 2023–2024: Open edition and 1/1 art NFTs bypass trait rarity entirely; rarity mechanics remain central to generative PFP collections

See Also