Reading holder distribution

How many addresses hold a token says almost nothing. How much the largest hold, and what kind of addresses they are, says almost everything.

This lesson has had no expert review. It was written for this platform and against the evidence it cites; nobody has gone through it independently.

Learning objectives

You can clean a holder list so the remaining numbers mean something.
You can distinguish contract addresses from addresses with a holder.
You can say which statement a holder distribution supports and which it does not.

Check your prior knowledge

Answer these for yourself before reading on. Wherever you hesitate is where this lesson pays off.

What sits behind the largest address of any given token?
Where do tokens deposited at an exchange sit?
How do you tell a contract address from a holder's address?

Core concept

Fact

The largest addresses are usually not holders

The top of a holder list is typically contracts: trading pools, lending markets, lock-up contracts for unlocks, bridges. Those addresses hold other people's balances or none in an economic sense. Reading the list unchecked as a distribution describes the infrastructure and calls it concentration — or conversely misses a real concentration because it sits behind a contract.

Interpretation

Cleaning means attributing, not deleting

A usable distribution comes from assigning each large address to a category: protocol contract, exchange custody, lock-up contract, project treasury, or unknown. Nothing is deleted — the balances exist. Only the question moves: for a lock-up contract it becomes “when does it unlock”, for an exchange address “whose balance is this”, and for “unknown” it stays open.

Risk

The number of holders is the weakest figure

“50,000 holders” can be produced for free: sending dust to many addresses creates the number without changing anything. What is informative is the distribution: what share falls on the largest ten cleaned addresses, and how much on addresses below an economically relevant amount. The second group moves no price, however numerous it is.

Uncertainty

What the distribution does not say

Even a properly cleaned distribution says nothing about people: several addresses can belong to one party, and one address can hold for many. It also says nothing about intent — holding a lot does not mean someone will sell. What is defensible is only the statement about how few decisions would be needed to move a given share of the circulating supply.

Definitions

Contract address in the glossary: An address backed by program code rather than a key — it acts only when called.
Cleaned distribution: A holder list in which every large address has been assigned to a category.
Dust balance: A balance too small to have economic effect — often created by distribution campaigns.

Model

List the largest addresses until 80 % of circulation is covered
Assign each address: contract, exchange, lock-up, treasury, unknown
Carry locked balances separately, with their unlock date
Report the remainder as the cleaned holder share — with the count of unknown addresses

Cleaning a holder list in four steps — Step four names the uncertainty; without it the result looks more precise than it is.

Formulas

Cleaned share of the largest holders

share_top10 = sum of the 10 largest non-contract holdings / circulating supply

non-contract holdings: Balances of addresses not assigned to an infrastructure category
circulating supply: The freely movable quantity, excluding locked balances

Limit: Assigning categories relies on publicly known labels; an unlabeled contract address wrongly counts as a holder, and several addresses of one party count as several holders. The result is therefore a lower bound on actual concentration.

Worked example

The same list, read twice

Address 1: 22 % — trading pool
Address 2: 17 % — lock-up contract, unlock in 11 weeks
Address 3: 9 % — exchange custody
Addresses 4 to 13: 12 % together — unlabeled
Remainder: 40 % across about 48,000 addresses

Read unchecked: “the largest three hold 48 %”. Cleaned: address 1 is infrastructure, address 3 holds other people's balances, address 2 is locked and has a date. The ten largest actual holders are the unlabeled ones with 12 % together.

“48 % concentration” becomes 12 % among holders plus a dated event covering 17 %.

Reading: Both findings matter and they differ: the 12 % concerns ongoing selling pressure, the 17 % concerns a date. Adding them together turns two checkable statements into one unusable one.

Retrieval

Concentration is high.Nothing directly — it is infrastructure holding others' balances; the question moves to that pool's providers.The token is widely distributed.

Very — a high holder count evidences wide distribution.Barely — the number can be produced by sending dust; what is informative is the share of the largest cleaned holders.It is false.

Exercise on real data

Sort the guiding questions into those answerable from a holder list and those needing the emission schedule.

Dimension 5: tokenomics →

Application

You receive a holder list and are to turn it into two sentences for an analysis. How do you proceed?

Related case studies

Institutional reading

Asset management: Which holder metric sits in your papers — the count or the cleaned share?
Bank: How do you document an address that could not be assigned to a category?