A golden Bitcoin coin inside an illuminated glass case on a pedestal, still and protected, with a candlestick chart in gold and blue tones in the background
Original illustration generated with AI — not representative of real data.

Bitcoin Long-Term Holder Supply: Conviction, Not a Price Call

NodeWitness Team 5 min read

Long-term holder supply rising is usually reported as a bullish call. It measures conviction accumulated up to today, not what happens next, and treating those as the same thing is the most common misread of this data point.

Some version of "long-term holders aren't selling" shows up in the on-chain news cycle almost weekly, tied to whatever the metric printed that day and framed as a reason price should move. The underlying data is real, it's the same thing HODL waves measure, supply grouped by how long each coin has sat unmoved, but the leap from "this supply isn't selling" to "so price will rise" is exactly the part the data doesn't support.

What the headline actually maps to

This piece pairs with HODL waves: what they are and how to read them, which has the full mechanics: the exact age bands, the 155-day line between short- and long-term holders, and how we compute it from our own node. This one is about something narrower: not the mechanism, but what the behavior it describes actually means, and where the "so price goes up" leap breaks down.

In practice, "holders aren't selling" means the share of supply sitting in the oldest cohorts (155 days or more) is staying flat or growing. That's a verifiable fact about the past, how many people have held on so far. It isn't a forecast of what happens from here.

Conviction and price are two separate questions

A large share of supply sitting unmoved for a long time measures conviction, not future price. Someone who didn't sell through a long, painful drawdown probably isn't going to sell over a smaller one either, they've already proven they can hold. That lowers potential selling pressure from that specific supply. It says nothing about whether enough new demand shows up to push price higher.

How many people are willing to sell and how many people want to buy are two independent variables. Long-term holder supply only measures the first one.

A market can hit a record share of dormant supply and still trade sideways, or keep falling, simply because new buyers aren't showing up in enough size. An absence of sellers doesn't manufacture demand on its own; they're different mechanisms inside the same market.

What this signal can't tell you

  • It doesn't say price will rise. Holders can keep not selling for months while price drifts sideways or keeps falling from weak demand, not from excess supply.
  • It doesn't separate conviction from lost coins. Part of the "unmoved" supply is coins gone for good (misplaced keys, old inaccessible addresses) and they count the same as patient holders in the same age bands.
  • It's not a reliable standalone signal. It gets used combined with other indicators, never alone: it's one of two sub-signals feeding the nodo_propio component of the NodeWitness Cycle Score, alongside wallet-size distribution.
Bitcoin supply grouped by holding age, from under a week to more than two years unmoved, with the long-term holder band highlighted
Original diagram illustrating the age-cohort concept — real supply computed directly from our own node's UTXO set, not an invented figure.

Why most coverage of this stops at the day's print

Explainers of the underlying concept already exist in English. What's harder to find is coverage that separates the mechanism from that day's chart. A lot of what circulates about long-term holder supply is a same-day rewrite: today's figure, a headline framing it as a bullish or bearish tell, and a close that implies a price direction. It's news-cycle content reacting to today's print, not an explanation of what the data structurally can and can't support, and it rarely spells out the "this doesn't guarantee price direction" part covered two sections up.

How we check this against real data

Our Score doesn't stop at the "holders aren't selling" narrative: the own-node sub-signal compares real snapshots taken directly from our Bitcoin node's UTXO set, not a third-party API. Today that comparison is sharper on the wallet-size cut (how much each cohort's share of supply moves between two dates, shrimp to whales) than on the age cohort itself. HODL waves still publish as a single snapshot refreshed weekly, without a historical series rolling backward in time yet (see the HODL waves article for the detail). That's a real limitation, stated plainly: today we can verify "who holds the supply, and whether that share is shifting" better than "how long each holder has held it," even though both questions point at the same underlying idea: whether Bitcoin is redistributing or concentrating into hands that aren't letting go.

The one real comparison we've run so far, between two snapshots taken a few weeks apart, shows exactly why a single day's number can mislead: whales (over 10,000 BTC) lost 4.86% of their share of supply and sharks (100-1,000 BTC) lost 5.90%, while orcas (1,000-10,000 BTC) gained 18.20% and every smaller cohort gained 20-42%. Read as a snapshot, that's "the largest holders are distributing." Read next to the same-period holder-supply data, it's closer to a redistribution toward mid-sized wallets, a different story than either "whales are selling" or "holders aren't selling" tells on its own, and neither headline alone captures it. Neither cut replaces the other; they answer different questions about the same underlying supply, and reading only one of them is how a single-snapshot comparison gets oversold as more certain than it actually is.

Want to see holder distribution and HODL waves with real data from our own node, updated weekly? Visit our Supply and Holders section, free, no sign-up required.

Last updated: August 31, 2026