Long Term Holders Bitcoin: LTH vs STH Cohorts Explained
Lesson 7 of 10 · Intermediate
A cohort groups coins, not people
Cohort analysis groups bitcoin's supply by coin age, meaning the time since each UTXO last moved. The first thing to understand is that this grouping applies to coins, not individuals. One person may hold both old coins and new ones.
- Short-term holder (STH): coins that moved recently.
- Long-term holder (LTH): coins that have not moved for a long time.
In the common definition used by Glassnode, the boundary between the two groups is about 155 days (nearly five months). The logic behind that number is statistical: the longer a coin stays still, the less likely it is to be spent soon, and beyond this period the probability of being spent drops markedly. The boundary is a convention, and different providers may define it slightly differently.
Each time a coin moves, its age resets to zero and it rejoins the short-term group, even if its owner only moved it from one of their wallets to another. That point becomes very important later.
The cost basis of each group
Using the same realized-price logic we saw in the MVRV lesson, an approximate cost basis can be computed for each cohort:
- STH realized price: the average price at which young coins last moved; an estimate of recent buyers' purchase cost. Because this group is more price-sensitive, the distance of market price from this level is often related to short-term market behaviour.
- LTH realized price: usually far below market price, because old coins are recorded at past prices. That is why long-term holders typically sit on large paper profits and are less often at a loss even in deep corrections.
Hypothetical example: if market price is $85,000 and the STH realized price is about $90,000, most recent buyers are on average at a loss, and this group may react to any rally by selling at breakeven. If market price is $100,000, the same group is in profit and its behaviour could differ.
Supply shifting between cohorts
The classic cycle pattern is described like this:
- In strong rallies: older holders sell part of their coins to new buyers. LTH supply falls and STH supply rises; this process is called distribution.
- In quiet or declining periods: coins held by new buyers gradually age and cross the 155-day boundary, so LTH supply rises again. This is called maturation, or accumulation.
Related tools:
- HODL waves: show the share of supply in different age bands.
- Coin days destroyed (CDD): when an old coin moves, its accumulated "coin days" are destroyed; a spike in CDD means old coins have moved.
- Dormancy: the average age of coins being spent; a high value means older coins are moving.
Custody and ETF effects: when 'ageing' and 'waking up' are false
All these indicators are built on coin age, and coin age resets to zero with every movement. That means:
- False awakening: when an exchange or custodian moves its old coins to a fresh wallet for security management, CDD spikes and it looks as if "old holders sold," though the owner has not changed.
- False rejuvenation: coins bought for an ETF and moved to a custodian have age zero and stay in the short-term group for 155 days, even if the ultimate owners, the fund's shareholders, intend to hold long term.
- Data revisions: according to CryptoDaily, Coin Metrics in September 2026 revised ETF wallet data for 19 months; corrections like this can change the cohort picture for the past too.
Long-term holders are also no longer only individuals. Companies and institutions may sell for their own financial reasons. According to CoinDesk, Strategy sold part of its bitcoin between May and August 2026, for the first time since 2022, to fund preferred-stock dividends. Such a sale appears as "distribution" in LTH data, but its motive differs from that of a retail holder who "sensed the top."
How to read cohorts cautiously
- Before interpreting a one-day CDD spike or a sudden fall in LTH supply, match it against the real transactions of that day.
- Separate multi-month trends of distribution and maturation from one-off movements.
- Treat the STH realized price as a contextual zone, not a firm support or resistance line.
- Where available, compare entity-adjusted versions with the raw version.
A useful rule of thumb: turn any narrative that begins "old holders are selling" into three questions. How many transactions are behind this number? Are the destinations exchanges or fresh unknown addresses? And did selling appear in the market over the following days? If the coins have not reached exchanges and no aggressive selling appeared, the "distribution" narrative is not yet proven. Conversely, if the fall in LTH supply continues for several weeks, is spread over many transactions and comes with inflows to exchanges, the distribution story has more support.
Common mistakes
- Assuming LTH and STH are groups of people rather than a grouping of coins by age.
- Interpreting every CDD spike as old holders selling, without checking custody moves.
- Overlooking that coins bought for ETFs count as "short-term" for 155 days.
- Using the STH realized price as a firm entry level or stop-loss.
- Assuming all long-term holders behave on one logic, when companies and institutions have different financial motives.
What do volume and data say here?
Cohort data tells you what kind of supply is moving; volume and order flow tell you how the market absorbed it:
- If LTH supply falls during a rally (distribution) while spot volume is high but price makes little progress, that is the "high volume, little progress" pattern, which in volume language signals heavy supply against demand.
- Near the STH realized price, watch spot CVD and volume: absorption and stabilisation, or a quick pass-through?
- Compare a CDD spike with the same day's exchange inflows and aggressive selling; without those two, the chance of an internal move is high.
These combinations shift probabilities; they do not guarantee an outcome.
📊 See real gold futures volume behind these ideas: free order flow course · today's gold volume map
Frequently asked questions
Who is a long-term holder of bitcoin?
In on-chain analysis, coins that have not moved for a long time, about 155 days in the common definition, belong to the long-term holder group. The definition is based on coin age, not the owner's identity.
What is the STH realized price used for?
It is an estimate of recent buyers' purchase cost and shows whether this group is on average in profit or at a loss. It is a contextual zone, not a firm level.
Why do ETFs distort holder data?
Because moving coins to custody resets their age to zero, and internal custodian moves may look like old holders selling.
Education only, not a recommendation to buy or sell; we never give signals. Price examples are illustrative. Test any method on a demo account first, with strict risk management.