Key Takeaways
- What it is: Cumulative coin days destroyed divided by cumulative coin days created, measured from genesis to today.
- Why it is bounded: A coin day can be destroyed only once and only after it is created, so the numerator can never exceed the denominator and the ratio lives between 0 and 1.
- Why it is slow: Today's activity is one day of numerator against a denominator holding every day the chain has ever existed. A single day can barely shift it.
- The neutral rate: The ratio holds flat only when the day's coin days destroyed equals liveliness multiplied by current supply. Below that it falls, above it rises.
- Quiet markets drift down: The denominator grows by the whole supply daily regardless of behaviour, so a quiet period pushes the series down by arithmetic, not by any change in conviction.
- Vaultedness is redundant: It equals 1 minus liveliness exactly. Plotting both is plotting one series and its mirror image.
Who This Guide Is For
This guide is for readers who have seen liveliness described as a measure of holder conviction and want to understand why the series moves the way it does. The answer is almost entirely arithmetic: both terms are running totals from genesis, and the metric's stiffness, bounds and drift all follow from that.
The underlying quantity, coin days destroyed, is defined on the coin days destroyed guide, which this page assumes rather than repeats.
Educational content about a measurement method. Not individualized financial advice. Every number below is hypothetical.
What Does Bitcoin Liveliness Measure?
Liveliness measures what fraction of all the coin days a network has ever produced have since been destroyed by spending. It is a cumulative accounting identity, not a snapshot of current behaviour.
Every coin outstanding produces exactly one coin day per day, so the network manufactures coin days at a rate equal to its supply. Some are later destroyed when the coins holding them are spent; some are never destroyed at all because their coins never move again. Liveliness is the running share in the first category.
What the two halves actually contain
- Numerator. Every coin day destroyed since the chain began, summed across every spend that has ever occurred. Nothing is ever removed from it. A burst of ancient-coin spending from years ago is still fully present in today's numerator.
- Denominator. The running integral of the supply curve. It increases by the full circulating supply every day, in a bull market, in a bear market, and on a day when the chain processes nothing at all.
Why the level says so little
Because the denominator carries the entire history of the chain, the current level is a weighted average of every regime the network has passed through, with the earliest years weighted just as heavily as this week. Two chains reporting the same liveliness have nothing in common except a coincidence of two integrals.
Vaultedness
Vaultedness is the complementary share: coin days created that have not been destroyed, divided by coin days created. Since every created coin day is either destroyed or not, the two shares sum to one, and vaultedness equals 1 minus liveliness by definition. It is a presentational choice, not a second measurement.
How Is Bitcoin Liveliness Constructed?
Two series are needed, both running from the chain's first block: daily coin days destroyed, and daily supply. Everything else is a running total and a division.
The history requirement
Liveliness cannot be computed from a recent window. A dataset that begins its index part-way through a chain's life produces a denominator missing every earlier day, and the resulting series is not a lower-quality liveliness but a different quantity that happens to look similar. Any provider publishing liveliness is implicitly claiming complete coverage back to genesis, which is worth verifying before comparing two published series.
How the issuance schedule shapes the denominator
Coin days created on a given day equal that day's supply, so the denominator is the area under the supply curve. On a chain whose issuance decays toward a cap, that area grows at an almost constant rate once issuance becomes small, since supply is nearly flat. On a chain with ongoing inflation the denominator accelerates, which pushes liveliness down over time even if spending behaviour is unchanged.
Permanently lost supply sets a ceiling
Coins that can never move again keep producing coin days every day and can never destroy any. They are pure denominator. If a fraction of supply is permanently inaccessible, the highest liveliness the chain can asymptotically reach is roughly one minus that fraction. The theoretical bound of 1 is unreachable on any real chain, and how far below 1 the real ceiling sits is unknowable.
Genesis and early-history artifacts
In a chain's first months the denominator is tiny, so a single spend can move the ratio by a large amount. Reading a long history as one continuous series therefore conflates a noisy early regime with a stiff later one. Standard reorganization, revision, and coverage discipline for the input series is covered in the on-chain data quality methodology guide.
Formula and Measurement Logic
Liveliness = (cumulative coin days destroyed) / (cumulative coin days created)
The useful form is not the ratio itself but its daily change, which can be derived directly. Writing C for cumulative destroyed, D for cumulative created, c for today's coin days destroyed and s for today's supply, one day's change in liveliness is (c - L × s) / D, where L is yesterday's liveliness and D is today's cumulative denominator.
That expression contains everything worth knowing about the metric:
- The neutral rate is
L × s. Spend exactly that many coin days today and liveliness does not move. The threshold rises as liveliness rises, so a chain at a high level needs more daily spending merely to stand still. - The denominator of the change is the whole history. Every additional year makes an identical day of spending move the metric less. The stiffness is built into the definition, not chosen.
- Falling is the default in quiet conditions. Ordinary quiet days destroy far fewer coin days than
L × s, so the ratio slides downward with no change in holder behaviour required.
| Property | Consequence | Practical effect |
|---|---|---|
| Both terms are cumulative | Nothing is ever forgotten | An old spending burst still sits in today's numerator |
| Numerator cannot exceed denominator | Bounded between 0 and 1 | Percentage-style intuition works; threshold-crossing intuition does not |
| Denominator grows by supply daily | Growth continues through total inactivity | Quiet periods produce a downward drift |
| Denominator grows with chain age | Responsiveness decays | The same event moves the series less every year |
| Lost supply is denominator-only | Effective ceiling below 1 | The attainable range is narrower than the nominal range |
| Issuance schedule shapes the integral | Chain-specific baseline | Levels are not comparable across networks |
How Should Bitcoin Liveliness Be Interpreted?
Read the slope over multi-month windows, and read it against the neutral rate rather than against zero. A rising series means spending has been running above liveliness multiplied by supply for a sustained period; a falling series means it has been running below.
Questions the level cannot answer
- Is this chain's supply held tightly? The level reflects issuance history and chain age at least as much as holding behaviour, so the same figure means different things on different networks.
- Is this level historically extreme? Today and a level from several years ago are ratios over denominators of very different length.
- Did something happen today? A single day, however dramatic, is one term against a running total spanning the chain's entire existence.
What the slope does support
A sustained multi-month rise indicates that older supply has been moving faster than the neutral rate for long enough to overcome the denominator's growth, which takes persistent activity rather than a single event. A sustained fall indicates the opposite, with the caveat that a genuinely inactive market produces a fall automatically. Telling a meaningful decline from arithmetic drift means comparing observed daily coin days destroyed against L × s.
Liveliness contains no price information whatsoever. A rise says old coin days were destroyed, which is compatible with selling, with moving to new custody, with a wallet upgrade, or with an exchange reorganizing its reserves. Pairing it with a price-aware series such as SOPR is how the intent question gets addressed, and even then only partially.
Step-by-Step Workflow
- Confirm the provider's series runs from the chain's first block: a truncated denominator is a different quantity under the same name.
- Compute the current neutral rate, liveliness multiplied by current supply, as a coin-days-per-day figure.
- Compare recent daily coin days destroyed against that rate rather than against its own past values.
- Look at slope over months, not level, and not day-to-day change.
- Check how much of the recent numerator came from a few very large old spends.
- Discount early-history volatility as a small-denominator artifact.
- Do not compare the level against another chain, or against a historical threshold without noting how much the denominator has grown since.
- Ignore vaultedness if you have already looked at liveliness. It is the same series inverted.
Worked Hypothetical Scenario
Every figure here is invented to demonstrate the arithmetic and describes no real network.
Take a hypothetical chain with a constant supply of 20,000,000 units. It creates 20,000,000 coin days per day, so 140,000,000 per week. Today it stands at 24,000,000,000 cumulative coin days destroyed against 40,000,000,000 cumulative coin days created, giving a liveliness of exactly 0.6000. The neutral rate is therefore 0.6000 × 20,000,000 = 12,000,000 coin days per day, or 84,000,000 over a week.
Now run three different weeks forward. In all three the denominator gains the same 140,000,000, because the denominator does not care what happened.
| Week | Coin days destroyed | Cumulative destroyed | Cumulative created | Liveliness | Change |
|---|---|---|---|---|---|
| Starting point | 24,000,000,000 | 40,000,000,000 | 0.6000 | ||
| Quiet | 20,000,000 | 24,020,000,000 | 40,140,000,000 | 0.5984 | -0.0016 |
| Neutral | 84,000,000 | 24,084,000,000 | 40,140,000,000 | 0.6000 | 0.0000 |
| Busy | 500,000,000 | 24,500,000,000 | 40,140,000,000 | 0.6104 | +0.0104 |
The quiet week averaged about 2.86 million coin days per day against a neutral rate of 12 million, so liveliness fell. Nothing changed about holder conviction; the denominator simply grew by 140,000,000 while the numerator grew by 20,000,000. The busy week destroyed twenty-five times as many coin days as the quiet week and still moved the metric by only 0.0104, roughly one point on a scale that runs from 0 to 1. That is why the level looks nearly flat on any chart covering more than a year.
The change formula reproduces both exactly. For the quiet week, (20,000,000 - 84,000,000) divided by 40,140,000,000 gives -0.00159. For the busy week, (500,000,000 - 84,000,000) divided by 40,140,000,000 gives +0.01036.
The same week on an older chain
Now suppose the same chain, same supply, same liveliness of 0.6000, but five times as much accumulated history: 200,000,000,000 cumulative coin days created against 120,000,000,000 destroyed. Run the identical busy week through it.
| Chain history | Cumulative created before | Busy week coin days destroyed | Liveliness after | Change |
|---|---|---|---|---|
| Younger | 40,000,000,000 | 500,000,000 | 0.6104 | +0.0104 |
| Older | 200,000,000,000 | 500,000,000 | 0.6021 | +0.0021 |
Identical spending, identical supply, identical starting level, and one fifth of the movement. The only thing that changed is how much history sits in the denominator. Any threshold or percentile applied to liveliness has to account for this: a move that was ordinary early in a chain's life is a large move later.
What Can Make the Interpretation Wrong?
- Reading a quiet-market decline as capitulation ending. The decline happens automatically when daily coin days destroyed sit below liveliness multiplied by supply, which is most days.
- Treating the level as a conviction gauge. The level encodes issuance schedule, chain age, and lost supply alongside behaviour, and cannot be decomposed into them from the outside.
- Comparing chains. Two networks' denominators are integrals of different supply curves over different lifespans. The ratio is on a chain-specific scale.
- Reusing a historical threshold. The metric stiffens every year, so a level once reached by ordinary activity now requires far more of it.
- Citing vaultedness as corroboration. It is 1 minus liveliness. Two views of one number are not two pieces of evidence.
- Reading a single day or week. One period against a genesis-to-date denominator is close to invisible, which is why almost all short-term liveliness commentary describes noise.
- Ignoring who spent. A single exchange migrating cold storage can supply a large share of a period's numerator without representing any broad behaviour.
Cross-Network and Provider Comparison
Liveliness travels across networks worse than most on-chain metrics, for two independent reasons.
The first is the denominator. It is the area under each chain's own supply curve since its own genesis, so it encodes the issuance schedule and the chain's age. A network with heavy ongoing issuance accumulates denominator faster and reports lower liveliness for identical spending behaviour, and no normalization removes that without discarding the definition.
The second is the numerator. Coin age is a native property on a UTXO chain, where each output records its creating block, but on an account chain it has to be reconstructed by imposing first-in-first-out or average-age accounting on a balance that carries no lot structure. The choice of convention changes coin days destroyed, and therefore changes liveliness, without any difference in the underlying chain data. The UTXO versus account model guide covers that asymmetry in detail.
Contract-heavy networks add a third distortion. Every deposit into a liquidity pool, lending market, staking contract, or bridge escrow moves coins and destroys their accumulated age, so a chain with an active decentralized finance ecosystem generates numerator from protocol plumbing rather than from holders deciding anything. Whether staked balances freeze or reset their age is a provider convention, rarely stated prominently, that changes the series materially.
Advanced Analytical Methods
Plot the excess instead of the ratio
The quantity c - L × s, daily coin days destroyed minus the neutral rate, is the numerator of the daily change and is far more readable than liveliness itself. It is centred at zero by construction, it responds immediately, and it does not stiffen with chain age. Anyone asking whether old supply is moving faster or slower than the network's own baseline should read this series rather than the ratio it feeds.
Windowed liveliness
Replacing both cumulative sums with rolling sums over a fixed window gives a responsive series that no longer carries the whole history. It gives up the clean bounded interpretation and comparability with published liveliness, and gains the ability to move. The window length must be published: it fully determines sensitivity.
Entity-adjusted numerator
Suppressing coin days destroyed by identified internal transfers removes custodial reorganizations from the numerator. It is the single most useful filter available on the metric: a large exchange migration can otherwise dominate a period's numerator on its own.
Pairing with a normalized spend-age series
Liveliness normalizes by cumulative coin days created, which is why it is slow. Coin dormancy normalizes the same numerator by the same day's transfer volume instead, which makes it fast and heavy-tailed. The two are complementary rather than confirmatory: they divide the same input by very different denominators and answer different questions.
Practical Checklist
- I confirmed the series is cumulative from genesis and not from an index start date.
- I calculated the current neutral rate as liveliness multiplied by supply.
- I compared recent daily coin days destroyed against that rate, not against zero.
- I read the slope over months rather than the level.
- I did not compare the level to another chain or to a threshold from an earlier era.
- I noted whether the numerator is entity-adjusted.
- I did not treat vaultedness as separate evidence.
Conclusion
Liveliness is a ratio of two running totals, and almost everything readers find puzzling about it comes from that shape. The single most useful thing to do with it is to stop reading the level and start comparing daily coin days destroyed against liveliness multiplied by supply, the threshold the ratio is silently measuring against.
Frequently Asked Questions
Why does liveliness barely move even during heavy spending?
Because the denominator holds every coin day the chain has ever created and one day of spending is a tiny addition to the numerator by comparison. In the hypothetical worked above, a week destroying twenty-five times as many coin days as a quiet week moved the ratio by about 0.010. The stiffness is a property of the definition, not a smoothing choice made by a data provider.
Can liveliness fall while people are still spending coins?
Yes, and this is the normal case. The ratio holds flat only when the day's coin days destroyed equals liveliness multiplied by current supply. Ordinary spending sits well below that threshold, so the series slides downward through most quiet periods without any change in holder behaviour. Reading such a decline as increasing conviction attributes to holders something the arithmetic did on its own.
What is the neutral level of daily coin days destroyed?
It equals current liveliness multiplied by current supply. On a hypothetical chain at 0.6000 liveliness with 20,000,000 units outstanding, that is 12,000,000 coin days per day. Days above it push liveliness up, days below push it down, and the threshold itself rises as liveliness rises, so a chain at a higher level needs more spending merely to hold still.
Why does liveliness become less responsive as a chain ages?
The daily change equals the day's excess over the neutral rate divided by the cumulative denominator, and that denominator grows every day forever. In the worked example, an identical busy week moved liveliness by 0.0104 on a chain with 40 billion cumulative coin days created and by 0.0021 on one with 200 billion. Any historical threshold applied to the series has to account for this or it will misread an ordinary early move as an extreme later one.
What is vaultedness, and how does it relate to liveliness?
Vaultedness is the complement of liveliness: where liveliness is the share of accumulated coin days that have been destroyed, vaultedness is the share that has not. The two sum to one by construction, so they carry identical information presented from opposite directions. Vaultedness is sometimes preferred because it rises when supply is being held rather than falling, which some find easier to read alongside a price chart. Neither adds anything the other lacks.
Is liveliness bounded, and what do its limits mean?
It sits between zero and one. A value of zero would mean no accumulated coin days have ever been destroyed, which describes a network where nothing has been spent. A value approaching one would mean nearly all accumulated coin days have been destroyed, describing supply that turns over constantly with no long-held component. Real networks sit between the extremes and move slowly, which is why the level of liveliness matters less than the direction and the rate of change.
How does liveliness differ from dormancy?
Both are built from coin days destroyed and they answer different questions. Dormancy is a daily ratio describing the average age of coins that moved on one day, so it is volatile and reacts immediately to a single event. Liveliness is a cumulative ratio over the entire history of the chain, so it changes slowly and describes the balance between spending and holding across the network's whole life. Dormancy is a flow measure, liveliness is a stock measure, and the two are not substitutes.
How do lost coins affect liveliness?
They hold it down permanently. Coins whose keys are gone keep accumulating coin days that can never be destroyed, so they add to the denominator without ever contributing to the numerator. On a network with a substantial unrecoverable share of supply, the ceiling on liveliness is lower than one and drifts lower still as those coins age. This makes any absolute interpretation of the level unreliable, and it strengthens the case for reading changes rather than levels.
Can liveliness be compared between two chains?
Not usefully as a level. Liveliness depends on how long a network has existed, how its supply was distributed, how much of that supply is lost, and whether its data model supports coin ageing at all. Two networks with identical holder behaviour but different ages will report different values. What can be compared is the shape of the change over a common window, and even that requires both networks to have been through comparable market conditions during it.
References
These sources should be reviewed during editorial verification. They support data structures and methods, not the hypothetical conclusion. Provider formulas, chain rules, and APIs can change. Confirm current documentation before publication.
- Bitcoin Developer Guide: Transactions: the output model that makes coin age a native property and coin days destroyed observable.
- Bitcoin Developer Guide: Block Chain: block height and confirmation structure, the basis for converting age into days.
- Glassnode: Metric Catalog: where liveliness, vaultedness, and related age-weighted series are catalogued.
- Coin Metrics: Network Data Glossary: supply definitions, which determine the rate at which coin days are created.
- Coin Metrics API v4: frequency and revision behavior, which matter for a series computed from genesis.