Cycle dashboard/Indicators/MVRV Z-Score
Bitcoin on-chain indicator
Bitcoin MVRV Z-Score
The MVRV Z-Score asks one question: how far has Bitcoin's market value stretched away from what the market actually paid for its coins? It takes the gap between market cap and realized cap and divides it by the standard deviation of market cap, so the answer comes out in standard deviations rather than dollars — which is what makes a 2013 reading and a 2025 reading comparable at all.
Reading the latest public snapshot. This page shows the raw MVRV Z-Score value in its own unit, not a normalised score.
- BTC price that day
- –
- Historical percentile
- –
- Full-history range
- –
- This cycle's peak
- –
- Site cycle zone
- –
- Cycle Compass verdict
- –
Public data refreshes every six hours.
Bitcoin MVRV Z-Score, full history in raw σ units
loading history…MVRV Z-Score (left axis, raw units) BTC price (right axis, log) 0 · 3.5 · 7.0 reference levels
Loading the public daily series…
What the MVRV Z-Score is, and exactly how it is computed
Three quantities go in, and CryptoTIDE derives all three from free CoinMetrics Community fields
(CapMrktCurUSD, CapMVRVCur). Nothing here comes from a paid or licensed feed.
Market capitalisation is spot price multiplied by circulating supply — what the whole network is worth right now. Realized capitalisation values every coin at the price it last moved on-chain instead of at today's price, which makes it a rough aggregate cost basis for the whole supply. The Z-score then expresses the difference between the two in standard deviations of market cap, using an expanding window: every market-cap observation from the start of the series up to that day, with a minimum of 50 observations before a value is published.
realized cap = market cap ÷ MVRV ratio MVRV Z-Score = (market cap − realized cap) ÷ σ(market cap over all history to date)σ is the population standard deviation. Because the window expands, the denominator only ever grows — a fact that matters a great deal for reading the chart, and one we come back to under limits.
Approximation disclosure
CryptoTIDE computes MVRV-Z in house from public market-cap and MVRV-ratio fields. It is a transparent approximation, not a replica of any commercial vendor's number. Dashboards that run proprietary realized-cap pipelines, or that use a sample rather than population standard deviation, or a fixed rather than expanding window, will print a different level for the same day. Expect small differences recently and larger ones in the earliest years. The same disclosure is in the project's methodology and data inventory.
The very first published readings are the least trustworthy of all: in October 2010 the expanding window held barely fifty market-cap observations, so σ was tiny and the score exploded to 11.75 on a $0.08 Bitcoin. That is a small-sample artefact of the formula, not a market event.
How the levels are read
CryptoTIDE does not act on a single number, and neither should a reader. The score feeds two places: the 0–1 TIDE valuation lens (normalised across 0 to 7) and the Cycle Compass evidence families, where it is grouped with MVRV ratio, NUPL and realized price into one realized-cost-basis family, because four views of the same underlying data should not get four votes.
| Band | What it describes | Role in CryptoTIDE | Share of all days since 2010 |
|---|---|---|---|
| Below 0 | Market cap sits under realized cap: the market in aggregate is underwater on its coins. | Counts toward the bottom-side realized cost-basis family | 13.4% |
| 0 – 2.5 | A modest premium over aggregate cost basis. | No family flag | — |
| 2.5 – 3.5 | Premium widening. | Top-risk watch | — |
| 3.5 and above | Historically wide stretch above cost basis. | Counts toward the top-risk realized-profit-stretch family | 10.4% |
| 7.0 and above | The extreme tail of the whole series. | Top-risk pressure flash level | 1.4% |
Those percentages are the honest context that a threshold on its own hides: "above 3.5" is not a rare event, it has happened on roughly one day in ten since 2010.
What every Bitcoin cycle actually printed
Here is the score at each cycle's own extreme, computed from the same public series this page charts.
| Cycle | Peak MVRV-Z | BTC price then | Post-peak trough | BTC price then |
|---|---|---|---|---|
| 2011 cycle | 11.75 2010-10-08 | $0.0800 | -0.69 2011-10-19 | $2.22 |
| 2013 cycle | 10.66 2013-04-09 | $231 | -0.60 2015-01-14 | $176 |
| 2017 cycle | 10.09 2017-12-07 | $17,032 | -0.49 2018-12-15 | $3,185 |
| 2021 cycle | 7.15 2021-02-21 | $57,501 | -0.36 2022-11-09 | $15,758 |
| 2025 cycle (open) | 3.35 2024-12-16 | $105,856 | 0.24 so far 2026-06-06 | $60,772 |
Cycle windows run low-to-low, using the same boundaries as the dashboard's cycle map: 18 Nov 2011, 14 Jan 2015, 15 Dec 2018 and 9 Nov 2022. The trough column is the lowest reading after that cycle's peak. The table is recomputed in your browser from the published series, so it stays correct as new data arrives.
Peak-to-peak, the MVRV Z-Score has fallen every cycle: 11.75 → 10.66 → 10.09 → 7.15 → 3.35. Each cycle top has printed a lower extreme than the one before it, so a level that was ordinary in 2013 has not been reached since.
The troughs have compressed too, and further than most readers expect: −0.69, −0.60, −0.49, −0.36 and, so far in the open cycle, a low of +0.24 that has not yet crossed below zero at all. Both tails are shrinking toward the middle.
One more detail from the current cycle that a normalised gauge hides entirely: the MVRV-Z high of 3.35 was set on 16 December 2024, roughly ten months before the price all-time high of $124,824 on 6 October 2025 — and on that all-time-high day the score read only 2.53. Price and this particular on-chain stretch measure peaked at different times.
Limits worth knowing before you lean on this
The denominator only grows. An expanding-window standard deviation gets larger as Bitcoin's history lengthens, so the same percentage premium over realized cap yields a smaller Z-score every year. Part of the decay in the table above is a real change in market behaviour and part is arithmetic. There is no clean way to separate the two, and anyone claiming a precise split is guessing.
Realized cap is not really "cost basis". Lost coins keep the price they last moved at forever. Exchange-internal transfers and self-transfers reprice coins that never changed hands economically. Realized cap is a useful proxy, not an accounting fact.
The sample is five cycles. Every threshold on this page is fitted to at most five observations of a market that has changed structure completely — from a hobbyist exchange in 2011 to spot ETFs and a listed-company treasury bid today. Treat the bands as a way of describing where a number sits, not as a rule that has been tested.
Questions people ask about the MVRV Z-Score
What is the Bitcoin MVRV Z-Score?
The MVRV Z-Score measures how far Bitcoin's market capitalisation has moved away from its realized capitalisation — a rough aggregate cost basis for all coins — and expresses that gap in standard deviations of market cap. Readings below zero mean the market in aggregate holds an unrealized loss; historically high readings have clustered around cycle peaks. It is a description of where valuation sits, not a signal to act.
How is the MVRV Z-Score calculated?
CryptoTIDE derives realized cap as market cap divided by the MVRV ratio, then computes (market cap − realized cap) ÷ σ(market cap), where σ is the population standard deviation of every market-cap observation from the start of the series to that day. Both inputs come from free CoinMetrics Community fields, so the whole calculation is reproducible from public data.
What MVRV Z-Score levels have appeared at Bitcoin cycle tops and bottoms?
Cycle peaks printed 11.75 in 2010, 10.66 in April 2013, 10.09 in December 2017, 7.15 in February 2021 and 3.35 in December 2024 — a lower extreme every cycle. Cycle troughs printed −0.69, −0.60, −0.49 and −0.36, with the open cycle so far bottoming at +0.24. Both extremes are compressing, so quoting a fixed "top level" from an older cycle is misleading.
Why does CryptoTIDE's MVRV Z-Score differ from other dashboards?
Because it is computed in house from public CoinMetrics market-cap and MVRV-ratio fields rather than from a proprietary realized-cap pipeline, and because the standard deviation here is a population statistic over an expanding window. Vendors that use a different realized-cap methodology or a different σ convention will print a different level for the same day. The approximation is disclosed rather than hidden.
Related Bitcoin cycle indicators
No single gauge carries a cycle read on its own. CryptoTIDE groups correlated metrics into families so that four views of the same realized-cap data cannot vote four times — the reasoning is on the methodology page, and the combined live read is on the Bitcoin cycle dashboard.