Sign in
Article On-chain

Bitcoin realized price: what it shows and how to read it

Bitcoin realized price explained: the market's on-chain average cost basis, the formula, every dip below it since 2011 and a worked example at $54,100 (Oct 27, 2026).

Bitcoin realized price: what it shows and how to read it
Contents
  1. How realized price is calculated
  2. Where the metric came from
  3. What bitcoin realized price shows
  4. What happened when price fell below realized price
  5. A worked example on October 2026 data
  6. Cohort versions: STH, LTH, ETF cost basis
  7. Limitations
  8. Frequently asked questions

Bitcoin realized price is the average price at which every coin in circulation last moved on the blockchain. It is calculated as realized capitalization divided by supply: each coin is counted not at today's quote but at the price on the day it was last transferred. That is why realized price is read as the market's average cost basis: when bitcoin trades above it, the average holder is in profit; below it, in a loss. On October 27, 2026, bitcoin's realized price was about $54,100 against a price of $80,800, so the market sat 49% above its own average cost.

This guide covers where the metric came from, how it is calculated, what happened every time price fell below it, the versions built for specific holder groups, and what realized price cannot tell you.

How realized price is calculated

Ordinary market cap is supply times the last price. Realized cap works differently. According to Glassnode's documentation, it values each unspent transaction output (UTXO) at the price when it was created, that is, when the coin last changed addresses. If a coin moved in 2017 at $4,000 and has sat still since, it is worth $4,000 in the realized cap no matter what bitcoin costs today. Coins last moved in 2009, before bitcoin had a price, are valued by Glassnode at zero.

Realized price is realized cap divided by supply. CryptoQuant defines it the same way: Realized Cap divided by total coin supply, a supply-weighted average of what market participants paid for their coins.

A three-coin example:

Coin Last moved Price at that time
A 2019 $10,000
B 2024 $60,000
C 2025 $110,000

The realized cap of these three coins is $180,000 and the realized price is $60,000. If the market price is $80,000, the ratio of market cap to realized cap (MVRV) is 1.33: holders are, on average, 33% in profit. If coin A moved to a new address today, its "price tag" would reset to $80,000 and the realized price would rise to $83,333. That is how the metric lives: it rises when old, cheap coins are sold at high prices and falls when coins bought at high prices change hands during a decline.

Where the metric came from

Realized capitalization was created at Coin Metrics in 2018. In the post "Introducing Realized Capitalization" dated December 14, 2018, the team writes that the idea came after Pierre Rochard asked for data on a "historically-weighted UTXO market cap," and Coin Metrics engineer Antoine Le Calvez worked out the methodology and coined the name (the metric had previously been called "Effective Cap"). The same post gives the first reading: in September 2018, bitcoin's market cap was $115 billion and its realized cap $88 billion. Glassnode's documentation also credits Le Calvez and dates the metric to September 2018.

The motivation was stated plainly: market capitalization is "often an empty metric" for cryptocurrencies because a meaningful share of coins is lost or abandoned, and realized cap "reduces the contemporary impact of long-lost coins." A whole family of indicators has since been built on top of it: MVRV, NUPL, SOPR and the cost basis of individual holder cohorts. How MVRV, the Mayer multiple and Pi Cycle work together is covered in our article on bitcoin cycle indicators.

What bitcoin realized price shows

Three readings come up most often.

  1. The market's average cost basis. Above realized price, the average holder has an unrealized profit; below it, a loss. Glassnode calls it "one of, if not the most widely recognized Bitcoin on-chain models" and the on-chain acquisition price (cost basis) of the supply (The Week On-chain, week 29, 2022).
  2. Capital that actually entered the asset. Realized cap grows only when coins change hands at a higher price than last time. Its growth is "new money" coming in; its decline is capital leaving, recorded on-chain. Market cap cannot do that: it changes with every trade on an exchange.
  3. An on-chain support or resistance level. CryptoQuant's metric guide says realized price can be interpreted as on-chain support or resistance: below it the average holder is at a paper loss, and in past cycles such periods coincided with accumulation phases.

It is important to understand this is not a chart level that mechanically "holds" price. It is the point where the average participant flips from profit to loss, and behavior around it changes: some capitulate, others buy more.

What happened when price fell below realized price

We counted these periods from daily Coin Metrics data (metrics CapMVRVCur and PriceUSD; realized price = price divided by MVRV). Only runs of three or more consecutive days with a daily close below realized price are included.

Episode Period Consecutive days Lowest MVRV Example
2011 bear market Sep 21 – Dec 31, 2011 102 0.39 —
2015 bear market Jan 1 – Jul 10, 2015 191 0.56 Jan 14, 2015: $176 vs. realized $312
2015 bear market, second leg Jul 31 – Oct 22, 2015 84 0.75 —
2018⁠–⁠2019 bear market Nov 19, 2018 – Apr 1, 2019 134 0.69 Dec 15, 2018: $3,185 vs. $4,613
COVID crash Mar 14⁠–⁠18, 2020 5 0.90 Mar 16, 2020: $5,013 vs. $5,566
2022 bear market, first leg Jun 13 – Jul 17, 2022 35 0.84 Jun 18, 2022: $19,014 vs. $22,642
After the FTX collapse Nov 6, 2022 – Jan 12, 2023 68 0.75 Nov 9, 2022: $15,758 vs. $20,901

In 2022 there were two more short dips in August–November lasting 22 and 52 days; in total, price spent 177 days below realized price that cycle. Two takeaways from the table. First, every major bitcoin bottom occurred during a period below realized price, and the trough was 25⁠–⁠45% beneath it (except 2011 and the five-day March 2020 episode). Second, these periods last for months: in the same report dated July 18, 2022, Glassnode put the average time below realized price in previous bear markets at 197 days, excluding March 2020, with a range from 7 days in March 2020 to 301 days in 2015. On the date of that report the realized price was $22,092, bitcoin traded at $21,060 and had been below it for 35 days, MVRV was 0.953, and the average bear-market MVRV was 0.85.

The gap between our count and Glassnode's (191 versus 301 days in 2015) comes down to methodology: Glassnode computes realized cap from its own data and apparently does not break a streak on brief pops above the level. The overall picture is the same.

A worked example on October 2026 data

Daily Coin Metrics data for October 27, 2026: price $80,801, MVRV 1.494, realized price about $54,100, realized cap $1.087 trillion against a market cap of $1.624 trillion. An independent calculation by bitcoin-data.com gave $53,573 for October 21, 2026; the methodologies differ by fractions of a percent.

What the past year of this number tells us:

  • Realized price peaked on November 26, 2025 at $56,463, after the price peak of $124,824 on October 6, 2025. Realized cap at the time was about $1.09 trillion.
  • Realized price then declined: $56,109 on January 1, 2026, $54,156 on April 1, $53,056 on July 1. Realized cap fell to $1.064 trillion by June 30, 2026 as coins bought at high prices changed hands lower and capital left the asset.
  • The cycle low by daily close, $58,525 on June 30, 2026, stayed above the realized price of $53,070: MVRV that day was 1.103. Glassnode's report of October 28, 2026 puts it this way: this bear never closed below the realized price.
  • By October 27, 2026, realized cap was back to $1.087 trillion: capital had been flowing into bitcoin again since July.

This is what separates the current drawdown from 2015, 2018 and 2022: back then the bottom was set 25⁠–⁠45% below the market's average cost, and this time price never touched it. Whether that means the low is already in or that the "real" capitulation has not happened yet, the metric does not say. It only shows that the average holder stayed in profit throughout 2026. Live MVRV and the short-term holder cost basis are in our Market Pulse block, and how this fits the four-year pattern is in our article on the bitcoin four-year cycle.

Cohort versions: STH, LTH, ETF cost basis

The aggregate realized price averages coins from 2011 with coins from last week, so analysts also compute it for groups.

Version Which coins Reading, October 2026 Source
Realized price (all coins) Entire supply ≈$54,100 on Oct 27 Coin Metrics
Short-term holder (STH) cost basis Coins moved less than 155 days ago $74,865 on Oct 21; $75.8K per Glassnode bitcoin-data.com, Glassnode
Long-term holder (LTH) cost basis Coins older than 155 days $48,720 on Oct 21 bitcoin-data.com
True Market Mean Cost basis of active investors, dormant coins excluded $77.9K Glassnode
US spot ETF cost basis Average price the ETFs paid for their coins $81.5K Glassnode

Glassnode chose the 155-day threshold because the longer a coin sits, the less likely it is to be spent; Bitcoin Magazine's 2021 explainer gives the same UTXO spend-probability rationale. The STH cost basis is the cost of "fresh" buyers, the most nervous part of the market: it acts as support in bull phases and resistance in bear phases. The bitcoin-data.com figures come from bitcoin-data.com; the Glassnode levels are from The Week On-chain, week 43, 2026. In the same report Glassnode notes that long-term holders hold a block of about 4 million BTC bought between $85K and $118K, and more than half of those coins were bought above the current price; on a recovery toward $85K some of them may sell to break even. How ETF flows reshape this picture is covered in our article on spot bitcoin ETF flows.

Limitations

  • Lost and dormant coins look the same. Coin Metrics admitted this from day one: realized cap "has a hard time differentiating between truly lost/abandoned coins and coins that are merely in yearslong deep cold storage." If Satoshi's coins ever move, realized price jumps without a single purchase.
  • Internal transfers count as trades. An exchange shuffles coins between its own wallets, an ETF custodian rotates addresses, a user moves savings to a new cold wallet: to the blockchain that is a "move at the current price," even though the owner did not change. Providers filter some of these transfers, but differently, which is why Glassnode, CryptoQuant and Coin Metrics disagree slightly.
  • Off-chain markets are invisible. Coins sitting with a custodian and changing owners inside its database (ETFs, corporate treasuries, OTC deals) do not move on-chain and never update their "price tag."
  • Lag. Free sources deliver data with a delay of one day to a week; in a sharp move, "yesterday's" realized price can differ from the real one.
  • It is not a trading signal. Price can sit below realized price for months (2015: more than half a year) and keep falling: in 2015 and 2018 it went 30⁠–⁠45% below the level that was "supposed" to hold.

Frequently asked questions

How is realized price different from MVRV? MVRV is market cap divided by realized cap, or equivalently price divided by realized price. MVRV of 1.49 means price is 49% above realized price. It is the same information in a different form.

Where can I see bitcoin realized price for free? Daily values are available in Glassnode Studio, at LookIntoBitcoin, on bitcoin-data.com (seven-day delay without a subscription) and in Coin Metrics community data via MVRV. We show MVRV and the short-term holder cost basis in the Market Pulse block and on the bitcoin page.

Why does realized price fall while bitcoin rises? Because it depends not on the quote but on which coins move. If the coins sold into a rally were mostly bought even higher (for example at the 2025 peak), their price tags reset lower and the average declines. That is what happened from November 2025 to July 2026.

Does bitcoin have to fall to its realized price? No. In past bear markets it happened every time, but the sample is four cycles, and market structure has changed: a meaningful share of coins sits with ETFs and companies and never moves on-chain. In 2026, price came no closer than 10% to realized price.

This is not investment advice.

Morning brief and news on TelegramEvery morning at 9:00 AM ET, before the US open: price, 24h scenario, squeeze risk. Key news around the clock.
Coins
Category
Tags
#realized price#realized cap#MVRV#holder cost basis#Coin Metrics
Ryan Mitchell
On-chain and mining

Breaks down blockchain and mining data: hashrate, difficulty, holder and whale behavior.

Comments

Read next