Bitcoin Price History by Year, 2010 to 2026

Bitcoin's first reference price was $0.09 on July 18, 2010. Its highest daily close so far is $124,824.45 on October 6, 2025, and it closed October 27, 2026 at $80,800.54, down 7.7% for 2026 as of that close, measured from the last close of 2025. Of the fifteen complete calendar years from 2011 to 2025, eleven ended higher than they started and four ended lower: 2014, 2018, 2022 and 2025.

Those numbers are only as good as the definition behind them, and many price-history tables online never state one. This page uses a single public series for every row, says exactly which price each column means, and shows where the answer changes if you pick a different price instead. Nothing here is a forecast. It is a record of what one reference rate did, year by year, so that you can quote a figure and know what it is.

The Price Series and the Rule Behind Every Number

Every figure in the main table comes from Coin Metrics' bitcoin PriceUSD metric, pulled from the free Coin Metrics community API on October 28, 2026. Coin Metrics' metric catalog describes it as "the fixed closing price of the asset as of 00:00 UTC the following day", generated by Coin Metrics' own fixing and reference-rate service. In plain terms, the value labelled with a date is the reference price at midnight UTC at the end of that date. The pull has 5,946 daily values, from July 18, 2010 to October 27, 2026, with no missing days.

The columns use these definitions:

A close-based series is smoother than an exchange's intraday chart, and that is the point. Intraday extremes are set in single minutes on single venues, sometimes on a few coins of volume. A daily reference close is the number you can reproduce, and it is the kind of price funds value against: the iShares Bitcoin Trust, for example, names the CME CF Bitcoin Reference Rate, New York Variant, as its benchmark on its product page (checked October 28, 2026).

Bitcoin's Price by Year, 2010 to 2026

Calculated by CryptosEyes from Coin Metrics PriceUSD daily reference closes, pulled October 28, 2026. Change uses the prior year's last close; 2010 uses July 18.

The 2026 change of -7.7% is measured from the December 31, 2025 close. Measured from the first close of 2026, on January 1, it is -8.9%. The opening paragraph uses -7.7%, the December 31 basis, which is the convention the table follows. The two numbers differ because the price moved between the last close of 2025 and the first close of 2026, and you will see both quoted for the same year. Neither is wrong; they answer different questions. The same gap explains why some sites report 2025 as -7.3% (first close to last close) and others as -6.3%.

What the Table Shows When You Read Across the Years

A table of seventeen rows invites pattern-hunting, and some patterns in it are real while others are artefacts of the calendar. Four observations survive a careful look.

The gains shrank by orders of magnitude. The three best years by change are 2013 (+5,286%), 2011 (+1,471%) and 2017 (+1,337%). On percentage change, no year since 2017 has come within a factor of four of those. The best year after 2017 was 2020, at +304.9%, and the best since 2020 was 2023, at +155.5%. That is what you would expect from an asset whose market value grew from millions of dollars to more than a trillion: doubling a $1.6 trillion asset takes far more new money than multiplying a $144 million one about sixty times over, which is roughly what happened to market value in 2013 (the year-end market values are in a table further down).

The most recent down year was by far the shallowest. The four completed down years fell 56.0% (2014), 73.5% (2018), 64.4% (2022) and 6.3% (2025); since 2018, each has been shallower than the last. The 2025 decline is the outlier: a year that set twelve record closes and still finished lower, because the record came in October and the price gave back the gain by December. Calendar years cut cycles at arbitrary points, which is why the peak-to-trough bear market table tells you more about drawdowns than any annual row. The deepest decline from peak to trough since 2013 was 84.5%, and none of the annual figures in this table comes close to it.

The yearly range has narrowed. Divide each year's highest close by its lowest close. In 2011 the ratio was 98 to 1 and in 2013 it was 85 to 1. In 2017 it was 25 to 1. Since 2018 it has never been above 6 to 1, and 2025 recorded the narrowest range in the series at 1.63 to 1 ($76,351 to $124,824). 2026 to October 27 sits at 1.66 to 1. Narrower ranges are consistent with what the volatility survival guide describes: lower realised volatility than in early years.

Record years come in runs of one or two. Record closes appeared in runs: 2010 and 2011, 2013 alone, 2017 alone, 2020 and 2021, and 2024 and 2025. Between those runs are stretches of zero-record years: 2012, 2014 to 2016, 2018 and 2019, 2022 and 2023, and 2026 so far. The runs line up loosely with the halving schedule, which the halving cycles page measures with a stated rule instead of by eye. Loosely is the right word: 2020 and 2024 were both halving years and record years, while 2012 and 2016 were halving years with no records at all.

Milestones: When Each Round Number Was First Reached and Last Seen

Round numbers carry more attention than they deserve, but readers ask about them constantly, and the dates are easy to state precisely on a close basis.

Calculated by CryptosEyes from Coin Metrics PriceUSD, pulled October 28, 2026.

The $1,000 line is a good illustration of how long a round number can stay in play. Bitcoin first closed above it in November 2013 and last closed below it in March 2017, three years and four months later. The $10,000 line took from December 2017 to July 2020 to leave behind. The $100,000 line, first closed above in December 2024, had 218 daily closes at or above it through October 27, 2026, and the latest close is below it again.

One detail in the $100,000 row trips people up. The first trade above $100,000 came days before the first close above it. Bitstamp's daily candle for December 5, 2024 (UTC) shows a high of $103,647, yet that day closed at $97,057 on Bitstamp and $96,949.66 on the reference rate. On the daily reference close used here, the first close above $100,000 was December 8. Both dates are true about different prices, which is the subject of the next section.

Price Times Supply: Market Value at Each Year-End

Price alone hides how much the asset grew, because the number of coins kept rising on the issuance schedule. Multiplying each year-end close by the supply Coin Metrics reports for the same day (its SplyCur metric, the coins issued so far) gives the market value usually labelled market capitalisation.

Calculated by CryptosEyes from Coin Metrics PriceUSD and SplyCur, pulled October 28, 2026. Selected years; the pattern in the omitted years is the same.

Two things follow from the supply column. First, a price return slightly understates the growth in market value, because supply grew too: from the end of 2016 to the end of 2017 the price rose 1,337% while market value rose about 1,400%. The effect fades as issuance shrinks. Supply grew 0.8% in 2025, against 7.0% in 2016. Second, "market value" counts every issued coin, including any whose keys have been lost, because the supply series counts issuance and cannot see who still controls a coin. It is a price multiplied by a count, not money anyone could take out. How much of that supply sits on corporate balance sheets, with each company's filing date, is on our list of public companies holding bitcoin.

Close, Intraday High, or Exchange Price: Why Sites Disagree

Search for bitcoin's 2025 high and you will find figures ranging from about $124,800 to about $126,300 on different pages. None of them is a typo. They measure different things:

To show how large the gap can be, the table below puts the Coin Metrics daily close next to the intraday high or low printed on one long-running US exchange, Bitstamp, taken from its public daily OHLC data for the BTC/USD pair (pulled October 28, 2026).

Sources: Coin Metrics PriceUSD and Bitstamp public OHLC API, daily candles in UTC, both pulled October 28, 2026. Gap is the Bitstamp extreme against the Coin Metrics close.

The March 2020 row is the extreme case. On the worst day of the pandemic crash, bitcoin traded as low as $3,850 on Bitstamp, but the reference close that day was $5,627.69 and the lowest close of the year was $4,959.31 the day before. Anyone describing "the March 2020 low" needs to say whether they mean the trade or the close, because the answer differs by more than a fifth.

The "different day" rows matter for anyone matching dates. In 2021 the highest close came on November 8, but the highest trade came two days later on November 10, when the close was lower ($64,756.08). A chart labelled "2021 all-time high, November 10" and a table that says "November 8" are both right about their own data.

Which should you use? For quoting history, comparing years or computing returns, use closes from one named series and say which. For questions about what was actually possible to trade, such as whether a stop order could have filled, only the venue's own trade data answers it, and the answer differs from venue to venue. Our page on why crypto prices differ across exchanges explains how those gaps arise and why they widen in fast markets.

How the Record Close Moved

The all-time high is the number people check first, and its history is short enough to list in full. These are the record closes that stood at the end of each run of record-setting, with the date the series first closed above them again.

Calculated by CryptosEyes from Coin Metrics PriceUSD, pulled October 28, 2026. The 2020 and 2024 runs of new records peaked at $29,022.67 (2020-12-31) and $106,115.91 (2024-12-17) before continuing into the following year without a 50% decline, so they are part of the 2021 and 2025 rows.

Three of the five completed waits ran between 847 and 1,177 days, roughly two and a third to three and a quarter years. At the latest close of $80,800.54, the price stood 35.3% below the October 2025 record. That is a description of where the series is, not a statement about where it is going. The bear market page has the same episodes with peak-to-trough depth and time to the bottom, which is the other half of this story.

Edge Cases That Change the Numbers

Several choices, each defensible, move the figures in this table. They are listed here so you can tell whether another source disagrees with this page because of an error or because of a rule.

Which day a year starts. A year's return from the last close of the previous December differs from a return measured from the first close of January. In fast years the difference is large: from December 31, 2012 to the first close of 2013, the price had already moved. The table uses the prior year's last close because it chains cleanly: multiply every year's change together and you get the full-period change.

Time zone. Coin Metrics uses midnight UTC. A US-based chart that closes the day at 4 p.m. New York time, or an exchange that closes at midnight in another zone, will produce different "closes" for the same date. The differences are usually small, and on volatile days they are not.

2010 is a partial year. Coin Metrics' reference price begins on July 18, 2010. Earlier trades existed on forums and informal venues, but there is no reference rate for them, and this page does not invent one.

Early-year prices rest on thin markets. The series begins in July 2010, when bitcoin changed hands in small amounts, and Coin Metrics' catalog entry for PriceUSD does not say which venues its earliest values draw on. Treat those early values as the best available record, not a precise one. Treat the 2010 to 2012 rows as approximate in a way the later rows are not.

Price is not value held. The table says nothing about what a typical holder earned. Someone who bought at the 2017 high and sold at the 2018 low lost 83.8%, while the 2017 row shows +1,337%. Returns depend on entry dates, which is why dollar-cost averaging guides show a distribution of outcomes rather than one number.

Reproducing This Table

Everything above can be rebuilt in a few minutes, which is the main reason to prefer it over an unsourced table.

If your numbers differ by a cent, check rounding. If they differ by more, you are probably using a different series, a different time zone or an intraday high. The script used for this page applies exactly the steps above, and the bear market history and halving cycle pages use the same Coin Metrics series, so their historical figures agree with this page's. For how this data fits with on-chain measures of cost basis, the realized price guide is the natural companion, and the on-chain and cycles hub lists every related guide.

Frequently Asked Questions

What was bitcoin's price in 2010?

The first reference price, on July 18, 2010, was $0.09. The lowest close that year was $0.05 on July 25, the highest was $0.40 on November 6, and the year ended at $0.30. Trades before July 2010 happened on informal venues with no reference rate.

What is bitcoin's all-time high?

On daily reference closes, $124,824.45 on October 6, 2025. Intraday trades went higher that day: Bitstamp's daily candle shows a high of $126,272. Sources that quote about $126,000 are using an intraday print; sources that quote about $124,800 are using a close.

Which year was bitcoin's best and worst?

By calendar change on this series, the best year was 2013 at +5,286% and the worst complete year was 2018 at -73.5%. Since 2018 the best year was 2020 at +304.9%, and the worst was 2022 at -64.4%.

How many years has bitcoin gone down?

Four of the fifteen complete calendar years from 2011 to 2025 ended lower: 2014, 2018, 2022 and 2025. 2026 was down 7.7% from the end of 2025 as of October 27, with the year not yet finished.

Why does my chart show a different high or low for the same year?

Many exchange charts plot intraday highs and lows from that one exchange, while this table uses one daily close per day from a reference rate. The gap is usually 1% to 3%, but on crash days it can be far larger: the March 2020 intraday low on Bitstamp was 22.4% below the lowest close of that year.

Where can I see how long bitcoin's declines lasted?

The bear market history table lists every decline of 50% or more on the same daily series, with the dates of each peak and trough, the depth, and how many days it took to close above the old high again.

CryptosEyes publishes general educational research, not financial or investment advice. The figures on this page describe what one price series did in the past; none of them is a forecast, and bitcoin can lose most of its value in a single year, as the 2014, 2018 and 2022 rows show.