Bitcoin Ethereum correlation by year, plus 8 more coins
BTC and ETH daily moves had a 0.82 correlation on Binance since 2019, from 0.77 in 2021 to 0.91 in 2026. See 8 more coins and what this means.
Bitcoin and Ethereum move closely together. On Binance perpetual futures, their daily price moves had a correlation of 0.82 over 2,518 days from late 2019 to October 2026. The yearly score changed: it was lowest in 2021 at 0.77 and highest in 2026 so far at 0.91, the closest link in any year we measured.
Most other big coins also moved more closely with Bitcoin in 2025 and 2026 than in earlier years. These numbers describe the past. They do not tell you what comes next; we explain why in why past crypto moves are not a forecast.
For how far each coin moves when Bitcoin moves 1%, see Do altcoins follow Bitcoin?.
What does correlation actually tell you?
A daily return is the percentage change from one day's closing price to the next. We compare those daily changes, not the prices.
Correlation is a score from −1 to 1. We use the most common kind, called Pearson correlation, written r.
- Near 1: on days one coin had a big move up, the other usually did too, and the same for down days.
- Near 0: the two coins' daily moves showed little straight-line link.
- Below 0: when one went up, the other tended to go down.
Two things correlation does not tell you:
- It is not the share of days both coins went up. A score of 0.8 does not mean "80% of days".
- It does not mean equal size. Two coins can move in the same direction while one moves twice as much.
How did Bitcoin Ethereum correlation change by year?
We measured BTC–ETH on two exchanges. Binance BTCUSDT and ETHUSDT are perpetual futures (perps), contracts that follow the coin's price and never expire. Coinbase BTC-USD and ETH-USD are spot markets, where the coins themselves trade. Each pair was compared on its own exchange only. Dates are in UTC, and 2026 runs through October 19.
| Year | Binance r | Days | Coinbase r | Days |
|---|---|---|---|---|
| 2016 (from May 18) | — | — | 0.26 | 224 |
| 2017 | — | — | 0.37 | 365 |
| 2018 | — | — | 0.82 | 365 |
| 2019 (Binance from Nov 27) | 0.87 | 34 | 0.82 | 365 |
| 2020 | 0.85 | 366 | 0.85 | 366 |
| 2021 | 0.77 | 365 | 0.77 | 365 |
| 2022 | 0.89 | 365 | 0.89 | 365 |
| 2023 | 0.83 | 365 | 0.83 | 365 |
| 2024 | 0.80 | 366 | 0.80 | 366 |
| 2025 | 0.82 | 365 | 0.82 | 365 |
| 2026 (to Oct 19) | 0.91 | 292 | 0.91 | 292 |
A dash means Binance ETH futures history in our test starts later. The Binance 2019 score uses only 34 days, so treat it with care.
What the table shows:
- The score moved up and down. It fell in 2021, rose in 2022, eased through 2024, then rose again.
- The two exchanges agree. From 2020 to 2026, the Binance and Coinbase scores match to two decimals.
- The early years were very different. On Coinbase, BTC and ETH were only loosely linked in 2016 and 2017.
The 2026 score covers part of a year, so the full year could differ.
Why can a full-history score mislead you?
Over the whole history, BTC–ETH was 0.82 on Binance (2,518 days) but only 0.70 on Coinbase (3,803 days). That is not a gap between exchanges. Coinbase's history starts in 2016 and includes the loosely linked years of 2016 and 2017. In the years both cover, their scores match.
A full-history score is also not the average of the yearly scores. It is computed from all days at once, so days with very large moves count for more. Dogecoin (DOGE) shows how far apart the two can be: its full-history score with BTC was only 0.38, yet it was 0.61 or higher in every year from 2022 to 2026. Its 2021 score was 0.29. We did not measure exactly which days pulled the full-history number down.
So when someone quotes one correlation number, ask which dates it covers.
For why the same coin can trade at different prices on different exchanges, see why Bitcoin prices differ between exchanges.
Do other coins move with Bitcoin the same way?
Not the same way, but most moved more closely with BTC lately. This table compares each coin's daily moves with Bitcoin's on Binance USDT perpetual futures. 2021 to 2025 are full years; 2026 has 292 days. The 2020 column covers only part of the year for every coin except ETH, because their futures started trading later (from 99 days for AVAX to 360 for XRP).
| Coin vs BTC | 2020 | 2021 | 2022 | 2023 | 2024 | 2025 | 2026 (to Oct 19) | Full history |
|---|---|---|---|---|---|---|---|---|
| ETH | 0.85 | 0.77 | 0.89 | 0.83 | 0.80 | 0.82 | 0.91 | 0.82 |
| SOL | 0.24 | 0.40 | 0.77 | 0.61 | 0.75 | 0.80 | 0.88 | 0.59 |
| XRP | 0.58 | 0.55 | 0.74 | 0.40 | 0.44 | 0.78 | 0.87 | 0.58 |
| LINK | 0.62 | 0.68 | 0.76 | 0.59 | 0.57 | 0.80 | 0.82 | 0.68 |
| BNB | 0.78 | 0.60 | 0.82 | 0.63 | 0.63 | 0.66 | 0.82 | 0.67 |
| DOGE | 0.45 | 0.29 | 0.65 | 0.61 | 0.78 | 0.80 | 0.79 | 0.38 |
| ADA | 0.72 | 0.57 | 0.78 | 0.65 | 0.66 | 0.75 | 0.79 | 0.67 |
| AVAX | 0.38 | 0.46 | 0.81 | 0.59 | 0.69 | 0.77 | 0.70 | 0.60 |
| TRX | 0.69 | 0.65 | 0.61 | 0.59 | 0.13 | 0.47 | 0.40 | 0.52 |
What stands out:
- SOL went from loose to tight. 0.40 in 2021, 0.88 in 2026.
- XRP dipped, then jumped. About 0.40 in 2023 and 0.44 in 2024, then 0.87 in 2026.
- 2022 was high for almost everyone. Seven of the nine coins were above 0.70 that year.
- TRX went the other way. Its link with BTC fell to 0.13 in 2024 and was 0.40 in 2026.
- For every coin except TRX, 2026 was above the full-history number.
We only measured coins that still trade today. Coins that collapsed or were delisted are not in this table, so it may make the group look more stable than the whole market was.
What does this mean for spreading risk?
Spreading money across different assets so they do not all fall at once is called diversification. It only works if the assets really move differently. If five coins each have a link of about 0.8 with Bitcoin, they tend to have good and bad days together. They spread risk less than the number of coins suggests.
Some practical points:
- Check the most recent year, not only the full history. For SOL, XRP and DOGE, the full-history score made the link look much weaker than it was in 2025 and 2026.
- Expect the link to change. Every coin's score changed from year to year, some by a lot.
- Correlation is not loss size. It does not tell you how far each coin can fall. For that, see crypto drawdowns and recovery times. A drawdown is a fall from a previous high.
What we did not test: whether correlations get higher on big down days, or how a group of coins held together would have done. So we cannot say how much risk a mix of these coins would have removed.
How we measured this
We used public daily price bars from Binance perpetual futures and Coinbase spot, up to an October 20, 2026 UTC cutoff. For each coin, we took the percentage change from one daily close to the next, only when both days had a price. We then paired each coin's daily changes with Bitcoin's on the same exchange and the same dates, and computed Pearson correlation for each year and for the whole period. Each change belongs to the year of the day it ends on, so the first change of a year uses the last close of the year before. We did not fill in missing prices, did not test any trades, and these scores do not show that one coin's move leads another's.
Try this in Comparand
You can get a correlation matrix for Hyperliquid coins with the hl_stats tool. Ask your AI agent:
"Using hl_stats, show the correlation matrix and volatility for BTC, ETH, SOL, XRP and DOGE on Hyperliquid for [a day you pick]."
What comes back:
- A correlation matrix: the link between every pair of coins you listed, up to 20 coins.
- Realized volatility: how much each coin's price actually moved.
- A volatility regime: a label for the level of price movement.
It costs $0.01 per call on pay-as-you-go. To see how links change, run it for a few different days and compare. hl_stats uses Hyperliquid prices, so its numbers will not match the Binance and Coinbase results above exactly.
You can try 3 coins free, and the docs explain each tool. These are past cases, not a forecast.
The short version
- BTC and ETH daily moves had a 0.82 correlation on Binance from late 2019 to October 2026.
- By year it ranged from 0.77 (2021) to 0.91 (2026 so far), and Coinbase showed the same yearly scores.
- Most other big coins moved more closely with BTC in 2025 and 2026; TRX was the clear exception.
- A full-history score can hide this. DOGE was 0.38 overall but 0.61 or higher in every year from 2022.
- Correlation compares same-day moves. It does not predict, and it does not measure how big losses can be.
Questions people ask
How correlated are Bitcoin and Ethereum?
Closely. On Binance perpetual futures, BTC and ETH daily moves had a correlation of 0.82 over 2,518 days from late 2019 to October 2026. By year it ranged from 0.77 in 2021 to 0.91 in 2026 (through October 19).
Does Bitcoin Ethereum correlation stay the same each year?
No. On Binance it went 0.85, 0.77, 0.89, 0.83, 0.80, 0.82 and 0.91 from 2020 to 2026. Coinbase spot prices showed almost the same yearly scores.
Do altcoins move with Bitcoin?
Most did, and most moved more closely with Bitcoin in recent years. In 2026 through October 19, SOL, XRP, LINK and BNB were all at about 0.82 or higher with BTC on Binance. TRX was the exception at 0.40.
Does a high crypto correlation predict the next move?
No. Correlation compares moves on the same day. It does not show whether one coin moves first, and it does not tell you what either coin will do next.
Does owning several coins spread risk?
Less than it may seem, if those coins move closely with Bitcoin. Many coins we measured had a correlation near 0.8 or higher with BTC in 2025 and 2026. Correlation also does not tell you how big each coin's losses could be.