Correlation Checker
Correlation between two assets
Correlation (r)
Overlapping days
BTC vol 30d
ETH vol 30d
Reading: they moved almost as one asset over this window
Correlation is computed on daily returns, not on price levels. Correlating levels is the common mistake: two assets that both drifted upward over a year score near 1.0 on levels whether or not they actually move together. The two series are aligned from the most recent day on whichever is shorter, because a coin listed part-way through the year has fewer closes. r describes one past window only — crypto correlations rise sharply in sell-offs, which is exactly when diversification is being relied upon. Not advice.
What this tool measures
The Pearson correlation of two assets’ daily returns over the days they both have history for. It ranges from +1 (they moved in lockstep) through 0 (no day-to-day relationship) to −1 (they moved in opposite directions).
Why returns and not prices
This is the part most correlation tools get wrong, and it is not a small difference. If you correlate price levels, any two assets that both trended in the same direction over the period score close to 1.0 whether or not they actually move together day to day — you are measuring a shared trend, not a relationship.
On our own data the gap is stark. Measured on daily returns, BTC and ETH come out around 0.88; measured on price levels the same two look like 0.99. BTC against a dollar stablecoin is roughly −0.21 on returns but −0.60 on levels, which would imply a hedging relationship that does not exist. Correlating returns is the only version that answers the question people are actually asking.
Alignment
The two series are lined up from the most recent day, over whichever is shorter. Assets listed part-way through the year have fewer closes, and zipping the two series from their start dates would compare different calendar days and produce a confidently wrong number. The tool prints how many overlapping days it used.
The limitation that matters most
Correlation is a single number describing one past window, and crypto correlations are not stable. They tend to rise sharply during sell-offs — assets that looked usefully independent in calm conditions frequently fall together when it counts, which is exactly when diversification is being relied upon. A low reading over the last year is not a promise about the next bad week.
Frequently asked questions
What counts as high? Above roughly 0.8 the two behaved almost as one asset over that window. Below about 0.2 in either direction there is little day-to-day relationship. The tool states a plain-language reading rather than implying that a second decimal place is meaningful.
Why is an asset compared with itself exactly 1? Because it is the same series. That is a useful check that the calculation is behaving.
Does correlation mean causation? No, and in crypto a high reading usually reflects both assets responding to the same market-wide conditions.
Is this advice? No. See our disclaimer.