If you are long three coins that all move as one, you do not have three trades — you have one, three times the size. This shows the hidden overlap between the majors over your chosen window.
Each cell is the Pearson correlation of two coins' bar-to-bar returns over your chosen window, on a scale from −1 to +1. The diagonal is always 1.00 — every coin is perfectly correlated with itself. Read across a row or down a column to see how one coin relates to all the others.
Deep red cells (near +1) move tightly together: long both and you are doubling a single bet. Cells near 0 are roughly independent, so they spread risk. Green cells (negative) move in opposite directions and act as a partial hedge. Shorten the window to see recent regime shifts; lengthen it for a steadier, structural picture of how the majors travel as a pack.
A correlation matrix is a grid that shows how closely the price movements of several assets track each other. Each cell holds a coefficient between −1 and +1 for the pair in that row and column: +1 means they move perfectly together, 0 means no linear relationship, and −1 means they move in exactly opposite directions. This tool computes Pearson correlation on the bar-to-bar returns of the major coins over your chosen window.
Stacking highly correlated positions multiplies the same risk. If you are long three coins that all move as one, you do not have three independent trades — you have one bet, three times the size, and a single adverse move hits all of them at once. Knowing the correlations lets you size positions sensibly, avoid hidden double-exposure, and spread risk across coins that do not move in lockstep.
A negative correlation means two coins tend to move in opposite directions: when one rises, the other tends to fall. A pair near −1 acts as a natural hedge, partly offsetting each other's swings. In crypto, genuine negative correlation between major coins is rare and usually mild — most coins are strongly positive versus Bitcoin — so a negative reading often reflects a short-term, window-specific divergence rather than a lasting relationship.