About this tag

Correlation is a measurement, not a property. It describes how two price series moved relative to each other across a specific window - nothing more. A 0.4 reading between Bitcoin and an altcoin over twelve calm months says they tended to drift apart; it says nothing about why, and nothing about the next 48-hour stress event. The number is accurate for the regime it was sampled in. The mistake is treating it as fixed.

That distinction matters most where it fails. In calm conditions, crypto assets genuinely move on separate theses - a maximalist holds Bitcoin, a yield farmer holds farm tokens, and demand differentials produce low correlation. In a crash, conviction stops setting price and liquidity takes over. Forced liquidations and risk-off selling hit whatever can be sold, indiscriminately, and correlations converge toward 1. The relationship did not break. It responded accurately to a new set of conditions: everything sold at once, by the same actors, for the same reason.

This tag collects observations on how correlation behaves across regimes. Why Bitcoin moves before altcoins because capital routes through it first. How dominance shifts and BTC-alt decorrelation mark the structural start of rotation before any altcoin prints. Why portfolios that look diversified in a spreadsheet act like one asset in a drawdown. And how Bitcoin's correlation to equities is a symptom of shared participants, not a structural link - rising when funds de-risk together, fading when crypto trades its own narrative.

The framing is mechanical, not predictive. Correlation does not tell you where assets go - it tells you when their independence is real and when it is borrowed. Notes here document the conditions: where decorrelation is structural versus statistical, when shared liquidity collapses the spread, and what a correlation table from calm markets quietly hides until stress arrives.