Bitcoin log-level correlation with S&P hits decade low

Bitwise’s Market Compass, dated Sept. 7, reported that Bitcoin’s 260-day log-level correlation with the S&P 500 reached its lowest reading since 2015. An independent calculation using FRED data produced a similar log-level result of about −0.616. Measuring percentage returns over the same 260 common trading observations produced a positive correlation of roughly +0.396.

The two statistics measure different relationships. The log-level correlation tracks how the two price paths moved over months. The percentage-return correlation tracks whether daily gains and losses tended to occur together. The dataset used in the independent checks ended on Sept. 4, 2026, and uses daily public prices recorded at different local times.

FRED records a Coinbase Bitcoin observation at 5 p.m. Pacific time and the S&P 500 at the U.S. market close, typically 4 p.m. Eastern time. That means the two series share a calendar date but not identical timestamps. Shifting Bitcoin’s dates by one calendar day changes the 260-return correlation to +0.14 when moved earlier and +0.03 when moved later. Over a 30-day window, the same shifts produce correlations of −0.30 and −0.37. Those changes show that return correlations are sensitive to timing.

Return correlations vary by window. Using overlapping windows that end on Sept. 4, 2026, the return correlations were +0.14 over 30 observations, +0.32 over 60, +0.34 over 126 and +0.13 over 500 observations. A negative log-level correlation can coexist with a positive return correlation for many window choices.

Daily return behavior across the 260-return sample showed repeated overlap in losses. The S&P 500 declined on 115 common trading dates; Bitcoin declined on 76 of those days and averaged a 1.03% loss across all 115 equity-down observations. When the S&P fell by more than 1% (25 observations), Bitcoin fell on 22 of those days and averaged a 2.59% loss. There were three instances when equities fell more than 2%; Bitcoin fell on all three and averaged a 4.23% decline. The sample for large drops is small.

A simple portfolio exercise used the common-date return windows with weights reset at every common observation. Replacing 5% of a 100% S&P 500 allocation with Bitcoin increased measured annualized volatility from 12.68% to 13.01% and raised the largest peak-to-trough drawdown from 9.10% to 9.98% over the 260-return window. Using a 500-return window, the same 5% Bitcoin allocation reduced volatility from 16.11% to 15.73% and reduced maximum drawdown from 18.90% to 18.66%. These calculations used prices recorded at different times, excluded fees, taxes and dividends, and the S&P series is a price index rather than a total-return series.

Bitwise’s report stopped short of calling the observed decoupling a hedge. The report said a repeat of Bitcoin’s roughly 9,800% rise from mid-2015 to late 2017 is unlikely and warned that periods of decoupling usually do not last. The report also noted that portfolio outcomes depend on allocation size, asset volatility, the measurement window and whether prices are synchronized.