• Bitcoin’s Cross-Asset Correlation Problem: Why Correlation Analysis Is Harder Than It Looks
  • The Rolling Correlation Problem
  • The Gold Correlation Illusion
  • The Causation Question
  • The Practical Investment Implication
  • Key Takeaways

§Bitcoin’s Cross-Asset Correlation Problem: Why Correlation Analysis Is Harder Than It Looks

“Bitcoin is correlated with stocks” or “Bitcoin is correlated with gold” — these statements appear constantly in financial media. The reality is that correlation depends critically on the time period, the specific assets being measured, and the mathematical method used. Most casual correlation claims don’t survive scrutiny.

§The Rolling Correlation Problem

Standard correlation measurements use a fixed time window — say, 90-day returns. But correlation isn’t stable. Bitcoin’s correlation with the S&P 500 in 2020 was near zero. In 2022 it spiked to 0.6+ during the Fed tightening cycle. In 2023 it returned closer to zero. In 2024-2026 it fluctuates between 0.1 and 0.4 depending on the specific time window.

If you measure the 2022 period, you conclude “Bitcoin is a risk asset that falls with stocks.” If you measure 2020 or 2023, you conclude “Bitcoin decouples from equities.” Both conclusions are correct for their respective periods and wrong as general claims.

§The Gold Correlation Illusion

Bitcoin’s correlation with gold has been similarly unstable. Post-ETF approval, institutional money treating Bitcoin as “digital gold” pushed the correlation to 0.5+. But during the 2022 bear market, Bitcoin fell faster and further than gold — correlation dropped.

The real question isn’t “what’s the correlation?” but “what’s driving the correlation and is it stable?” Gold’s correlation with Bitcoin partly reflects shared exposure to dollar strength and real interest rates. That relationship is more structural than Bitcoin’s relationship with tech stocks.

§The Causation Question

Correlation doesn’t imply causation, and in volatile assets like Bitcoin, even genuine correlation can be spurious. When both Bitcoin and tech stocks fall during a risk-off event, it’s tempting to conclude they’re driven by the same factors. But:

• Bitcoin’s volatility is 3-5x higher than stocks
• Bitcoin’s fundamental value drivers are completely different from software company valuations
• The correlation could reflect that “things that make people sell assets broadly” affect both, without any specific relationship between them

§The Practical Investment Implication

For portfolio construction, the unstable correlation creates both opportunity and challenge:

• Adding Bitcoin to a traditional 60/40 portfolio improves the efficient frontier during periods of low correlation
• During periods of high correlation, the diversification benefit disappears exactly when you might need it most
• A static 1-5% Bitcoin allocation that doesn’t adjust for correlation changes captures average benefits but misses timing opportunities

§Key Takeaways

• Bitcoin-stock correlation is unstable: 0.0 in 2020, 0.6+ in 2022, 0.2-0.4 in 2024-2026
• The gold correlation is similarly unstable — it’s driven by shared macro exposures, not fundamental linkage
• Most correlation claims are period-specific and don’t generalize
• Unstable correlation creates both diversification opportunity and timing risk
• Static allocation models capture average benefits; dynamic models risk mistiming

⚡ If this was useful, a zap is always welcome. tomford@rizful.com