🔗 Bitcoin vs Stocks, Gold & Dollar

How Bitcoin and Ether moved with the Nasdaq, S&P 500, KOSPI, gold, the dollar, US 10Y yield, oil and USD/KRW: correlation bars, matrix, rolling chart. Yahoo Finance data, not advice.

Loading Data: Yahoo Finance (via our server, may be delayed)

Base

Window

Bitcoin's correlation with each asset

BitcoinBTC-USD
–
EthereumETH-USD
–
Nasdaq Composite^IXIC
–
S&P 500^GSPC
–
KOSPI^KS11
–
Gold futuresGC=F
–
US Dollar IndexDX-Y.NYB
–
US 10-year yield^TNX
–
WTI crude futuresCL=F
–
USD/KRWKRW=X
–

−10+1

In short

–

Correlation matrix

−1+1

Rolling correlation

Compare with

Performance (window start = 100)

Data status

–

A correlation coefficient measures, from past data, whether two assets' daily moves went the same way (near +1) or opposite ways (near −1). It shifts a lot over time and shows neither causation nor what prices will do next. Crypto trades at weekends, so only dates on which both assets have a close are used, and returns are taken between consecutive common dates.

For information only, not investment advice. Prices come from Yahoo Finance via our server and may be delayed 15–20 minutes or more depending on the exchange, or interrupted. Charts: TradingView Lightweight Charts™

What this tool does

Measures, from real daily prices, how closely Bitcoin (or Ethereum) has moved with the Nasdaq, the S&P 500, KOSPI, gold, the US Dollar Index, the US 10-year Treasury yield, WTI crude and USD/KRW. It shows 30-, 90-, 180- and 365-day correlations as bars and a 10×10 matrix, a 30- and 60-change rolling correlation with a chosen asset over the past one or two years, and each asset's performance rebased to 100 over the same window. Prices are five years of daily bars that this site's server fetches from Yahoo Finance.

How it is calculated

Each daily close is dated in its exchange's time zone (crypto by UTC date; gold and crude futures, whose session starts the evening before, are shifted six hours onto their trading date). The last bar is skipped while it is still forming (today's crypto bar, an intraday bar). For each pair only the dates on which both assets have a close are kept, and changes are taken between consecutive common dates: a percentage return for prices (close ÷ previous common close − 1) and a difference in percentage points for the US 10-year yield, which is a rate rather than a price. The correlation is the Pearson coefficient of the two change series, Σ(x − x̄)(y − ȳ) ÷ √(Σ(x − x̄)² × Σ(y − ȳ)²), and is not shown when there are fewer than 10 changes. An N-day window ends on the latest last closed date among all assets and is anchored on the last common date on or before N days earlier; only changes after the anchor are used. The rolling correlation uses the latest 30 or 60 changes up to each date, and the performance chart sets each asset's anchor close to 100.

Choose Bitcoin or Ethereum as the base.

Choose a window (30, 90, 180 or 365 days); the bars, matrix and performance chart are recalculated for it.

In the rolling correlation card pick the asset to compare and how far back to look (past one or two years).

Click legend items on the performance chart to hide or show lines. The page address keeps the base, window and compared asset, so a shared link opens the same view.

How to read each section

Base and window

Pick the base (Bitcoin or Ethereum) and the window (30, 90, 180 or 365 calendar days) at the top. The window ends on the latest last closed date among all assets. The status pill shows how many assets loaded and that last date.

Correlation with each asset (bars)

The centre is 0, the right end +1 and the left end −1. A longer bar to the right means the asset more often moved the same way as the base on the same day; to the left, the opposite way. The small number is the count of changes used; with fewer than 10 the value is left blank as too few.

Correlation matrix

Every pair of the ten assets over the same window. Yellow cells are positive, blue cells negative, and darker means a larger absolute value. Each pair uses only dates on which both have a close, so sample sizes differ by cell. On a phone, scroll the table sideways.

Rolling correlation

The base's correlation with the chosen asset, recalculated for each date from the latest 30 (faint line) and 60 (solid line) changes between common dates. The vertical axis is fixed from −1 to +1 and the dashed line is 0. The more the lines swing, the more the relationship differed from one period to the next.

Performance

Each line is set to 100 at the window's anchor close; 120 means 20% above it. Stocks, indices and futures have no points at weekends or holidays, so their lines skip those days. The US 10-year yield is left out because it is not a price.

Data status

For each asset: the last date used, its close and the change over the window (basis points for the yield). Assets that failed to load are marked as not loaded and are retried automatically.

Things to know

Crypto trades at weekends but stocks, indices and futures do not, so between crypto and stocks Friday's close to Monday's close counts as one change. A big weekend move in crypto lands in that Monday change.

KOSPI closes in the Korean afternoon, US indices in the Korean early morning and the crypto daily bar at UTC midnight. The same date covers different hours, so assets with distant closing times tend to look less correlated than they are.

USD/KRW (KRW=X) is the price of one dollar in won, so a rise means a weaker won and a stronger dollar. The US Dollar Index (DX-Y.NYB) measures the dollar against six currencies including the euro and yen.

Shorter windows have fewer samples and jumpier numbers. A stock pair in a 30-day window has only about 20 changes, so read it next to the 90- and 180-day values.

Correlation says how often the direction matched, not how big the moves were. A highly correlated asset can still have moved far more (see the performance chart).

Prices are reloaded every 30 minutes. Assets that fail to load are marked in the bars, matrix and data status table, and the rest are still calculated.

Frequently asked questions

How are weekends handled?

For every pair only the dates on which both assets have a close are kept. For Bitcoin and the Nasdaq that leaves weekdays, and since changes are taken between consecutive kept dates, Friday to Monday is one change. A pair that both trade at weekends, such as Bitcoin and Ethereum, keeps the weekends. That is why sample sizes differ by cell in the matrix; hover a cell to see it.

Does a high correlation mean Bitcoin follows the stock market?

No. It measures how often the daily direction matched over a past period, not which one drives the other or whether it will continue. The rolling chart shows the same pair swinging widely over time. Use it to see how relationships between markets have changed, not as a basis for investment decisions.

Why is the US 10-year yield treated differently?

A yield is a rate, not a price, so calling a move from 4% to 4.4% a 10% rise is misleading. Its correlation is measured on daily changes in percentage points, it is left out of the rebased performance chart, and its change in the data status table is shown in basis points.

Why do the numbers differ slightly from other sites?

Correlations depend on the window's start and end, daily versus weekly returns, how dates are aligned (here: only dates on which both assets have a close) and when the crypto daily bar closes. This tool uses Yahoo Finance daily closes with the method above and leaves out today's unfinished bar.

For information only, not investment advice. Correlations change over time and imply neither causation nor prediction. Prices come from Yahoo Finance via this site's server and may be delayed or interrupted.