Bitcoin's Correlation with the Stock Market: How to Use It

Bitcoin's 30-day correlation with the S&P 500 as of August 28, 2026 sits at 0.02. Six months ago, in March, it was 0.74. At the end of 2025 it went negative, down to -0.299.

Three numbers in nine months for the same asset. Which is why the phrase "bitcoin and the stock market are connected" means nothing until someone names the lookback window and the date.

Let's break down how the metric is calculated, where it actually works, and where it turns into a pretty chart with no trading value. And most of all, what to do with it intraday, once your position is already open.

What correlation means in trading

Correlation is a measure of how closely two assets move together. It's calculated with the Pearson coefficient on returns, not on prices, and it fits on a scale from -1 to +1.

Three things break most traders' understanding of this metric.

First, the lookback window. A 30-day and a 200-day correlation on the same pair of assets can differ by a factor of several, and sometimes by sign. A short window catches the current regime and is noisy, a long one smooths and lags. Someone shows you a chart labeled "bitcoin and equities correlation"? The first question is always the same. What window?

Second, correlation says nothing about the size of the move. It's about directional synchrony, not amplitude. The S&P 500 drops 1%, bitcoin drops 4%, and the correlation is still 1.0. Amplitude is beta's job, a separate coefficient. Bitcoin's historical beta to the index runs around 2-3: for a one percent move in the index, bitcoin gives you two or three. Confusing the two gets expensive.

Third, correlation is not causation. Two assets can react in sync to a third factor without affecting each other at all. Bitcoin doesn't fall "because the Nasdaq fell." Both are pricing the same shift in rate expectations and risk appetite, the Nasdaq just reflects it faster.

How to calculate it yourself

Five minutes in any spreadsheet. Do it by hand at least once, and other people's pretty pictures stop being impressive.

Take bitcoin's daily closes and the S&P 500's daily closes for the last thirty trading sessions. Convert them into returns with the formula (today's price / yesterday's price - 1). The index has no weekend data, so days without quotes get dropped from both series. Then run the standard CORREL function on the two columns.

Two mistakes come up most often. People calculate on prices instead of returns, and then two rising assets give you 0.95 simply because both went up. And they don't align the series in time: bitcoin closes its day at midnight UTC, the US market at 21:00 UTC, and three hours of difference on a volatile asset distorts the result noticeably.

One last detail that rarely gets said out loud. Correlation is non-stationary. It isn't a constant property of an asset like its ticker, it's a snapshot of market conditions at a specific moment. A poor foundation for a strategy, a decent regime filter.

If the order book and footprint charts are still hard to read, start with the free lesson on the order book and entry points. It's part of the full free course "Trading from Scratch | Free Course on Crypto Trading and Scalping" on our YouTube channel.

Bitcoin and the S&P 500

I'll start with the current numbers, they show the problem best. According to Newhedge data as of late August 2026, bitcoin's 30-day rolling correlations look like this.

Right now bitcoin has no connection to equities. Zero. Meanwhile with gold it's 0.76, and that's already a strong correlation.

Some perspective on the range: in March 2026 the bitcoin S&P 500 correlation reached 0.74, the year's high, and the intraday coefficient of determination touched 0.94 (hence the headlines about "bitcoin tracking equities 94% of the time"). Around the turn from 2025 into 2026 the reading fell to -0.299. The average 90-day value over five years holds near 0.30.

That number is the honest answer to the question of whether bitcoin correlates with the stock market. Weakly, on average, with rare outliers in both directions.

Why the connection exists at all

The first mechanism, and the main one, is shared sensitivity to liquidity. The Fed tightens, money gets more expensive, multiples compress on assets with no cash flow. Tech stocks and bitcoin land in the same bucket of long-duration risk bets. How rate decisions travel into crypto is covered in detail in a separate piece "Macroeconomics and Bitcoin: How Fed Rates Move Crypto".

The second, shared holders. After the spot ETFs launched in January 2024, bitcoin ended up in portfolios sitting next to equities and bonds, and assets under management in those funds have reached roughly $90 billion. A portfolio manager needs to cut risk, so he trims the Nasdaq position and the IBIT position. One move, one risk limit. A mechanical link that physically didn't exist before. How the funds themselves are structured and where the flows come from is covered in a separate article "Bitcoin ETFs: What They Are and How They Affect the Market".

The third, trading hours. Most of the correlation is accumulated during the US session, from 16:30 to 23:00 Kyiv time. Bitcoin trades around the clock, the index doesn't. In the Asian session the tape is noticeably thinner, and moves are more often driven by internal factors: funding rates and liquidations.

What history shows

The pattern is visible. The metric spikes during market-wide stress and collapses to zero when crypto gets a storyline of its own. March 2020 showed it in its purest form: in a liquidity crisis every risk asset's correlation goes to one, because people sell everything that can be sold fast. The conclusion is unpleasant for anyone holding bitcoin "to diversify the portfolio." The diversification disappears exactly when it's needed most.

What to do with it intraday

I usually look at the index futures before the US session opens, but not as a signal, as a warning about the character of the day. ES down a percent or more in the morning? The odds that you'll have to close your long earlier than planned go up noticeably.

Then comes the gap between correlation on daily charts and reality in the order book. An inflation print at 15:30 Kyiv time produces a synchronized reaction for the first five to fifteen minutes. Then the link falls apart: in bitcoin, after the first impulse, its own machinery kicks in, stops get hit, leveraged positions start getting liquidated (forced closure of a position by the exchange when margin runs short), and the move is driven by open interest structure, not by the macro data.

Trading bitcoin off the index chart is pointless. Tracking the macro calendar is mandatory.

Bitcoin and the DXY (dollar index)

The DXY measures the dollar against a basket of six currencies, and the weights there are far from even.

• euro, 57.6%
• Japanese yen, 13.6%
• British pound, 11.9%
• Canadian dollar, 9.1%
• Swedish krona, 4.2%
• Swiss franc, 3.6%

More than half of the index is effectively EUR/USD flipped upside down. Keep that in mind when you read about "dollar strength." Often what's being described is euro weakness.

For years the inverse logic looked like this: the dollar rises, financial conditions tighten, dollar funding gets more expensive, investors move to cash, bitcoin falls. The historical correlation held in the -0.5 to -0.7 range, one of the most reliable intermarket relationships in crypto.

In 2026 it broke.

The 2026 break

JPMorgan analysts documented that bitcoin's correlation with the DXY flipped into positive territory. It didn't weaken, it changed sign.

A telling day, March 3, 2026. The dollar index hit a three-month high of 99.4, the Nasdaq 100 lost a percent, gold fell 3.6%. Bitcoin didn't budge, held above 68 thousand, and on that same day $1.5 billion of fresh money went into the spot ETFs. Classic intermarket logic would have predicted a decline. It rose against three factors at once.

In the three weeks before that, the DXY traveled from 96.6 to 99.4, while bitcoin's 30-day correlation with the Nasdaq 100 fell from 92% to 69%. Crypto was decoupling from tech and losing its fear of the dollar over the same stretch.

The reason I see is structural. While bitcoin was a retail speculative asset, it responded to the cost of dollar funding directly. Once pension funds and endowments came in through the ETFs, it became a portfolio allocation position, and positions like that don't get cut over a two-point move in the dollar index. They get reviewed once a quarter, per mandate.

The DXY has more practical value than the S&P 500 for one reason. The dollar index trades almost around the clock, while the stock market runs eight hours a day, five days a week. For an asset that lives 24/7, that matters.

A separate note on weekends. Equity and FX markets are closed, bitcoin trades, and any talk of correlation at that point is meaningless. Weekends are exactly when the order book is empty, liquidity drops several times over, and one large order moves price a full percent.

Bitcoin and gold

Right now this is bitcoin's strongest link to the outside world. A 30-day correlation of 0.7558.

The reason is specific and datable. In August 2026 the theme the market calls the debasement trade came back. The trigger was Treasury Secretary Scott Bessent's actions in the debt market, in particular building a reserve of roughly $950 billion in accounts at the Fed. The 10-year Treasury yield held at 4.74% on August 21, and the real yield on 30-year paper stood at 2.97% on August 24.

The assets' reaction over the week speaks for itself: IBIT added 22.57%, GLD 5.23%. Gold rose more than 18% off its yearly lows and approached resistance at 4,716, its fourth straight up week. Bitcoin over the same stretch traded around 78,750-80,270.

Both rose on the same narrative. Hence the high correlation.

Where the resemblance ends

Cold shower. Take the same period on a longer horizon.

A 70 percentage point gap over a year. These are assets that happened to line up in the moment, not assets with a shared role in a portfolio. The high 30-day correlation between gold and bitcoin reflects short-term positioning, not structural kinship.

There's also a purely mechanical difference that any scalper notices. Gold's daily volatility in a calm regime holds around 0.8-1.2%, bitcoin's in the same regime is 2-3.5%. Even at a correlation of 0.9 you can't work them the same way, position size and stops are calculated on different scales.

The BTC to gold ratio

A simple derived metric I keep on a separate chart: bitcoin's price divided by the price of an ounce of gold. It strips the shared dollar-weakness factor out of the picture.

When both assets rise, it's unclear whether that's money flowing into bitcoin or simply the debasement of the currency both are priced in. The ratio settles the question. If it rises, bitcoin is winning the competition for capital inside the inflation-hedge theme itself. If it falls, capital prefers the metal and bitcoin is just riding the same wave. Right now it's around 17 ounces per bitcoin with the metal near 4,700, and over twelve months the reading has compressed dramatically.

Briefly on "digital gold," then. The narrative sounds good and holds up poorly against the data: over the year gold did +37% on geopolitics and central bank buying, while bitcoin dropped a third. Two instruments expressing one idea don't diverge like that.

When correlation breaks

A break almost always means one thing. Bitcoin has found its own reason to move, and it outweighs the general backdrop. There are four types of reason.

Crypto-specific events. A halving, a fund launch, a major hack, a regulatory decision. In May-June 2019 bitcoin rose 62% in anticipation of the halving while the S&P 500 lost 6.5%. In 2023 speculation on spot ETF approval delivered +147% against the index's +26%. The mechanics of how the halving affects supply are covered in a separate article "The Bitcoin Halving: Mechanics and Price Impact".

Liquidation cascades. The fastest way to zero out a correlation within an hour. On August 19, 2026, roughly $1.29 billion of short positions were liquidated during the densest stretch of the cascade, 92% of liquidations were shorts, and open interest dropped about 15%. What the S&P 500 was doing at that moment was irrelevant.

Weekends and holidays. A physical constraint: the index doesn't trade, so there's nowhere for a correlation to come from. This is exactly where the most deceptive moves show up, because the order book is empty and price jumps levels with almost no trades.

Structural shifts. The rarest and the most interesting. A change in who holds the asset changes its reaction permanently, not for a week. That's precisely what happened with the DXY in 2026: $56.9 billion of inflows into the spot funds since January 2024 didn't change the price, it changed who sets it.

All four types share one problem. The break is visible after the fact, once someone has calculated it. In the moment, all you see is bitcoin failing to deliver the expected reaction to macro data, and you have to decide whether that's noise or a regime change. What helps you decide faster than the chart is the tape: the divergence between the expected reaction and how density levels in the order book actually behave shows up in the first minutes after the release, while on the daily chart it becomes obvious a week later.

How to use it in trading

Most material on correlation ends with pretty charts and never answers the question of what to do with it on Monday morning. Four working rules.

Rule one, correlation sets the backdrop, not the entry

A reading of 0.74 with the S&P 500 doesn't mean "buy bitcoin when the index goes up." It means that in the current regime the odds of a long working out are higher on days the index rises and lower on days it falls. An adjustment to probability, not a signal.

The entry was and remains in the order book and the tape. Correlation answers the question of which trades I'm even considering today, not when to click.

Rule two, the calendar matters more than the coefficient

Events where the bitcoin and stock market link works almost every time:

• US consumer price index, 15:30 Kyiv time
• the FOMC rate decision, 21:00, and the press conference half an hour later
• non-farm payrolls, the first Friday of the month
• speeches by the Fed chair, including the Jackson Hole symposium
• producer price index and retail sales

In the plus or minus fifteen minute window around these releases, correlations across all risk assets spike, whatever the reading was the day before. Afterward everything drifts back apart.

My approach is simple. Ten minutes before the release I either close the position or take half off and move the stop to breakeven. Slippage at the moment macro data hits easily reaches 0.3-0.5% even on BTC/USDT, where liquidity is the best in the market.

Rule three, divergence tells you more than agreement

Moves lining up is backdrop. Divergence is a signal.

That March 3, 2026 case is instructive. The dollar at a three-month high, tech down, gold off 3.6%, and bitcoin standing still. Three factors against it, and the asset doesn't fall. That means there's a buyer ignoring all three, and that day it turned out to be $1.5 billion of ETF inflows. An asset that doesn't fall on bad news usually rises once the news runs out.

The reverse works too. The index is rising, bitcoin can't lift, which means there's selling inside crypto that the external backdrop isn't covering. Check the funding rate (periodic payments between longs and shorts on perpetual futures) and open interest before you climb into a long.

Rule four, count your real portfolio risk

Holding a bitcoin position and a tech stock position at a correlation of 0.8? You don't have two positions, you have one at double size. Plenty of people find this out on the day both go red in sync. With r near zero the risks add up almost linearly, at r = 0.8 total risk is substantially higher than the sum of the parts.

A worked example

The setup. Late August 2026, bitcoin has approached resistance at 80,589. The bitcoin S&P 500 correlation is 0.02, with gold 0.76. Gold is in its fourth up week and closing in on its own resistance at 4,716.

What follows from that. Switch off the stock market as a reference point, there's no link. The main external driver is the debasement trade, and within it bitcoin is secondary to gold. So gold becomes the leading indicator, and as it approaches 4,716 the odds of profit-taking across the whole pair rise.

A long at resistance 80,589 in that situation requires either reduced size or confirmation in order flow. What counts as confirmation: a density level in the order book (a cluster of limit orders at one price) above 80,589 that has been sitting for more than twenty minutes and has started getting eaten by market buys. Not an order blinking for four seconds, but steady size with a verified lifetime. Plus positive delta in the footprint (volume grouped by price levels inside the candle) on the breakout itself. The three tools only work together: the order book shows intent, the tape shows execution, the footprint shows who won.

No confirmation, and gold running into resistance? Then the long is buying the top of two assets at once.

Common mistakes

Using correlation as a trading signal. Buying bitcoin because the Nasdaq is rising means trading with a lag behind the faster market. Crypto prices risk-on earlier, not later.

Ignoring the lookback window. The argument over whether bitcoin correlates with equities is usually run by two people, one looking at a 30-day window and the other at a yearly one. Both are right and both are useless.

Believing "digital gold" is a permanent property. Over the last twelve months gold is +37%, bitcoin -33%. The narrative and the data have parted ways, and the data wins.

Diversifying a risk portfolio with bitcoin. In a liquidity crisis every correlation goes to one. March 2020 demonstrated it in a week.

Trading crypto on the stock market's schedule. Bitcoin runs 24/7, and a substantial share of the year's movement happens during hours when the New York exchanges are closed.

Transferring BTC's correlation onto altcoins. Alts correlate first with bitcoin and only through it with the outside world. The direct link between ETH or SOL and the S&P 500 is almost always weaker and lagged.

A breakdown of how professionals read the market through the order book and footprint charts is in the free lesson from the same course. It's about the mechanics that replace correlation at the moment of entry.

FAQ

• Does bitcoin correlate with the stock market?

  Weakly, on average. The 90-day correlation with the S&P 500 over five years holds near 0.30, and as of August 28, 2026 the 30-day reading has dropped to 0.02. Yet in March 2026 it reached 0.74. The relationship exists, but it's unstable and shows up mainly during periods of market-wide stress.

• What is bitcoin's correlation with the S&P 500 right now?

  0.0217 on a 30-day rolling window as of late August 2026. Practically zero. At the same time the correlation with gold is 0.7558, and with Treasury bonds 0.3293. Right now bitcoin behaves like a hedge against currency debasement, not like a risky tech stock.

• Why did bitcoin stop falling when the dollar rises?

  Because of a change in who holds it. Assets under management in the spot ETFs have reached roughly $90 billion, and bitcoin has become a portfolio allocation position for institutions rather than only a leveraged retail speculation. JPMorgan documented in 2026 that BTC's correlation with the dollar index flipped positive. A telling day was March 3, 2026, when the DXY hit 99.4 and bitcoin held above 68 thousand on $1.5 billion of fund inflows.

• Is bitcoin digital gold?

  By correlation, right now it looks that way. By results, no. The 30-day correlation with gold is 0.76, and both assets are rising on the fiat debasement theme. But over twelve months gold added 37.38% while bitcoin, via IBIT, lost 32.62%. Lining up in the moment doesn't mean a shared role in a portfolio.

• How do you use correlation in scalping?

  As a regime filter, not as a signal. Correlation determines which trades you're considering today and at what size, while the entry still comes from the order book and the tape. The practical application runs through the calendar: in the fifteen-minute window around an inflation print or an FOMC decision, correlations across all risk assets spike, and sitting in a position there isn't worth it.

• What matters more for bitcoin, the S&P 500 or the DXY?

  Historically the DXY, because it trades almost around the clock and directly reflects the cost of dollar liquidity. But in 2026 that link broke and the sign of the correlation turned positive. Right now gold is the more informative reference point. The general rule: watch not for a specific asset, but for which asset is currently leading the pair.

• How does correlation affect altcoins?

  Indirectly. Altcoins correlate with bitcoin at 0.7-0.9 and only through it with external markets. An alt's direct link to the S&P 500 is usually weaker and delayed. In practice that means the macro backdrop is secondary for trading alts, while bitcoin dominance and liquidity rotation inside crypto come first.

Bitcoin's correlation with the stock market isn't a property of the asset, it's a description of the current regime. A number without a lookback window is useless, and divergence tells you more than agreement. And it never gives you the entry: the entry is where it always was, in the order book and the tape.

Watch the market where liquidity is visible. Secret Terminal brings together the order book with a density map to a depth of 5%, the tape, footprint charts with delta, and funding rates across Binance, Bybit, OKX, MEXC and WhiteBIT in one window. The order lifetime timer separates confirmed capital interest from spoofing, and recalculating your filters for higher volatility takes seconds. At the moment macro data lands, when the spread widens and depth drops, the difference between a real density level and a blinking order decides the trade. The terminal is free, runs on Windows and macOS, and keeps your data and API keys stored locally.