I still remember the first time someone told me crypto would act like digital gold when stocks tanked. It sounded convincing at the time. Then the data started rolling in, and the picture looked nothing like the story we had been sold. Bitcoin has fallen on 92 percent of the U.S. stock market’s worst days since 2022. That single figure should make any investor pause and rethink the idea of crypto as reliable protection during equity selloffs.
What The Latest Numbers Actually Reveal
Recent research dug into the relationship between Bitcoin, Ether, and major U.S. stock indices across a long stretch of market history. The results are hard to ignore. When equities suffered their heaviest losses, both leading cryptocurrencies frequently moved in the same direction, often with greater intensity. On ordinary trading days the two assets behaved more independently. During the worst sessions the connection tightened sharply.
Looking at the full sample of sessions since mid-2018, researchers identified more than a hundred days that ranked among the S&P 500’s poorest performers. Bitcoin declined in roughly four out of five of those sessions. Ether showed a similar pattern. The gap between normal days and stress days stood out clearly. On routine sessions each cryptocurrency finished lower about half the time. During severe equity declines the loss frequency jumped by roughly thirty percentage points for both assets.
Restricting the window to 2022 onward produced an even starker result. Bitcoin fell on 92 percent of those worst equity days. Ether declined on 95 percent of them. The study used the period’s own threshold for the worst five percent of sessions rather than simply carrying over the longer-term cutoff. That choice makes the finding more relevant to the market environment many investors face today.
I’ve found that numbers like these tend to cut through the noise faster than any narrative. People can debate theory all day. When the same pattern appears across dozens of stressful sessions, the practical message becomes harder to dismiss.
Average Correlations Tell Only Part Of The Story
Over the entire multi-year period the average correlation between Bitcoin and the S&P 500 sat around 0.29. Ether’s reading came in slightly higher at 0.31. Those figures look moderate on paper. They even turned slightly negative during parts of 2018 and 2019. By 2026 the readings had climbed closer to 0.50. Moderate averages can hide important differences in behavior during the moments that matter most to portfolios.
The November 5 selloff offered a fresh example. The S&P 500 dropped 4.4 percent while the Nasdaq-100 fell 6.4 percent. Bitcoin declined about 5 percent and Ether dropped more than 7 percent over the same window. Liquidations across the broader crypto market exceeded one billion dollars in a single day, with the large majority coming from long positions. The episode illustrated how quickly leveraged exposure can amplify losses when the correlation tightens.
Perhaps the most interesting aspect is how consistently the pattern holds once you focus on the worst equity sessions. Ordinary correlations give a calm average. Stress periods reveal a different relationship. Investors who rely only on the calm average may find themselves surprised when markets turn turbulent.
Bitcoin Versus Ether During Equity Stress
Within the crypto market itself the research uncovered a clear difference in relative performance. Since 2022 Ether underperformed Bitcoin on 79 percent of the stock market’s worst days. Median losses reached 3.6 percent for Bitcoin and 5.6 percent for Ether. That two-percentage-point gap may sound modest until you consider the frequency with which it appeared.
The exchange rate between the two assets told a similar story. On the worst equity days the median change in the Ether-to-Bitcoin rate was negative 1.52 percent. On the best equity days the median change turned positive 0.67 percent. The spread between those two groups exceeded two percentage points. Removing the extreme moves of March 2020 did not erase the pattern.
In my experience, relative strength during stress periods often carries more weight for portfolio construction than average returns over calm stretches. Holding the stronger of the two assets during equity selloffs still produced negative median returns. It simply produced less severe ones. That distinction matters when every percentage point of loss compounds across multiple stressful sessions.
Both assets posted negative median dollar returns during the recent sample’s worst equity sessions. Relative strength does not equal absolute safety.
Why U.S. Trading Hours Matter More Than Expected
Another layer of the research examined hourly price variation after the launch of U.S. spot Bitcoin exchange-traded products in early 2024. The analysis defined the core cash-session window as the period from 10 a.m. through the actual close of the New York equity market. That window represents only 17.1 percent of observed hours. Yet it accounted for 33 percent of Bitcoin’s total price variation and 30 percent of Ether’s.
In a typical U.S. trading hour Bitcoin’s average absolute move measured more than twice the size of its weekend-hour move. Ether’s ratio came in a bit lower at 1.78. Stripping out the largest one percent of hourly moves barely changed the results. The concentration of volatility during overlapping equity hours suggests that institutional flows and risk-management activity linked to traditional markets play a larger role than many retail participants assume.
The study measured the timing of volatility without claiming that stock trading directly causes every crypto move. Still, the disproportionate share of variation that occurs while U.S. equities are open is difficult to overlook. When equity desks are active, crypto prices tend to swing harder. When those desks quiet down, the swings often moderate.
I’ve watched this pattern play out in real time more than once. A quiet weekend can give way to sharp moves the moment the New York open approaches. That rhythm has implications for anyone trying to time entries or manage leverage across the 24-hour crypto clock.
Exchange Flows And The Stablecoin Backdrop
Alongside the correlation findings, the same research brief noted substantial net inflows onto exchanges. Bitcoin saw more than three billion dollars in net transfers to trading platforms in a single week. Ether inflows added another billion. Combined, the figure reached roughly 4.2 billion dollars. Exchange inflows do not automatically equal selling. They do increase the supply of coins available for immediate trade.
At the same time, the circulating supply of stablecoins declined by nearly five billion dollars, settling near 288 billion. The contraction stretched across more than two consecutive weeks of daily declines. A shrinking stablecoin base can remove a source of potential buying power even while coins move onto exchanges. The combination creates an environment in which selling pressure can meet thinner demand.
These flow figures sit in the background of the price action rather than explaining every move. Yet they help illustrate the broader conditions under which the high correlation during equity stress can become more costly. More coins on exchanges and fewer stablecoins in circulation leave less cushion when risk appetite suddenly fades.
Rethinking The Diversification Narrative
For years the case for including crypto in a traditional portfolio rested partly on the idea of low or even negative correlation during equity downturns. The data since 2022 challenge that assumption more directly than the longer-term averages suggest. When stocks experience their sharpest losses, Bitcoin and Ether have tended to join the decline rather than offset it.
That observation does not mean crypto has no place in a portfolio. It does mean the justification needs to rest on other grounds. Growth potential, technological exposure, or long-term adoption trends may still appeal to certain investors. Treating the asset class primarily as a hedge against equity selloffs looks harder to defend on the recent evidence.
In practice, many portfolios that added crypto for diversification discovered that the protection failed exactly when they needed it most. The psychological impact of watching both the equity sleeve and the crypto sleeve fall together can be larger than the pure mathematical effect. Risk models that rely on average correlations often underestimate the joint losses that appear in the left tail of the distribution.
- High loss frequency during the worst equity sessions undermines the hedge narrative
- Relative underperformance of Ether versus Bitcoin still leaves both assets negative on median
- Concentration of volatility in U.S. trading hours points to shared risk channels
- Elevated exchange inflows and falling stablecoin supply reduce potential buffers
None of these points require abandoning crypto entirely. They do call for clearer thinking about the role the asset is expected to play. If the goal is pure diversification against equity stress, the recent track record raises legitimate questions.
Practical Implications For Portfolio Construction
Investors who still want exposure can adjust position sizing and risk controls with the correlation data in mind. Smaller allocations reduce the damage when the joint selloff arrives. Explicit limits on leverage become more important once the tendency for simultaneous declines is recognized. Some participants may choose to treat Bitcoin as the relatively stronger of the two major assets during stress periods while still accepting that both can lose value.
Time-zone awareness also gains importance. Because a disproportionate share of price variation occurs during U.S. equity hours, risk decisions made overnight or over weekends can face a different volatility regime once the cash market opens. Monitoring tools that flag rising correlation in real time may help some active managers respond faster, though no tool eliminates the underlying relationship.
I’ve seen portfolios that treated crypto as a permanent diversifier struggle more than those that viewed it as a satellite growth position with its own risk budget. The second approach tends to produce fewer unpleasant surprises when equity markets turn sharply lower. The first approach often leaves investors wondering why the protection they expected never materialized.
Position sizing remains the most controllable variable for most individual investors. A five-percent allocation that declines alongside stocks still hurts, yet the absolute impact stays limited. A twenty-percent allocation that behaves the same way can dominate the overall portfolio return for that period. The correlation data simply make the sizing decision more consequential.
Looking At The Longer Historical Context
The full sample that begins in 2018 already showed elevated loss frequencies during equity stress. The post-2022 window simply intensified the pattern. Correlations that were modest or even negative in earlier years moved higher as institutional participation grew and as more traditional capital entered the space through regulated products. That shift is consistent with the idea that broader adoption can bring crypto price behavior closer to that of other risk assets.
None of this proves the relationship will remain fixed forever. Market regimes change. New participants arrive. Liquidity conditions evolve. Yet investors who dismiss the recent evidence as temporary may be underestimating how structural the shift has become. The concentration of volatility during U.S. trading hours points to channels that are unlikely to disappear overnight.
One useful way to think about the data is to separate average behavior from tail behavior. Average correlations can remain moderate while the probability of joint losses during the worst equity sessions stays high. Portfolio stress tests that focus only on average correlation will miss the episodes that actually drive most of the pain.
In my view, the more honest framing is that crypto has behaved like a high-beta risk asset during many of the equity market’s most difficult days. That characterization does not make the asset unattractive for every investor. It simply replaces an optimistic diversification story with a clearer description of observed behavior.
What Investors Can Do With The Information
First, update any mental model that still treats Bitcoin primarily as an equity hedge. The post-2022 evidence does not support that view. Second, examine current position sizes against the possibility of simultaneous declines. Third, pay attention to the hours when most of the volatility occurs and adjust monitoring or execution plans accordingly. Fourth, watch relative performance between Bitcoin and Ether during stress periods if the portfolio holds both.
None of these steps require perfect foresight. They simply incorporate the empirical pattern into decision-making. Markets will continue to produce both calm stretches and sudden shocks. The correlation data help set more realistic expectations for the second category.
Some investors may decide the growth case still justifies an allocation even after acknowledging the lack of diversification during equity stress. Others may reduce exposure or restructure how they hold the asset. Either choice becomes clearer once the 92-percent figure is taken seriously rather than explained away.
I’ve found that the most durable portfolios are the ones built on observed behavior rather than preferred narratives. The recent research supplies a useful set of observations. Ignoring them because they conflict with an earlier story does little to improve outcomes when the next sharp equity selloff arrives.
The Emotional Side Of Simultaneous Declines
Numbers alone never capture the full experience of a portfolio. Watching equities and crypto fall together can feel more destabilizing than the same total loss spread across independent assets. The sense that “nothing is working” often arrives faster when the supposed diversifier moves in lockstep with the core holdings. That psychological pressure can lead to poorly timed exits that lock in larger losses than necessary.
Understanding the statistical tendency in advance can reduce the surprise element. When investors already know that joint declines have been common during the worst equity sessions, the same price action feels less like a betrayal of the strategy and more like a known feature of the current market regime. That shift in framing sometimes helps people stick with a carefully sized plan instead of abandoning it at the bottom.
Of course, awareness does not eliminate the discomfort. It simply removes one layer of unexpectedness. For many participants that difference still improves decision quality under stress.
A Balanced View Going Forward
The data do not declare crypto dead or irrelevant. They do challenge one of the more popular justifications for owning it alongside traditional stocks. Growth-oriented investors can still make a case based on long-term adoption, technological development, or asymmetric upside. Investors seeking uncorrelated returns during equity crises will find the recent record less supportive.
Future regimes could look different. Greater institutional involvement might eventually stabilize certain aspects of the market, or it could further align crypto with other risk assets. No one can say with certainty which path will dominate. What can be said with greater confidence is that the period since 2022 has produced a high frequency of joint losses on the stock market’s worst days.
For now, the practical response is straightforward. Size positions with the observed correlation in mind. Avoid relying on crypto as a primary equity hedge. Monitor the hours when most of the price variation tends to occur. Keep relative performance between the major assets on the radar. And update the narrative to match the evidence rather than the other way around.
That approach will not guarantee perfect outcomes. It does increase the odds that expectations remain realistic when the next round of equity stress arrives. In a market that already contains enough surprises, realistic expectations remain one of the more useful edges available to ordinary investors.
The 92-percent figure is not a prediction. It is a description of what has already occurred across dozens of difficult sessions. Treating that description as relevant information rather than an inconvenient outlier seems like the more durable way to navigate the current environment.
