Quick Answer: The cited moves had opposite directions on November 16: the Nasdaq Composite fell about 2.5% from its prior close, while Phemex BTC/USDT spot rose about 0.9% from the 14:30 UTC five-minute bar's open to the 20:55 UTC bar's close. These are different return windows, so the observation does not establish a matched-session correlation. One day cannot establish or disprove a lasting relationship between Bitcoin and stocks.
What happened on November 16?
US stocks closed lower, but Bitcoin did not mirror that direction across the US cash session. The Nasdaq Composite closed at 24,509.76, down 626.52 points, or 2.49%, from Friday's close, according to the Associated Press closing report. Phemex's public BTC/USDT spot candles show an open of 64,508.64 USDT for the 14:30–14:35 UTC bar and a close of 65,098.59 USDT for the 20:55–21:00 UTC bar. That is a 0.91% increase, calculated as (65,098.59 / 64,508.64 − 1) × 100.
| Series | Comparison window | Start | End | Change |
|---|---|---|---|---|
| Nasdaq Composite | Previous US close to November 16 close | 25,136.28 | 24,509.76 | −2.49% |
| Phemex BTC/USDT spot | November 16, 14:30–21:00 UTC | 64,508.64 USDT | 65,098.59 USDT | +0.91% |
Method note: The Nasdaq return includes any gap from its previous close to Monday's opening; the BTC figure covers only Monday's regular US session. The table answers whether their cited moves shared a direction, not whether their session-only returns were statistically comparable. BTC prices come from Phemex's public five-minute spot candles, retrieved November 17, 2026. The API reports prices in scaled integers; the displayed values divide those integers by 100,000,000, following Phemex's API documentation.
Why does the choice of time window change the answer?
Bitcoin trades through the weekend and outside stock-market hours. The Nasdaq Composite represents a market that has set US trading hours. A BTC change from midnight to midnight UTC includes hours when US equities were closed; a Nasdaq close-to-close change includes the overnight gap. Calling those two numbers a daily correlation introduces a timing mismatch.
To compare short-term co-movement, use returns from the same timestamps: for example, BTC at 14:30 and 21:00 UTC and an equity index at its open and close. That requires a reliable intraday equity open and close, or index futures data measured at matching times. This article does not infer an intraday equity return from the Nasdaq close-to-close figure.
The November 16 contrast is therefore narrower than a headline about “decoupling.” It says that the quoted BTC spot session return was positive while the quoted Nasdaq close-to-close return was negative. It does not say BTC was immune to equity news or that the relationship changed permanently.
How would a matched-session calculation work?
Define the return before looking at the chart. For each instrument, use return = (price at 21:00 UTC / price at 14:30 UTC) − 1. The equity leg needs a verified Nasdaq opening level and closing level from the same session. The BTC leg needs a spot price at those two timestamps, with a rule for choosing a five-minute candle: its opening trade at 14:30 and its last trade before 21:00. Rounding should happen only after the calculation.
Then repeat the exercise at a finer interval, such as 30-minute returns, if the question concerns intraday co-movement. Align the bins, exclude missing bins rather than filling them with a fabricated price, and state whether a quote or a completed trade defines each observation. An index print and a BTC venue trade are different data types. They can be compared, but the distinction belongs in the method.
Finally, separate direction from magnitude. Two returns can have the same sign but very different sizes. A correlation calculation uses variation across many pairs of returns; a single pair only provides a dated comparison. The numbers in this article are useful because their sources and timestamps are explicit, not because they settle a general correlation question.
Can one day prove that Bitcoin is uncorrelated with stocks?
No. Correlation measures how two return series move together across many observations. With one paired observation, a correlation coefficient is undefined. Even a rolling correlation can vary by sample length, sampling frequency, market regime, and whether the analyst compares BTC with the Nasdaq, S&P 500, or equity futures.
A repeatable test would choose a period first, align timestamps, calculate log or percentage returns at a fixed interval, and report the number of observations. It would then repeat the calculation across several windows and include confidence intervals. Phemex Academy's asset-correlation guide explains the general measure; the dated calculation above illustrates why selecting a convenient single day is insufficient.
For example, a study of 30 US sessions could take the percentage change from market open to close for each day in both assets. A second study could take 30 full UTC days. The two answers need not match because the first excludes nights and weekends while the second includes them. Neither is automatically the “correct” correlation: the choice depends on whether the reader is managing risk during US cash hours or over an around-the-clock portfolio.
The distinction also matters for the equity benchmark. The Nasdaq Composite and Dow Jones Industrial Average contain different companies and use different index methodologies. A Bitcoin comparison with one cannot stand in for the other. If the original search asks about the Dow, the analysis should use Dow data; if it asks about technology stocks, the Nasdaq may be more relevant. The Google Trends related queries identify topics people searched together, but do not select an analytically valid benchmark.
What can explain opposite moves without proving a cause?
Several mechanisms are plausible, but the prices alone cannot identify which mattered. Bitcoin-specific flows, derivatives positioning, weekend liquidity, the timing of equity news, and differences in who trades each market could all affect returns. One may also see Bitcoin respond before the US opening bell and then recover during the session. Those hypotheses require order-flow, event-time, and cross-venue evidence before they become an explanation.
For readers managing exposure, the useful question is practical: do holdings share risk in the time window that matters to the portfolio? A trader hedging an hourly position needs hourly matched returns; a long-term allocator needs a longer sample and must expect the relationship to change. Neither use case supports treating one day's price contrast as a forecast.
What would falsify the “stocks drove Bitcoin” explanation?
An explanation needs an event sequence and a mechanism, not only two lines on a chart. If BTC moved before the US equity information became public, that equity event cannot explain the earlier BTC move without a separate evidence trail. If BTC moved in the opposite direction during the relevant minutes, a simple one-for-one transmission claim is unsupported. If both moved together but many other risky assets did too, a broad market factor may be more plausible than a direct stock-to-Bitcoin channel.
The checks would include timestamped news, equity-index futures, BTC spot prices from more than one venue, and derivatives funding or liquidation data where available. Each series should use the same timezone and an identified observation frequency. A result that depends on switching from 15-minute to daily data should be described as sensitive to the measurement window. This is a research framework, not a claim that any particular catalyst caused November 16's BTC move.
What should a reader take from the November 16 snapshot?
The observed divergence is a reason to question a blanket statement that Bitcoin “always follows stocks.” It is not a reason to assume the opposite relationship will hold tomorrow. In a portfolio, cross-asset behavior can change during volatility, and the measurement interval can change the apparent direction. The actionable analytical habit is to label the asset, venue, timestamp, benchmark and return definition before discussing correlation.
Readers using the Phemex spot chart can reproduce the BTC leg of this article. To test a portfolio hedge, add the appropriate equity benchmark and a longer matched sample. To decide whether to trade, consider liquidity, fees, and the possibility that historical co-movement changes. A chart comparison does not provide an entry or exit signal on its own.
FAQ
Did Bitcoin rise while the Nasdaq fell on November 16?
Across the cited windows, yes. Phemex BTC/USDT spot gained about 0.91% during the US cash-session hours, and the Nasdaq Composite lost 2.49% measured from its previous close. The differing start points limit direct comparison.
Does a negative one-day comparison mean Bitcoin is a hedge?
No. A hedge needs a defined exposure, a consistent comparison window, and evidence that the relationship holds when the underlying risk appears. One day does not provide that evidence.






