<!-- Generated from src/ data by scripts/generate-llm-files.ts. Do not edit by hand. -->
# Stock market crash or correction? Why the Nasdaq fell 5.5% on November 5

> The Nasdaq fell 5.52% and the S&P 500 4.36% on November 5, 2026, their worst day since April 2025. Why stocks fell, how it compares to past crashes, and what to watch next.

By Maya Koeva · 2026-11-06

<Tldr cta="See today's AI stock sentiment">
No, not by the usual definition. The S&P 500 fell 4.36% on November 5, 2026, its worst day since April 2025, and the Nasdaq fell 5.52% into a correction, 11.9% under its October 6 record; a crash usually means 20% or more in days or weeks. The finance accounts we track had already turned less bullish on the AI stocks that led the fall, while staying upbeat on everything else.
</Tldr>

<KeyFacts>
  <Fact label="S&P 500">7,353.73 on November 5, 2026, down 4.36%</Fact>
  <Fact label="Nasdaq Composite">24,315.00, down 5.52%, 11.9% under its October 6 record</Fact>
  <Fact label="Dow Jones">51,523.06, down 1,833.06 points (3.44%), erasing November 4's 3.03% gain</Fact>
  <Fact label="Worst day since">April 4, 2025, the tariff shock, for all three indexes</Fact>
  <Fact label="What drove it">Fear that a Democratic Congress slows the AI data center build-out</Fact>
  <Fact label="Next on the calendar">Jobs report November 6, Nebius earnings November 12, Nvidia earnings November 17</Fact>
</KeyFacts>

![A thin hairline pendulum hanging from a short bar at the top of a dark charcoal frame, swung far to the right, its round bob filled flat coral, with a faint dotted arc tracing the swing back toward the center.](/blog/stock-market-crash.webp)

US stocks had their worst day in 19 months on Thursday, November 5, 2026. The S&P 500 fell 4.36%
and the Nasdaq Composite 5.52%, which left the Nasdaq 11.9% under its October 6 record. A drop of
10% from a high is a correction, so that is where the Nasdaq stands. A crash usually means 20% or
more in days or weeks, and neither index is there.

AI stocks took the worst of it, on fears that the new Democratic Congress will slow the data
center build-out. The finance accounts we track had turned less bullish on those stocks in the
three days before the drop, and kept their calls on the rest of the market upbeat.



## What happened on November 5

The selling started in chips and spread. All 11 S&P 500 sectors closed lower, and technology fell
7.30%, its worst day since March 16, 2020, CNBC reported.

- **Chips:** the Philadelphia Semiconductor Index fell more than 12%. Nvidia fell 9.26% to $181.60
  and lost more than $440 billion of market value. AMD fell 17.3%.
- **The biggest tech names:** the Magnificent 7 lost more than $1 trillion of value between them.
  Tesla fell 10.77%.
- **AI data centers:** CoreWeave fell 9.8%, and IREN, Applied Digital and Nebius each fell more
  than 14%.
- **Losers beat winners:** on the Nasdaq, 4,166 stocks fell and 855 rose.

The Dow lost 1,833.06 points, which wiped out its 3.03% gain on November 4, the day after the
election. On Friday morning in Asia, Japan's Nikkei fell 5.88% and South Korea's Kospi 6.06%, after
a circuit breaker paused trading in Seoul at the open.

## Why the stock market fell

Democrats won the House and the Senate on November 3. On November 5, investors sold the companies
that sell into AI data centers, on the fear that a Democratic Congress will slow new
projects through tighter rules on power, water and permits.

In Texas, Gina Hinojosa ran for governor on halting new data center
approvals in the state until lawmakers set rules for them. Texas is one of the biggest data center
markets in the country, and the chip, networking and cloud companies that sold off hardest depend
on those projects getting built.

AI stocks were already down before the vote. The Nasdaq peaked on October 6, and AI stocks spent October
falling on rates, earnings and debt worries, which our
**[AI bubble breakdown](/blog/ai-bubble)** covers. Specialist GPU clouds such as CoreWeave and
Nebius, known as [neoclouds](/learn/what-is-a-neocloud), borrow heavily to build, so they have the
least room if projects slow.

## How it compares to past stock market crashes

| Selloff | Index | Worst day | Peak to low | Back to the old high |
|---|---|---:|---:|---:|
| 1929 crash | Dow | -12.8% (Oct 28) | -89% by July 1932 | 1954 |
| 1987, Black Monday | S&P 500 | -20.5% (Oct 19) | -34% by December | 1989 |
| 2000 to 2002, dot-com bust | Nasdaq | -9.7% (Apr 14, 2000) | -78% by October 2002 | 2015 |
| 2007 to 2009, financial crisis | S&P 500 | -9.0% (Oct 15, 2008) | -57% by March 2009 | 2013 |
| 2020, Covid crash | S&P 500 | -12.0% (Mar 16) | -34% by March 23 | August 2020 |
| 2025, tariff shock | S&P 500 | -6.0% (Apr 4) | -19% by April 8 | June 2025 |
| 2026, AI selloff | Nasdaq | -5.5% (Nov 5) | -11.9% so far | |

On this scale, November 5 sits at the small end. Black Monday in 1987 took more than a fifth
off the S&P 500 in one session, and on Black Tuesday in 1929 the Dow lost another 11.7% after
falling 12.8% the day before.

The recoveries split two ways. The 2020 and 2025 drops came from a shock outside the market and
were back at their highs within months. The dot-com bust and 2008 began with prices or debt that
had run too far, and took years. AI bears argue this one is the second kind, spending that ran
ahead of revenue; the bulls point to the profits the biggest buyers earn.

## What the accounts we track said before the drop

We looked at 13 AI chip, optics and data center stocks: Nvidia, AMD, Broadcom, Micron, TSMC,
Marvell, CoreWeave, Nebius, IREN, Oracle, Coherent, Lumentum and Applied Optoelectronics.

*[Chart: Mentions of 13 AI chip, optics and data center stocks across the accounts Quantral tracks, October 7 to November 5, 2026, by US Eastern day.]*

Over the 30 days to November 1, the finance accounts we track called those stocks bullish 70% of
the time, 2,855 calls to 1,216. From November 2 to 4, the three days before the drop, that fell to
62%, 408 to 252, while daily mentions of them rose 60%. Their calls on every other stock went the other
way, to 69% bullish from 65%.

The accounts with a graded track record moved further. They were 80% bullish on the 13 AI stocks
in the prior 30 days and 67% from November 2 to 4, while their calls on other stocks held at about
75%. Lumentum shows the turn: 208 bullish calls to 25 over the 30 days, then 69 to 57 from
November 2 to 5.

On November 5 itself, the accounts we track called the AI names 65% bullish, 162 to 89, up from the
three days before. Graded accounts were at 62%. They had cut their bullish share on these stocks in the days before
chips fell 12%, and stayed cautious on the day itself.

<MidCta heading="See who turned cautious on AI stocks" text="The Quantral app shows each stock's score with every call behind it and the record of each account that made one." />

## What long-term investors can take from it

None of this says what to buy or sell. A few things hold across the table above.

- **Corrections are common.** CFRA has counted 24 falls of 10% or more from a record since World
  War II that stopped short of a 20% bear market. On average, the S&P 500 took about eight months to
  get back to its high, the Associated Press reported in March 2025.
- **Big up days sit next to big down days.** The Dow rose 3.03% on November 4 and fell 3.44% on
  November 5. Selling after a day like that often means missing the rebound, which
  **[long-term investing and market sentiment](/learn/long-term-investing-and-market-sentiment)**
  walks through.
- **A first bounce can fail.** Sharp rallies inside a falling market are common, and a
  **[dead cat bounce](/learn/what-is-a-dead-cat-bounce)** explains how to tell them apart.
- **Know what you own.** A 12% day for chip stocks shows what heavy AI exposure feels like. If it felt
  like more risk than you signed up for, check your mix before the next one. Investors who keep adding
  on a schedule use **[dollar-cost averaging](/learn/what-is-dollar-cost-averaging)** to take the
  timing out of it.

## What to watch next

1. **The October jobs report,** Friday November 6 at 8:30 a.m. ET. A strong number can push rates up
   and hit tech, which **[what is the jobs report](/learn/what-is-the-jobs-report)** explains.
2. **Asia's reaction.** The Nikkei and Kospi each fell about 6% on Friday.
3. **Nebius earnings,** Thursday November 12, before the open. Our **[Nebius stock page](/blog/nebius-stock-sentiment)** has the setup.
4. **Nvidia earnings,** Tuesday November 17, after the close. The bull and bear cases are in our
   **[Nvidia stock forecast](/blog/nvidia-stock-forecast)**.
5. **The rules themselves.** What the new Congress and Texas do on data center approvals,
   which will take months to show.

The stock market is open on Veterans Day, November 11; the bond market is closed. Other dates are in
**[stock market holidays](/learn/stock-market-holidays)**.

## Common questions

### Is the stock market crashing?

Not by the usual definition. The S&P 500 fell 4.36% and the Nasdaq 5.52% on November 5, 2026, the
worst day since April 2025. The Nasdaq closed 11.9% under its October 6 record, which is a
correction. A crash usually means a fall of 20% or more in days or weeks.

### Will the stock market crash more?

No one can say ahead of time. Since World War II, most falls of 10% stopped
short of 20%. Friday's jobs report, the Nebius and Nvidia earnings in the next two weeks, and any
action on data center rules are the things most likely to move AI stocks from here.

### What should I do if the stock market crashes?

We don't give buy or sell advice. Long-term investors usually start by checking whether they still
believe the reasons they bought, and whether their mix of stocks matches the risk they can live
with. Selling right after a big down day has often meant missing the rebound.

### Is this like the dot-com bubble?

In part. Both times, a new technology pulled in money faster than it earned revenue. The
difference is profits: the biggest AI companies earn a lot and pay for much of the build-out from
cash flow, which most dot-com companies could not. The Nasdaq is down 11.9%; in the dot-com bust it
fell 78%.

### How long do stock market corrections last?

CFRA's count of 24 corrections since World War II that did not become bear markets puts the average
at about eight months to get back to the old high. Bear markets vary more: after the 2020 shock the
S&P 500 was back at its high within six months, while the 2000 and 2008 bear markets took years.

## Where to follow the mood

<Stores />

Quantral is an investment ideas app built on alternative data. It shows how the finance accounts
and communities we track feel about each stock, with a score that gives more weight to authors with
a proven track record, and every call behind it. Nvidia's 7-day score was 65 early on November 6.
It runs on the web at [app.quantral.com](https://app.quantral.com), on iOS and on Android, with a
7-day free trial on the yearly plan. For the rest of the group, start with
**[our AI infrastructure stocks list](/blog/ai-infrastructure-stocks)**, or see the monthly
**[stock market sentiment](/blog/stock-market-sentiment)** reading.

---

*Index closes, stock moves and market breadth for November 5, 2026, are from
[CNBC](https://www.cnbc.com/2026/11/05/stock-market-selloff-ai-fears.html),
[Fox Business](https://www.foxbusiness.com/markets/stocks-nov-5-2026) and
[NBC](https://www.nbcbayarea.com/news/national-international/november-5-stock-market/4170982/).
Historical crash figures are closing levels; the CFRA correction count is from the
[Associated Press](https://www.spokesman.com/stories/2025/mar/14/us-stocks-face-uneven-recovery-from-tariff-trigger), March 2025. Mention counts cover the accounts Quantral tracks, by
US Eastern calendar day: October 3 to November 1, 2026, for the 30-day comparison, and November 2 to
5 for the days around the drop. Graded accounts have more than ten resolved calls, most of them
right. Positions attributed to accounts are their views, not Quantral's. Nothing here is financial
advice or a recommendation to buy or sell. Past signals are not indicative of future results.*
