This website uses cookies

Read our Privacy policy and Terms of use for more information.

If you only watched the S&P 500, the last two weeks looked like a bad stretch. If you owned the companies that sell AI chips, they looked like something much worse.

From the close on 28 October to the close on 10 November, the S&P 500 fell 2.9% and the Nasdaq 5.6%. Nvidia, the face of the AI boom, fell 21.6%. In between came the S&P 500's worst day since April 2025.

Below: what happened, day by day, and why three different shocks landed on the same group of stocks. Then what Graham's Mr. Market says to do with a fortnight like this, and what to ignore.

Two weeks, day by day

Two weeks, day by day (index daily change, Nvidia closing price)
DayWhat moved itS&P 500NasdaqNvidia
Thu 29 OctAlphabet miss revives AI spending doubts−0.3%−1.4%$213.66
Fri 30 OctIntel plunges; Amazon’s AI bill−1.8%−2.9%$204.64
Wed 4 NovDemocrats win Congress; oil falls+1.6%+1.4%$200.13
Thu 5 NovAI chip selloff; worst day since Apr 2025−4.4%−5.5%$181.60
Fri 6 NovSmall bounce+0.3%+1.1%$186.64
Mon 9 NovOil drops, chips rally; Nvidia CEO news+2.3%+2.7%$178.67
Tue 10 NovHot inflation report; 10-yr yield above 5%−0.4%−0.7%$168.89
What to notice: The market’s two best days, 4 and 9 November, were both down days for Nvidia. By the end, the selloff had stopped being about AI chips in general.Source: Smart X Terminal closing prices. Event notes from AP, Reuters and New York Times reporting. 2–3 Nov (quiet days before the election) left out. As of 10 Nov 2026.

It started with a question about AI spending

The selling didn't start with the election. On 29 October, Alphabet reported revenue below expectations alongside faster spending on AI. Its shares fell 12.6%, leaving them roughly 30% below their May peak, the New York Times reported.

The next day, Intel fell 28.6%, its biggest one-day drop since 1974, after a weak outlook, and Amazon's results put its AI spending bill back in focus. The Nasdaq fell 2.9%, its worst day since 5 June, and closed 7.8% below its 6 October record.

The question underneath both days was simple. Will the hundreds of billions going into AI data centres earn a return, and how soon? Nobody can answer that yet, which is exactly why prices swung on it.

Money didn't leave the market. It moved. Consumer staples rose almost 2% on 30 October while tech fell, according to the Times.

The election, and the reversal

On 3 November Americans voted, and Democrats won control of Congress. On 4 November the S&P 500 rose 1.6% and the Dow had its best day since April 2025. Oil fell about 9%. The big cloud companies rallied, led by a 6.8% jump in Microsoft, on a bet that a Democratic Congress would cool the AI spending race. The chipmakers that sell into that spending lagged, and Nvidia slipped.

Then came 5 November, and the same result was read the opposite way. Investors sold on worries that the new majority could tighten AI rules and slow approvals for new data centres. Weak outlooks from two chip designers added to it. Nothing had been passed into law. The market was pricing a possibility.

  • The S&P 500 fell 4.4% to 7,353.73, its worst day since the April 2025 tariff shock.

  • The Nasdaq fell 5.5%, leaving it 11.9% below its record. That's a correction.

  • The Dow lost 1,833 points, and the main chip index fell more than 12%.

  • Nvidia fell 9.3% to $181.60, AMD 17.3% and Broadcom 10.4%. Companies building and powering data centres fell too: CoreWeave 9.8% and GE Vernova 7.5%.

Why did a sector selloff become an index selloff? Because AI chip companies are now among the largest companies on the market. A 1% move in Nvidia moves the S&P 500 more than a 1% move in almost any other stock. When the biggest names fall together, the index has nowhere to hide.

Friday brought a small bounce. The S&P 500 still closed the week at 7,376.86, its fourth straight weekly fall since the 6 October record.

Then it became about one company

On Monday 9 November, oil fell sharply and the market rallied hard. The S&P 500 rose 2.3%, the Nasdaq 2.7% and the chip index almost 7%. AMD gained 13% and Broadcom 8.8%. Nvidia fell 4.3%.

The difference was news about Nvidia itself. Over the weekend, chief executive Jensen Huang acknowledged a personal relationship that US officials reportedly link to Chinese intelligence, and Nvidia's independent directors opened an investigation. On Tuesday, The Information reported that the relationship may have helped chips reach Chinese firms. Reuters said it could not verify that report, and nothing has been proven.

Tuesday also brought a hot inflation report. Core prices rose 0.3% for a third month running, the 10-year Treasury yield climbed back above 5%, and the S&P 500 slipped 0.4%. Nvidia fell another 5.5% to $168.89. That's about 29% below its 6 October record close of $239.24.

After the worst day: change from the 5 Nov close to the 10 Nov close
AMD
 
+27.4%
Broadcom
  
+13.9%
Nasdaq Composite
  
+3.1%
S&P 500
  
+2.3%
Nvidia
  
−7.0%
What to notice: Nvidia is the only one of the five that kept falling after the worst day. AMD and Broadcom closed 10 November above where they stood before the selloff began. Grey bar = a fall. Not a ranking of businesses.Source: Smart X Terminal closing prices, 5 and 10 Nov 2026.

That split is the clearest thing the fortnight has shown. Five weeks ago, the market treated AI chips as one trade. This week it priced them one company at a time. That isn't a reason to own any of them. It's a reminder that a theme and a business are different things.

Nvidia is due to report quarterly results on 17 November.

What this does and doesn't tell you

Three separate things hit the same group of stocks: an open question about whether AI spending pays off, a political shock that may or may not become policy, and a governance story at one company. A fourth sat underneath all of them, which was the cost of borrowing. With the 10-year yield back above 5%, money tomorrow is worth less today. That hits hardest the companies whose value rests on profits far in the future.

What it doesn't tell you is what any of these businesses is worth. A falling price answers one question: what will someone pay today? It doesn't answer the one that matters, which is what the company will earn over the next twenty years.

In weeks like this I also check the credit market. In Ray Dalio's debt-cycle framework, the real damage starts when borrowers can't refinance. The high-yield spread, the extra yield investors demand on riskier company debt, was about 3.4 percentage points on 6 November in the Federal Reserve data I track. That isn't what a credit market under strain looks like. So far this is a stock-price event, not a debt event.

The Smart X Terminal cycle reading still places us at Peak, about year 14 of the 18.6-year land cycle.

Where the 18.6-year cycle stands today
Position in the cycle
  
Year 14 of 18.6
StagePeak
Reading date10 Nov 2026
What to notice: This is a stage, not a date. It says where the framework places the cycle today, and says nothing about when anything happens.Source: Smart X Capital, 18.6-year cycle framework · stage reading as of 10 Nov 2026

Peak readings have come before multi-year drawdowns in past cycles, but they don't give a date. Four down weeks and a 4% day near a peak fit a turn. They also fit a sharp pullback in a market that's still expensive. The S&P 500's CAPE ratio was about 40.7 heading into the election, the highest on record going into a midterm. I don't know which this is, and anyone who tells you with certainty is guessing.

What Graham said to do

Benjamin Graham gave investors a character to keep in mind: Mr. Market, a business partner with mood swings. Every day he offers to buy your share or sell you his, at a price that depends on his mood. You don't have to accept, and on most days you shouldn't care.

Three things follow from that.

  1. Treat the price as an offer, not a verdict. Mr. Market quoted very different prices for the same chip companies on 4, 5 and 9 November. The businesses didn't change much in five days. His mood did. Check our analysis of Nvidia here: https://smartxterminal.com/analyze/NVDA

  2. A margin of safety is about what you pay, not how far it fell. A stock 29% below its record isn't automatically cheap. If it was priced well above its value at the top, it can fall a long way and still sit above it. I'm not making a call on any of these companies' value here. That takes a full look at the business, which is a separate piece of work.

  3. Decide your sizing before the week, not during it. If you own businesses you understand, at sizes you can live with, a fortnight like this is mostly noise. If it made you want to do something drastic, that's useful information about your plan. Better to learn it now than in a real bear market.

What I'd ignore

  • The daily explanation. On 4 November the election was good news. On 5 November the same election was the reason for the worst day in 18 months. When the story flips in 24 hours, the story isn't the cause.

  • The forecast. "This is the start of the big one" and "this is just a pullback" are both guesses dressed as analysis.

  • The urgency. Nvidia's results on 17 November will bring a fresh wave of headlines. Nothing here requires a decision before then, or after.

The cycle tells us roughly where we are. Graham tells us what to do whatever it does: know what you own, know what you'd pay for it, and size it so a fortnight like this one doesn't decide for you.

Quick note: the readings and figures here are as of 10 November, when I wrote this. Markets move, so for the live picture I keep Smart X Terminal updated in real time.

To see the cycle reading next to a watchlist of the businesses you'd want to own at the right price, open Smart X Terminal.

Next week, for Smart X Insiders, I'll take on the harder question: what would have to be true for this AI selloff to mark real value, rather than the start of a longer slide.

If you find this article interesting, please let us know your thoughts down below here. We also have other similar articles on our website too.

Tools I use

Sharing with you the tools I’m using at the moment.

🧭  Why I’m Building the Smart X Capital Platform

I’m building something for investors who want to move smarter — not faster.

  • This isn’t for everyone. It’s for those who want to understand wealth through time, not tactics

A place where we’ll track these cycles together, share real-time insights, and learn how to invest with the cycle — not against it. I’ll be offering workshops, tutorials, and in-depth guides to help you build a timeless investing system that grows through every boom and bust.

📚 The Smart X Capital Platform is coming soon — a place to learn, connect, and stay ahead of every major market cycle using data, history, discipline and our community.

Because when every major cycle converges — the prepared don’t panic. They profit.

Talk soon,

Ace — Smart X Capital’s Founder

❝

Disclaimer: This newsletter is for educational and informational purposes only. It is general in nature and has been prepared without taking into account your personal objectives, financial situation, or needs. Nothing in this newsletter constitutes financial product advice, a recommendation to buy or sell any security, or a solicitation to invest. You should not rely on this content as the basis for any investment decision. Before making any financial decision, consider whether the information is appropriate for your circumstances and seek advice from a licensed financial adviser. Past performance referenced in this newsletter is not indicative of future results. All investing involves risk, including the possible loss of principal. Smart X Capital does not hold an Australian Financial Services Licence (AFSL). This publication is intended for a global audience of self-directed investors. It is not directed at Australian residents specifically. © Smart X Capital. All rights reserved.

Reply

Avatar

or to participate