I. The Setup: A Record Year for Single-Stock Blowups

Start with one number: 24.

Our chart tracks 88 of the largest US companies that have traded since 2010. It counts days when one of them fell 10% or more while the broad market held up. From 2010 to 2023, that happened about 9 times a year. In each of the past two years it happened more than 20 times. Through Oct 23, it has happened 24 times in 2026, a new record and 2.8 times the old yearly average.

The worst of this year's drops:

None of these came on a crash day. Each was one company repricing on one quarter or one forecast.

Last week added more. On All-In, Jason Calacanis noted that Meta fell 6% after earnings amid worry about its AI spending. AT&T beat on earnings and still, per Invezz, "declined 6.23% to $21.96, making it the worst-performing stock in the S&P 500 during the session." The company added 379,000 postpaid phone subscribers, down from 432,000 the quarter before. Beating the forecast was not enough.

The index can look calm while each report carries the risk. That is the backdrop for the week.


II. Event One: The Big Spenders Report

Three of the largest cloud companies report this week. For any one of them, the cost of a disappointing report is a single bad day, and that cost has rarely been this high.

Morgan Stanley's breakdown, shared by ZeroHedge, puts S&P 500 earnings growth at 27% for the third quarter. It splits the credit three ways:

Micron alone accounts for 18.8% of the growth and Nvidia for 15.0%. A large share of the market's profit story runs through a handful of names.

Cameron Dawson of NewEdge Wealth argued on the RiskReversal Pod that the big cloud companies' return on what they invest is "diluting from about 40% to the low-to-mid 20s," and that it "could fall further if they absorb neocloud capacity." On The Compound, a host said IBM's CEO described large enterprises as "scrutinizing AI ROI and spend."

Meta showed what a soft report costs. A Compound host called its latest quarter the "first negative free cash flow quarter since 2012," meaning the company spent more cash than it brought in. The stock fell 6%. If another big spender shows costs outrunning revenue, our chart says the market will settle the question in one session.


III. Event Two: France's Rating, With a Deadline

The second event is overseas.

On All-In, Calacanis said France's 10-year yield passed 5%, its highest since 2002, while the US 10-year slipped just under 5%. He cited Polymarket odds of 91% for a French downgrade by Oct 31, a figure the source marks as unverified. Chamath Palihapitiya said France has proposed 50 billion euros of austerity, and he urged bond vigilantes to push French yields to about 8%.

David Friedberg put the US next to France. The US runs a deficit equal to 6.2% of its economy with 2.3% growth. France runs 5.4% with 0.5% growth.

A downgrade could cut two ways for American investors. Calacanis cited a WSJ view that money is fleeing European debt for Treasuries, which would help hold US borrowing costs down. Dawson is less comfortable. She calls higher rates "structural" and says yields remain in an uptrend even after pulling back from a 10-year peak of roughly 5.3%.

Our second chart shows where nervous money has already gone. Health care beat the S&P 500 by 10 points over three months and still leads by 7.1 points. Dawson says risk appetite is shifting toward defensives. History says that kind of lead usually fades. Since 1998 it happened 14 times, and health care trailed the market over the next three months in 9 of them, by 2.4 points on average. In the 4 cases that came in a rising market, the lead faded 3 times.


IV. The Case Against Our Read

The honest counterargument is that the market is doing its job, and the damage stays contained.

A host on What Are Your Thoughts? argued last week's AI selloff was healthy credit discipline rather than a demand problem. Another host, citing Yardeni, said rising Treasury yields mostly reflect better-than-expected growth, not stress. On The Compound, the hosts and Jill Schlesinger said the midterms are unlikely to move markets, and cited Wolfe as seeing at most a temporary AI selloff. Schlesinger also said the US lacks overlevered banks, so a multi-year recession seems unlikely to her.

Our own charts have limits too. A record count of single-stock drops says nothing about where the index goes. The health care base rate rests on 14 cases, and in 5 of them the lead held. One stretch saw health care beat the market by another 11.1 points. Dawson herself says chip stocks are oversold and a bounce is likely.


V. What to Watch, and What Would Change the Read

Three things this week:

What would change the read: a week of big reports with no single-day collapses, and a health care lead that keeps growing in a rising market. Either would argue the pattern of the past two years is easing. Calacanis also flagged midterm risk on November 3, the week after, which would move the conversation from earnings to politics.

Disclosure: Nova One Capital may hold positions in companies we discuss. See novaonesignal.com/disclosures.

This document is for informational purposes only and does not constitute investment advice. Nova One Capital makes no representation as to the accuracy or completeness of the information herein, and past performance is not indicative of future results. October 26, 2026.