2026-08-19

Nvidia -2%, AMD -5%, Broadcom -3%, Meta -3% - Yet the Equal-Weight S&P 500 Rose 0.2% as the 30-Year Yield Hit a 19-Year High

Written & reviewed by the TradeSmrt editorial team · 2026-08-19

In this article

What Happened
Why AMD Fell Twice as Hard as Nvidia on the Same News
What to Take Away From This
FAQ
Sources

What Happened

On Tuesday, August 18, the AI chip and megacap tech names that have led this year's rally sold off together. Nvidia (NASDAQ: NVDA) fell 2%. AMD (NASDAQ: AMD) dropped roughly 5%, touching as much as 5.5% intraday. Broadcom (NASDAQ: AVGO) and Meta Platforms (NASDAQ: META) each slid about 3%. The VanEck Semiconductor ETF (SMH) tumbled 4.3%, and the Philadelphia Semiconductor Index (SOX) fell by a similar margin. The damage spread to the broader tape: the Nasdaq Composite closed down 1.33% at 26,289.71, its lowest level in roughly two weeks, while the S&P 500 fell 0.69% to 7,691.76, its third straight losing session. The Dow Jones Industrial Average held up comparatively well, slipping just 0.22% (116.38 points) to 53,343.40.

But one detail from that same session stands out. The equal-weight version of the S&P 500 - which holds all 500 constituents at roughly the same size rather than weighting by market capitalization, tracked by the RSP ETF - actually closed up 0.2% that day. That's the mirror opposite of the Invesco QQQ Trust, which tracks the megacap-heavy Nasdaq 100 and fell 1.4%. In other words, this wasn't a broad market rout where everything sold off together - it was a concentrated selloff in roughly the ten largest AI and semiconductor names. In the standard, cap-weighted S&P 500, the top 10 stocks now account for more than a third of the index's total market value, up from around 18% a decade ago. That concentration means a headline like "the index fell for a third straight day" can obscure the fact that the large majority of the 500 underlying companies actually held steady or even rose that same day.

A second detail worth noting is the gap between Nvidia and AMD, which were hit by the exact same macro trigger yet moved by very different magnitudes - Nvidia down 2%, AMD down more than twice that at roughly 5%. Adding to the day's unease, it also emerged that billionaire investor Dan Loeb's hedge fund Third Point had disclosed, in its Q2 13F filing covering positions as of June 30, that it had fully exited Nvidia, Meta, Broadcom, KLA, and Lam Research - fueling a narrative that "smart money" was stepping away from the AI trade's biggest names.

Why AMD Fell Twice as Hard as Nvidia on the Same News

The proximate driver was the bond market. The 30-year US Treasury yield climbed to roughly 5.32-5.33% intraday, its highest level since 2007 - about 19 years. Rising long-term yields hit hardest wherever valuations depend on earnings growth many years out, because the discount rate used to convert those future profits into today's dollars rises along with the yield. AI chip and megacap tech stocks are unusually exposed to this channel, since their valuations largely rest on years of continued data-center buildout still to come.

That explanation alone, though, doesn't account for why Nvidia and AMD reacted so differently to an identical rate shock. The real answer lies in the two companies' valuation and balance-sheet differences. By several valuation measures, AMD was trading roughly 70%-plus above estimated fair value heading into the selloff - a premium that only makes sense if the company executes close to flawlessly on the AI opportunity. The richer that premium, the more sensitive a stock naturally becomes to a rising discount rate. On top of that, AMD had recently priced a $4.75 billion bond offering, more than triple the size of its previous debt sale, raised to fund AI-platform acquisitions and infrastructure expansion. That decision raised AMD's leverage right as rates were climbing, meaning the company was hit through both channels at once - a higher discount rate on its future earnings and a higher cost of servicing its now-larger debt load. Nvidia, by contrast, carries a comparatively stronger balance sheet and less incremental debt exposure, which helps explain its smaller decline.

A Jefferies note published around AMD's "Advancing AI 2026" event a day earlier added another layer. Jefferies suggested the AI and server roadmap AMD unveiled could put the company ahead of Nvidia in some areas - which, read one way, is bullish competitive news. But for a stock already facing valuation-premium scrutiny, a signal that AMD needs to keep pouring capital into an intensifying arms race with Nvidia was apparently read as a reason to sell rather than buy, at least on this particular day.

Third Point's 13F disclosure deserves a caveat too. A 13F is filed up to 45 days after quarter-end, so the positions revealed this month are already a nearly two-month-old snapshot as of June 30. The filing doesn't reveal exactly when Third Point sold, or whether it has since rebuilt any of those positions. Even so, a headline reading "prominent hedge fund exits every major AI stock" was enough to add fresh selling pressure to a market already nervous about valuations.

What to Take Away From This

A falling index headline doesn't mean the whole market is falling. The S&P 500's third straight losing day sounds broad, but the equal-weight version of the same index rose that day. Always check whether a decline is concentrated in a handful of megacaps or spread across the market before drawing conclusions.
Cap-weighted indexes and ETFs can already be far more concentrated than they appear. With the top 10 S&P 500 stocks now worth more than a third of the index, buying "the index" is a much more concentrated bet than it may look. Investors seeking real diversification should understand how equal-weight products differ from standard cap-weighted ones.
The same headline can hit two stocks very differently depending on leverage and valuation premium. AMD and Nvidia faced an identical rate shock, but AMD's steeper valuation premium and its recent debt offering made the decline sharper. Sector-wide news still needs to be checked against each company's individual balance sheet.
A 13F filing is a stale, two-month-old snapshot, not a live trade alert. A famous hedge fund's exit can feel like a strong signal, but the gap between the filing date and the actual trade date means it should be weighed carefully, not acted on reflexively.

FAQ

What's the actual difference between the equal-weight and standard S&P 500?

The standard S&P 500 (tracked by SPY) weights each company by its market capitalization, so giants like Apple, Microsoft, and Nvidia drive a disproportionate share of the index's movement. The equal-weight version (tracked by RSP) holds all 500 constituents at roughly 0.2% each and rebalances quarterly, which sharply reduces how much any single megacap can sway the index. RSP has outpaced SPY by roughly 5 percentage points in 2026 so far, a gap widely attributed to this reduced exposure to megacap concentration risk.

Does Third Point selling Nvidia and Meta mean I should sell too?

Not necessarily. A 13F is filed up to 45 days after quarter-end, so this month's disclosure reflects positions as of June 30 - it doesn't show exactly when Third Point sold or whether it has bought back in since. One hedge fund's call isn't automatically correct, and every investor's time horizon and risk tolerance differ from a hedge fund's.

Does AMD falling more than Nvidia mean AMD's fundamentals are weaker?

Not necessarily. This was, if anything, a day when analysts flagged AMD's AI and server roadmap as increasingly competitive with Nvidia's. But capturing that opportunity requires heavy capital spending, and AMD's recent large bond offering - combined with its already-rich valuation - made the stock more sensitive to rising rates. A bigger one-day decline doesn't automatically translate into worse underlying fundamentals.

Sources

This article is an original synthesis and analysis based on the reporting below, not a reproduction of the original articles. Please check the source articles directly for the most current figures and details.

⚠️ This article is for informational purposes only and is not investment advice. Market conditions change constantly - always verify the latest data before making investment decisions.