Nvidia's 4.6% Drop Meets a Bull Case Pointed at 2028
Nvidia closed at 217.55, down 4.58%, on Aug. 28. A bull case published days later argues the stock's multiple is modest by its own history and won't look stretched by 2028.
A Nasdaq-published bull case argues Nvidia's valuation is modest against its own historical levels given its growth rate, and that the stock will not look overvalued by 2028 — published after NVDA closed at 217.55, down 4.58% on the day, on Friday, Aug. 28, 2026.
Nvidia (NASDAQ: NVDA) ended the week on the back foot, closing at 217.55, down 4.58% from the prior close of 227.98 and near the bottom of a session that ranged from 216.81 to 229.26. That was the tape into which a fresh bull case landed: an argument, published by Nasdaq Markets, that Nvidia only looks expensive, that its valuation is relatively modest measured against its own historical levels given the pace of its growth, and that by 2028 today's price will not look like an overpayment.
It is a familiar shape of argument for a company that has spent three years being called too expensive on the way up. What makes it worth taking apart is the mechanism it rests on — not whether Nvidia is cheap in absolute terms, which almost nobody claims, but whether earnings growth is arriving fast enough to make the multiple shrink without the share price having to fall.
The argument is about the denominator, not the price
Every valuation multiple has two moving parts. A stock can get cheaper because the price drops or because earnings rise. The 2028 case is entirely a claim about the second: that Nvidia's profit base compounds quickly enough that a price paid today is absorbed by forward earnings within a couple of years.
That framing has an important consequence for anyone weighing it. It converts a valuation debate into an execution debate. If the earnings arrive, the multiple takes care of itself and the entry price becomes a rounding error. If they slip — because hyperscaler capital spending pauses, because customers shift to in-house silicon, because supply or export policy bites — the multiple does not compress gently. It re-rates, and the price does the work instead.
The summary's specific comparison is against Nvidia's own historical levels rather than against the market or the semiconductor group. That is a narrower and more defensible test than a cross-sector one. A company whose growth rate has changed character over the past few years is genuinely hard to compare with a mature chipmaker. But it also means the conclusion is only as good as the historical window chosen, and readers should treat "modest versus its own history" as a relative statement, not an absolute one.
Friday's close sits inside a soft week for the megacaps
Nvidia's decline was not a solo event, though it was a much larger one than the indexes registered. As of the last trade on Friday, Aug. 28, 2026 at 20:00 GMT, the Nasdaq 100 tracker (QQQ) closed at $716.43, down 0.65% from $721.11, with a day range of $715.09 to $724.13. The S&P 500 tracker (SPY) closed at $769.35, down 0.23% from $771.10. The Dow tracker (DIA) was effectively flat at $535.06, off 0.03%.
So the tech-heavy benchmark fell roughly three times as hard as the broad market on the day, and Nvidia fell many multiples of that. All three index trackers finished at or near the low end of their daily ranges, which suggests selling into the close rather than an early scare that recovered. For a stock of Nvidia's index weight, a 4.58% drawdown is itself a meaningful contributor to the Nasdaq 100's move — the causation runs at least partly from the stock to the index, not the other way around.
Markets are closed; these are the most recent prints, not live quotes. Anyone reading the bull case should note that it was written into weakness, which is the usual habitat of long-horizon valuation arguments. Multi-year cases get published when the near-term tape is ugly, because that is when the gap between price and thesis is widest.
What a 2028 target date actually commits the bull to
Setting 2028 as the year the valuation "catches up" is a specific and testable commitment, and it is worth being clear about what it does and does not require.
- Sustained growth, not a single blowout year. A two-to-three-year runway means the earnings base has to keep compounding across multiple fiscal years, through at least one product cycle transition.
- Margin durability. Revenue growth that arrives at lower gross margin does far less for a multiple than revenue growth at current margin. Competitive pricing pressure would show up here first.
- Customer concentration holding up. The demand behind the thesis is capital spending decisions made by a small number of very large buyers. Those decisions are reviewed annually, not contractually locked for three years.
- No multiple collapse independent of earnings. Even if profits triple, the stock can disappoint if the market decides to pay a lower multiple for the same growth — which is what happens when investors start treating a growth story as a cyclical one.
Setting 2028 as the year the valuation "catches up" is a specific and testable commitment, and it is worth being clear about what it does and does not require.
None of those are predictions. They are the conditions the argument quietly assumes, and they are the items to check off as quarters print.
How to test the thesis rather than take it on faith
The honest version of this debate does not resolve in a headline. It resolves in the data flow. A few things are worth tracking against the claim.
First, the direction of the multiple itself. If the bull case is right, Nvidia's forward multiple should be drifting down over the coming quarters even in periods when the share price is flat or up, because earnings estimates are climbing faster than the price. A multiple that stays static while the price rises is the market simply paying more for the same growth — a different and less comfortable outcome.
Second, whether the growth is being funded or bought. Data-center demand rests on the capital budgets of a handful of very large customers. Watch what those buyers say about spending trajectory, and watch how much of Nvidia's own growth is coming from customers who are also building alternatives in-house.
Third, the shape of drawdowns. Friday's 4.58% single-day fall is a reminder of the volatility attached to a position of this size. Investors who accept a three-year thesis are, in practice, agreeing to sit through repeated moves of that magnitude in both directions. The valuation argument says nothing about the path — only about the destination — and the path is where most positions get abandoned.
The lead's claim is modest in one important respect: it does not assert Nvidia is cheap. It asserts the stock is less expensive than its reputation implies once growth is set against price, and that time is working in the buyer's favor. That is a coherent position. It is also one that hands the entire burden of proof to the next several earnings reports.
Key facts
- NVDA last close: 217.55, down 4.58% (Fri, Aug 28, 2026, 20:00 GMT)
- Prior close / day range: 227.98; range 216.81–229.26
- Nasdaq 100 (QQQ): $716.43, down 0.65% on the day
- Thesis horizon: Bull case argues stock won't be overvalued by 2028
Frequently asked questions
What is the bull case for Nvidia being made here?
The argument, published by Nasdaq Markets, is that Nvidia only appears expensive. Measured against its own historical valuation levels and set against its extraordinary growth rate, the multiple is relatively modest — and by 2028, the piece contends, today's share price will not look overvalued.
Where did Nvidia stock close?
Nvidia last traded at 217.55 as of Friday, Aug. 28, 2026 at 20:00 GMT, down 4.58% from the prior close of 227.98. The session range was 216.81 to 229.26, meaning the stock finished near the low of the day. Markets were closed at the time of writing.
How did the broader market perform that day?
The S&P 500 tracker SPY closed at $769.35, down 0.23%. The Nasdaq 100 tracker QQQ closed at $716.43, down 0.65%. The Dow tracker DIA was nearly unchanged at $535.06, off 0.03%. The tech-heavy Nasdaq 100 fell hardest of the three.
Why does a multiple shrink without the price falling?
A valuation multiple divides price by earnings. If earnings grow while the price stays put, the multiple falls on its own. That is the entire mechanism behind the 2028 argument: growth in the earnings denominator, rather than a decline in price, is what makes today's entry look reasonable later.
What would break the 2028 thesis?
Four things: earnings growth slowing before 2028, gross margins eroding under competitive pricing, a small group of very large customers trimming capital spending or shifting to in-house chips, or the market simply deciding to pay a lower multiple for the same growth rate — treating Nvidia as cyclical rather than secular.
Does the argument say Nvidia is cheap?
No. The claim is relative, not absolute. It says Nvidia is less expensive than its reputation suggests when growth is weighed against price, and that its multiple is modest versus its own history. That is different from calling the stock cheap on conventional measures, which the summary does not do.
Sources
CC0 1.0 — source
