AMD Stock Is Crushing Nvidia's in 2026. Will That Continue?

The article argues Nvidia is the better AI stock despite AMD's 144% year-to-date rally in 2026, citing Nvidia's stronger fundamentals and cheaper valuation at 21.6x forward earnings versus AMD's 71x. Nvidia's data center revenue grew 92% year over year to $75.2B in Q1, far outpacing AMD's 57% growth to $5.8B. The piece is bullish on Nvidia and skeptical of AMD's elevated expectations, but it is opinion-driven rather than new company-specific news.

Analysis

The market is pricing AMD like a structural winner before the evidence supports that rerating. That usually works until it doesn’t: when multiple expansion gets ahead of earnings revision velocity, the stock becomes vulnerable to any deceleration in order growth, gross margin cadence, or a softer-than-expected capex budget from hyperscalers. Nvidia’s lower multiple matters less as a standalone bargain call than as a signal that the better risk/reward may be in the incumbent whose earnings power is still compounding faster, not in the challenger that has already been awarded much of the future.

Second-order effects favor NVDA across the ecosystem. If customers remain performance-sensitive, Nvidia’s stronger software moat and installed base should keep pulling through higher-margin networking, platforms, and attach revenue, which can cushion any unit growth moderation. Meanwhile, AMD’s outperformance may be telling us more about positioning and narrative compression than about durable share gains; that leaves semicap equipment, memory, and supply chain names exposed to a sentiment snapback if AI spend concentrates back into the leader rather than broadening out.

The key risk to the bearish AMD / bullish NVDA framing is a genuinely material product-cycle surprise: if AMD sustains design wins and turns them into repeatable volume, the market can keep paying up for another 1-2 quarters. But that would need to show up in accelerating revenue mix, not just headline deals. If it doesn’t, the likely next catalyst is valuation normalization rather than a fundamental collapse, which is why the trade should be framed as relative, not absolute.

Contrarian view: consensus may be over-discounting Nvidia’s size and underestimating how long AI capex can remain concentrated. In the near term, leadership can stay expensive, but over 6-12 months the more likely outcome is multiple compression for AMD than a rerating higher for Nvidia from these levels. That makes the spread trade more attractive than an outright short on AMD or outright long on NVDA.

Key Decisions for Investors

• Initiate a relative-value pair: long NVDA / short AMD, sized market-neutral, for a 3-6 month horizon. Thesis: NVDA can compound earnings faster while AMD’s 71x forward multiple leaves little room for execution misses; target is 10-20% spread capture with downside protection from sector beta.
• If forced to choose outright exposure, overweight NVDA on pullbacks below the low-20s forward multiple and use a 3-6 month horizon. Risk/reward favors the incumbent with better earnings visibility and a lower expectation bar.
• Avoid initiating fresh long AMD at current levels unless you have a 1-2 quarter catalyst calendar. Best entry would be after a post-earnings reset or if the stock de-rates into a materially lower forward multiple with evidence of sustained data center acceleration.
• For more convex exposure, buy NVDA call spreads 3-6 months out rather than stock if you expect AI capex leadership to reassert itself. This captures upside from a multiple re-rating while limiting drawdown if the market continues to favor AMD in the short term.
• Set a watchpoint on next hyperscaler capex commentary: if customer spend broadens beyond the current AI leaders, AMD can retain narrative support; if not, rotate back into NVDA and consider a tactical short on AMD into strength.