Why Nvidia's Third Consecutive Win Against S&P 500 Could Extend Into 2026

The Rubin Chip Revolution Is Just Getting Started

When 2025 wrapped up, Nvidia had already cemented its dominance for the third consecutive year, delivering 38.9% returns compared to the S&P 500’s 16.4% gain. But here’s the thing: the real catalyst for the 3rd leg of Nvidia’s outperformance cycle may just be hitting the market.

Back in October 2025, Nvidia projected it would book $500 billion in combined Blackwell and Rubin orders, with revenue realization stretching into fiscal 2026. That forecast? It’s already outdated. By January, the chipmaker announced that orders continue to accelerate beyond prior guidance, and the Rubin architecture just entered full production, with customer deliveries ramping up in H2 2026.

Why Rubin Changes the Game

What makes Rubin such a big deal isn’t just incremental improvement—it’s a meaningful leap in efficiency. The new chip generation achieves up to 90% reduction in AI inference token costs while cutting GPU requirements for model training by as much as 75%.

These aren’t minor optimizations. For hyperscalers already spending hundreds of billions on AI infrastructure, this translates to substantially lower data center operating expenses. Nvidia’s ability to command premium pricing becomes justified because customers actually save money by using superior hardware.

The technical foundation here matters: Rubin packs more transistors per GPU and employs “extreme codesign”—essentially vertical integration across the entire stack from physical infrastructure to software algorithms. It’s the kind of engineering advantage that’s difficult for competitors to replicate quickly.

The Earnings Estimate Surprise

Analysts have already started recalibrating their forecasts. Consensus estimates for Nvidia’s fiscal 2027 EPS jumped to $7.60, up from $5.76 just six months prior. Fiscal 2026 estimates rose to $4.69 from $4.29. That’s not gradual improvement—that’s recognition of Rubin acceleration pulling forward earnings recognition.

Since Nvidia’s fiscal year ends in late January, the Rubin revenue windfall should become material in the second half of fiscal 2027, which translates to calendar year mid-2027 for most investors tracking real-time results.

Is the Valuation Still Reasonable?

With a forward P/E ratio around 39x, skeptics might question whether Nvidia’s valuation leaves room to run. The counterargument: at 39x multiples, you’re paying for a company that combines industry-leading margins with explosive growth tied to structural AI infrastructure buildout.

The Rubin order surge suggests demand for AI chips remains insatiable. Customers aren’t rationing orders—they’re competing to secure supply. That pricing power and demand visibility justify holding premium valuation, at least until capital expenditure cycles show signs of deterioration.

The Real Risk Worth Monitoring

The legitimate threat to this thesis isn’t innovation or competition—it’s customer saturation. At some point, hyperscalers will need to convert all these capital expenditures into actual business earnings. The capex-to-EBITDA conversion will eventually require a slowdown in spending growth.

Until that inflection arrives, there’s little evidence suggesting the AI infrastructure spending wave is cresting. Rubin’s early market reception and accelerating bookings suggest the 3rd leg of Nvidia’s outperformance could have real staying power through 2026 and beyond.

The window to capture this next phase of chip leadership may be narrowing, but the momentum indicators—both in orders and analyst revisions—continue pointing upward.