Nvidia Stock Has Underperformed the Semiconductor Sector in 2026. Will It Skyrocket in the Second Half?

Nvidia's Vera Rubin processors are expected to ship this fall and could materially lift revenue and earnings, with management citing a $1 trillion order pipeline for Blackwell and Vera Rubin across 2026-2027. The article highlights projected fiscal 2027 EPS growth of 88% to $8.96 and fiscal 2026 revenue growth of 82% to $392 billion, alongside a potential 25% price increase for Rubin systems. At 29x earnings versus 75x for the iShares Semiconductor ETF, the stock is presented as undervalued with potential upside to $358, or 85%.

Analysis

The setup is less about a single product cycle and more about a monetization inflection: when the installed base moves to a materially higher power-efficiency node, Nvidia can defend pricing even if hyperscalers push harder on procurement discipline. That matters because the stock’s current discounting appears to assume either slower unit growth or margin compression; neither is the base case if the new platform becomes the default training and inference standard over the next 2-3 quarters.

Second-order winners are the infrastructure layers that ride the capex wave without bearing the same product-qualification risk: optical interconnect, networking, advanced packaging, and foundry-adjacent capacity providers. The risk is that the first-order enthusiasm around launch timing underestimates a digestion period after initial deployments; even a strong product can see order lulls if customers pause to benchmark, integrate software, and wait for rack-level validation, which can create 6-12 week volatility in sentiment.

The contrarian angle is that the market may be correctly discounting the headline growth but still underpricing the earnings durability. If pricing steps up while mix shifts toward higher-system content, earnings can compound faster than unit shipments, making consensus margin estimates too conservative. The main bear case is not product failure but execution friction: any slip in ramp, yield, or customer deployment cadence would hit a crowded AI leadership trade and could compress the multiple before fundamentals catch up.

Key Decisions for Investors

• Go long NVDA on pullbacks over the next 2-6 weeks; target a 3-6 month move as Rubin shipments convert backlog into revenue. Risk/reward favors buying weakness because the downside is likely sentiment-driven while upside can re-rate on margin surprise.
• Pair trade: long NVDA / short SMH or the weakest AI-infrastructure beneficiary basket if you want to isolate company-specific execution upside versus broader semiconductor beta over the next quarter.
• Buy a call spread in NVDA with 3-6 month tenor to express upside from launch and pricing power while limiting premium burn if the market stalls during the initial ramp.
• Overweight optical/networking and advanced-packaging proxies as secondary beneficiaries for the next 2-4 quarters; these tend to lag the headline name but capture the capex spillover with less valuation risk.
• If Rubin launch commentary turns to yield or deployment delays, reduce NVDA exposure quickly: that would likely be a 1-2 week de-risking event for crowded AI longs rather than a multi-month thesis break.