Forget Nvidia's Stock Price. This Is the Number That Actually Matters.

Nvidia reported $68.1 billion in fiscal Q4 2026 revenue, up 73% year over year, with data center revenue of $62.3 billion rising 75%. The article argues the stock looks inexpensive at a 23.9 forward P/E versus a three-year mean of 79, while highlighting a potential $700 billion in customer capex that could support further AI demand. Jensen Huang also projected Nvidia could reach $1 trillion in AI revenue by calendar 2027.

Analysis

The key second-order read-through is that NVDA is no longer a simple “sell GPUs, grow revenue” story; it has become a derivative of hyperscaler budget discipline. As long as the largest buyers keep converting AI ambition into capex, NVDA’s earnings power remains resilient even if unit growth moderates, but the stock’s multiple is now highly sensitive to any sign that those budgets are being deferred rather than reduced. That makes customer guidance the real catalyst calendar over the next 1-3 quarters, not Nvidia’s own product cycle.

The market appears to be pricing NVDA as if peak growth is already behind it, which creates a setup where incremental evidence of re-acceleration can re-rate the multiple faster than fundamentals change. The asymmetry is that the downside from a single miss is cushioned by the breadth of end-demand, while the upside from sustained capex flow is amplified by operating leverage and investor positioning that is still underexposed relative to NVDA’s earnings trajectory. In other words, the stock may be “cheap” only if capex stays sticky; if it does, cheap can become very cheap in hindsight.

The more interesting risk is not chip competition in the near term, but capex mix-shift. If hyperscalers redirect spend toward data-center buildout, networking, or inference optimization, NVDA can still win, but the revenue capture rate may fall versus headline capex growth. A longer-dated risk is that customers start demanding better payback discipline on AI infrastructure, which could slow order visibility even while AI demand remains strong.

Contrarianly, the consensus is probably underestimating how fast this can turn from a narrative-stock into an earnings compounding story again if 2026 guidance from the big four remains elevated. The market is treating the three-year valuation mean as a ceiling, but if capex holds and Rubin ramps smoothly, the right comparison is not historical P/E averages but the earnings duration of a platform monopolist with expanding installed base economics.

Key Decisions for Investors

• Maintain/overweight NVDA into the next hyperscaler earnings cycle; best risk/reward is a 3-6 month horizon where any upward capex revisions can expand multiple and EPS simultaneously.
• Buy NVDA call spreads 6-9 months out, financed with out-of-the-money call sale; express upside to renewed capex confidence while limiting premium burn if AI budgets flatten.
• Pair trade: long NVDA / short an AI infrastructure basket with lower earnings torque (e.g., MSFT or GOOGL on a relative basis) if you expect spend to remain strong but market still underprices NVDA-specific leverage over the next 1-2 quarters.
• Set a hard risk trigger on the next round of customer capex guidance: if combined guidance from AMZN/MSFT/META/GOOGL rolls over meaningfully, reduce NVDA exposure by 25-35% as order visibility will likely compress before revenue does.
• If looking for lower-beta exposure to the same theme, prefer NVDA over INTC for a 12-month view; INTC is more likely to suffer from capex being concentrated in AI accelerators rather than broad semis refresh.