If You'd Invested $10,000 in Nvidia Stock 10 Years Ago, Here's How Much You'd Have Today

Nvidia reported fiscal 2026 Q2 adjusted EPS of $1.05 and revenue of $46.74B, beating Wall Street consensus despite a slight revenue miss in the data center division. Management indicated it still sees strong AI infrastructure demand, offsetting the disappointment. Nvidia also trades at about 41x forward earnings, with debate over valuation even as investors continue to pay for AI-linked growth.

Analysis

The main market takeaway is not that the franchise is intact — that is already priced — but that the burden of proof has shifted to acceleration. At this size, every incremental dollar of AI demand has to overcome multiple compression risk, so a modest data-center miss matters more for sentiment than for fundamentals. In practice, that means NVDA is increasingly a momentum/quality compounder rather than a clean “beat-and-raise” trade; upside now depends on whether the next leg of AI spend broadens beyond the current hyperscaler cohort.

Second-order winners are likely the less-owned infrastructure names with operating leverage to the same capex cycle: AVGO, ANET, MRVL, and parts of TSM’s AI supply chain. If investors conclude that compute is maturing while networking, interconnect, and memory remain underpenetrated, capital can rotate there without needing a pause in AI spending. The flip side is that any digestion in hyperscaler capex would hit the entire complex, but the most vulnerable names are those with the highest inventory and balance-sheet sensitivity, such as SMCI and other high-beta hardware suppliers.

The contrarian point is that a very good quarter can still be a bad setup when the valuation already discounts perfection. Over the next 1-3 months, the key catalyst is not the print itself but whether MSFT/AMZN/GOOGL confirm durable capex growth into the next budget cycle. If that confirmation stalls, NVDA’s multiple can compress even with strong absolute results; if it re-accelerates, the cleaner upside may actually be in adjacent beneficiaries that have not yet re-rated as aggressively.

Key Decisions for Investors

• Do not add aggressively to NVDA at current levels; wait for a 5-8% pullback or a confirmed re-acceleration in hyperscaler capex before increasing exposure.
• Relative-value idea: long AVGO / short NVDA for 1-3 months if NVDA remains range-bound while AVGO’s AI mix and diversification get rewarded by the market.
• Watch-list trade: buy ANET or MRVL on weakness as second-order AI beneficiaries; thesis is broader networking/interconnect spend, with a falsifier being any flattening in next-quarter cloud capex commentary.
• Avoid chasing SMCI-style high-beta hardware names until inventory turns improve; if AI spend digests, these names should underperform NVDA on both margin and multiple.