Nvidia just clinched its first record high in months on an unstoppable AI boom — ride the momentum

Verdict: the AI demand story is real, but the entry hasn't earned your money yet

The demand thesis is the strongest part of this idea: revenue of $96.2B in the July quarter, up 17.9% quarter over quarter, with operating margin expanding to 66.2% and the dividend stepped up from $0.01 to $0.25 per share — all consistent with the structural story. But the freshest signals cut the other way: free cash flow fell 56% quarter over quarter to $21.4B, net margin compressed 9.4 points to 62.0%, and debt-to-equity rose from 0.04 to 0.15, reversing a multi-year deleveraging trend. The ownership file for the period ended September 30, 2026 — flagged as provisional — shows roughly $569.5M of net open-market insider selling across nine reporters, a flow worth watching rather than treating as current sentiment. Mechanically, the pullback-and-reclaim entry never triggered across 60 months and 1,236 daily bars, so the reader is watching for a first-ever trigger, not buying a recurring edge; with NVDA at $236.41 versus its 20-day average at $226.27, RSI at 70.4 and ADX at 16.8, the setup is waiting for its conditions. Verdict: wait — let the market bring the entry to you rather than chasing a record close. A dip to the $226 area that holds, with RSI crossing back above 50 and ADX above 20, is what changes the answer; a close below the second support at $220.54 kills the setup.

Trade now: NVDA pullback-and-reclaim is armed, but price has not pulled back yet

**Nothing to buy today.** NVDA closed at $236.41, about $10.14 above its 20-day moving average of $226.27, so the first piece of the pullback-and-reclaim entry — a day whose low touches that average while the close holds above it — has not happened. The 14-day RSI stands at 70.4, well above the 50 level it would need to cross back above after a dip, and the 14-day ADX at 16.8 is still below the 20 trend-strength threshold (it needs to move above it, not sit below). In short: the setup is waiting for its conditions, not failing them. If the entry does trigger, the risk plan is mechanical: the fixed stop sits at a 2.4% loss on the position and the first take-profit at a 4.9% gain — roughly 2-to-1 reward-to-risk at the fixed levels. The chart-based exits refine that: profits are taken near the first resistance level (currently $236.54, essentially at the last close, so it would reset higher after a breakout), and the position is cut if price closes below the second support level (currently $220.54). A time stop closes any position after 60 trading days regardless. **What "wait" means concretely:** let NVDA trade down to the $226 area (the 20-day average), watch whether the close holds above it, and confirm that RSI has crossed back above 50 and ADX has pushed above 20. Buying today at $236.41 is not the strategy — it is chasing an entry the rules deliberately avoid. The same logic applies to the SMH basket alternative, which closed at $631.61, 5.6% below its range high, with its own 20-day average at $595.46 and ADX at 19.0, just shy of its threshold. Note that the parameter-sensitivity work on this rule set did not complete within its budget, so no robust nearby setup was established and no adjustment was recommended — the published rules stand as written.

Why the Bull Case Still Has Support

The bullish thesis rests on a demand story that the fundamentals actually corroborate. In the quarter ended July 26, 2026, Nvidia reported revenue of $96.2B, up 17.9% from $81.6B the prior quarter, and net income rose to $59.7B. Zoom out and the trajectory is staggering: full-year revenue through January 2026 was $215.9B, up 73.2% year over year — a growth rate that sits in the top 8% of nearly 900 Information Technology peers, per the supplied peer percentiles. This is not a momentum pop on hope; it is a company compounding at triple-digit-billion scale. Profitability supports the 'structural, not cyclical' framing the idea leans on. Gross margin held essentially flat at 75.0% in the latest quarter (versus 74.9% prior), and operating margin actually expanded to 66.2% from 65.6% — Nvidia sits at the 99th percentile for operating margin among its sector peers. Even after net margin compressed to 62.0% from 71.5%, ROE of 26.1% in the quarter remains elite. The most recent reported full-year quarter (ended January 25, 2026) showed $42.96B of net income on $68.1B of revenue, with free cash flow of $34.9B at the 99.5th percentile among 833 peers. The balance sheet and shareholder-return posture strengthen the case rather than undermine it. Debt-to-equity of just 0.146 at fiscal year-end 2026 means the AI capex boom is being absorbed without meaningful leverage. Meanwhile the dividend tells its own story: after paying roughly $0.04 per share over the prior twelve months, the company stepped up quarterly payouts from $0.01 to $0.25 per share — an annual run-rate of $0.51, a signal of management confidence in cash generation. Share count also ticked down from 24.2B to 24.1B, indicating ongoing buybacks rather than dilution. On the mechanics of the idea itself: the pullback-and-reclaim entry evaluated on daily bars across NVDA and SMH did not trigger in the 60-, 24-, or 12-month windows (1,236 bars evaluated), so this is a watch-list…

Scores

• Conviction score breakdown: 47
• Thesis support: 65
• Trade readiness: 40
• Risk quality: 45
• Trigger proximity: 30
• Fundamentals trend: 55

Watch items

• NVDA — Low vs 20-day EMA
• NVDA — RSI (14)
• NVDA — ADX (14)
• NVDA — Close vs second support
• SMH — Low vs 20-day EMA
• SMH — ADX (14)
• NVDA — Insider net open-market flow
• NVDA — Quarterly revenue growth