Chip rout spooks investors, but AI demand just proved itself — buy the dip on ASML and Nvidia

Verdict: The AI dip-buy story is credible, but the entry hasn't fired — wait for confirmation

The thesis has real fundamental support — ASML raised its sales forecast for the second time in 2026 on AI chip demand (per the July 15 CNBC report), fiscal-2025 revenue rose 15.6% to $32.7B with free cash flow up 21.8% to $11.1B, and NVDA's July-quarter revenue grew 17.9% to $96.2B — and the rule set has a completed backtest of +124.1% over five years on four trades. The strongest case against is ownership and cash-flow tension: NVDA's June 30, 2026 filings show roughly $565.4 million of net open-market insider selling, and its latest quarter saw free cash flow fall 56% to $21.4B while net margin compressed to 62.0% from 71.5%. Meanwhile the entry itself is not live — ASML closed at $1,729.52 with its 10-day average ($1,724.76) still below the 20-day ($1,732.75) and price $24.37 below the 50-day average ($1,753.89); NVDA is closer but needs a close above $224.33, and the QQQ filter cannot be confirmed from the data supplied. The completed backtest is thin — four trades in 60 months at a 50% win rate, with exits filled on daily bars only — and sensitivity testing produced no recommendation, so the published parameters are the only tested version. What would flip the verdict is confirmation: an ASML close above roughly $1,733 (the crossover) plus a reclaim of $1,753.89, and an NVDA close above $224.33 with the QQQ filter confirmed, would make this actionable; a close below $1,701.70 on ASML would kill it.

Trade now: two setups, one live, one eight points away

ASML closed at $1,729.52 and is the laggard of the pair: its 10-day average ($1,724.76) still sits below the 20-day ($1,732.75), so the crossover is roughly $8 away from triggering, and price is $24.37 below the 50-day average ($1,753.89) — that momentum filter is not met either. NVDA is much closer: at $224.17, its 10-day average ($224.33) is already above the 20-day ($221.02) and price is only $0.16 below the 10-day line, with the 50-day condition comfortably met ($12.08 above $212.09). The broader Nasdaq-100 filter needs QQQ above its own 50-day average, which cannot be confirmed from the live data supplied here, so that leg remains the wildcard. The strategy's own exits define the risk math: a 2.4% stop and a 4.8% take-profit, with a 30-day maximum hold. In price terms, an ASML entry near $1,729.52 implies a stop near $1,688 and a target near $1,813; NVDA near $224.17 implies a stop near $219 and a target near $235 — roughly 2-to-1 reward to risk on each leg. The completed backtest on this exact rule set supports the waiting game: over five years on ASML it returned +124.1% across 4 trades (50% win rate, 15.2% max drawdown), and over the last 12 months +7.6% across 3 trades (66.7% win rate, 5.9% drawdown). 'Wait' means concretely: do not pre-position. For ASML you need the 10-day average to cross above the 20-day, price to reclaim the $1,753.89 50-day line, and QQQ to hold its 50-day average. For NVDA, a close above $224.33 plus a confirmed QQQ filter is the practical trigger set. The idea argues AI demand proved itself after the rout — ASML's fiscal-2025 revenue rose 15.6% to $32.7B and NVDA's July-quarter revenue rose 17.9% to $96.2B — but the entry rules exist precisely to avoid paying for that story before momentum confirms it.

The fundamentals kept improving right through the panic

The idea's core claim — that AI demand proved itself immediately after the July 2026 chip rout — is directly supported by ASML's numbers. For fiscal 2025, ASML grew revenue 15.6% to $32.7B, lifted net income 26.9% to $9.6B, and expanded gross margin to 52.8% from 51.3%. Per the CNBC report dated July 15, 2026, ASML raised its sales forecast for the second time that year specifically on strong AI chip demand — the exact fundamental confirmation the thesis leans on. On peer comparison, ASML sits in the 99.9th percentile for free cash flow among industrial-sector peers, with $11.1B of free cash flow and free cash flow up 21.8% year over year. NVIDIA's supply-side story is even more extreme. Revenue for the fiscal year ended January 25, 2026 hit $215.9B, up 73.2% year over year, with a 75.0% gross margin in the latest quarter — the 79th percentile among technology peers — and a 65.0% operating margin at the 98.9th percentile. The idea's thesis also cites NVIDIA brushing off Kyber delay rumors to hold a $5 trillion market cap (per the July 13 Yahoo Finance report), and a potential China reopening flagged by Yahoo Finance on July 15 as possible upside surprise. Fundamentals, at least through the latest filings, are consistent with demand still outrunning supply. The completed backtest supports the entry logic itself. Over a 60-month daily window on the ASML pair, the strategy returned 124.1% across four trades with a 50% win rate and a maximum drawdown of 15.2%. On the shorter 12-month window, the same rules returned 7.6% over three trades with a 66.7% win rate and only a 5.9% maximum drawdown. That pattern — small trade counts but positive expectancy in both short and long windows — is what you want to see from a momentum-recovery rule set that waits for the 10-day EMA to cross…

Scores

• Conviction score breakdown: 58
• Thesis support: 75
• Trade readiness: 40
• Risk quality: 50
• Backtest evidence: 55
• Fundamentals trend: 70

Watch items

• ASML — ASML 10-day EMA vs 20-day EMA crossover
• ASML — ASML close vs 50-day SMA
• ASML — ASML close vs second support level
• NVDA — NVDA close vs 10-day EMA
• NVDA — NVDA close vs second support level
• NVDA — NVDA insider net open-market activity
• QQQ — QQQ close vs its 50-day SMA
• NVDA — NVDA quarterly XBRL fundamentals