AI companies are set to power most of this quarter's profits — buy proven earnings beaters before they report

Verdict: strong fundamentals, but the buy signal has never fired — wait

The idea argues that a market rising into earnings season has priced in AI-driven profit growth, and per Reuters, AI-related companies are expected to drive most third-quarter gains. The strongest evidence for the thesis is fundamental: NVDA's revenue rose 17.9% quarter over quarter to $96.2B with a 66.2% operating margin, and MSFT grew revenue 8.6% to $90.0B with free cash flow up 24.3% to $19.6B. The strongest case against is the adverse insider posture — provisional filings for the period ended September 30, 2026 show net open-market selling at all three names, roughly $569M at NVDA, $57.4M at META, and $47.7M at MSFT — plus META's own quarter showing free cash flow down 86.8% to $1.7B and net margin falling to about 26%. Critically, this is a watch-list setup, not an active signal: across 1,235 daily bars over 60 months, the entry rules never fired, and no position should be taken until one does. Thesis support is decent, fundamentals are strong but mixed at the edges, and risk quality is fair given defined 2.5% stops — but trade readiness and trigger proximity score low, because the entry needs, for example, NVDA RSI at or below 45 (currently 47.3) and price at or below its $222.40 50-day EMA (currently $229.28).

Trade now: the entry stack is still waiting on momentum

Nothing is live yet. This is a watch-list setup: the strategy buys a pullback-and-reclaim, and right now the pullback hasn't finished. For every name in the basket (META, MSFT, NVDA, QQQ), the entry needs price at or below its 50-day EMA plus a 14-day RSI reading at or below 45, with momentum turning back up. None of the four currently qualifies. Closest to a trigger is NVDA at $229.28 — about $6.88 above its 50-day EMA of $222.40, with RSI at 47.3 versus the 45 requirement. META ($718.67, about $51.61 above its $667.06 EMA; RSI 48.5) and QQQ ($750.20, about $23.08 above its $727.12 EMA; RSI 59.1) need a deeper cooling-off. MSFT is furthest away: at $535.07 with RSI at 71.5, it sits about $43.56 above its $491.51 EMA — a genuine correction, not a dip, would be needed. When an entry does fire, risk is defined mechanically: a stop at a 2.5% loss and a take-profit at a 4.9% gain — roughly 2:1 reward-to-risk — with no position larger than 25% of the book. The exit rule can also close a trade once a 60-bar holding period has passed and price closes below the lower Bollinger (20) band. What "wait" means concretely: no orders today. Track NVDA's RSI against 45 first — it is the nearest catalyst. Note that the parameter-sensitivity review did not establish a robust alternative setup, so the published thresholds stand as written.

AI Earnings Power Is Real: Why This Watch-List Setup Has a Fundamentally Loaded Bull Case

The thesis, per the Reuters report that AI-related companies will drive most third-quarter US earnings gains and the Bloomberg piece on stocks rising into earnings season, is fundamentally aligned with what the numbers show. NVIDIA's latest quarter ending July 26, 2026 delivered $96.2B in revenue, up 17.9% quarter over quarter and up 73.2% year over year — the 92nd percentile of its sector — with a 66.2% operating margin and a 63.0% net margin. Microsoft's quarter ending June 30, 2026 showed $90.0B in revenue, up 8.6% sequentially with a 45.1% operating margin (the 98.8th sector percentile) and free cash flow rebounding to $19.6B, up 24.3% from the prior quarter. If profits are concentrated in AI names, these two carry the fundamentals to justify it. META also fits the beat-prone candidate profile despite a soft quarter. Its trailing snapshot shows 23.8% year-over-year revenue growth (86th sector percentile), an 81.8% gross margin, and a 41.3% operating margin at the 97.3rd percentile of Communication Services peers. The company returned to sequential revenue growth in the quarter ending June 30, 2026 ($60.8B, up 8.0% from $56.3B), and holds $81.6B in cash and short-term investments against $58.7B of long-term debt with a 2.6 current ratio — balance-sheet room few companies have. The timing argument is the third leg. Per the CNBC screen of names with a history of earnings beats and post-report share gains, buying ahead of reports has historically skewed the odds in buyers' favor for this type of company. QQQ's look-through confirms the thesis's concentration claim: the top 10 holdings cover 45% of the fund's weight, and the covered constituents show 29.5% year-over-year revenue growth with a 28.4% net margin — earnings power that rises with the sector the market is already lifting into. One important scope note: the entry rules evaluated on real daily bars over the last 60 months (1,235 bars) did not fire — this is a watch-list setup, not an active signal. The idea waits for a pullback-to-support state (price at or below the 50-day average with momentum turning back up), so today's rising-into-earnings tape means patience is built into the design. The bull case is that when that entry eventually arrives, it would be buying fundamentally strong,…

Scores

• Conviction score breakdown: 45
• Thesis support: 65
• Trade readiness: 25
• Risk quality: 55
• Trigger proximity: 20
• Fundamentals trend: 62

Watch items

• NVDA — RSI (14)
• NVDA — Price vs 50-day EMA
• META — RSI (14)
• META — Price vs 50-day EMA
• MSFT — RSI (14)
• MSFT — Price vs 50-day EMA
• QQQ — Price vs 50-day EMA
• META — Free cash flow trend
• NVDA — Insider net open-market activity
• MSFT — Dividend ex-date