When a company reports spectacular results and still falls, it tells you the money was positioned for more — and that money tends to move to the next-best name rather than leave the sector. Broadcom's 221% AI revenue surge being punished for soft guidance

Verdict: rotation thesis is real, but wait for the crossover

The idea's core rotation claim is well grounded: all three companies posted accelerating P&Ls, with Nvidia's quarter ended 2026-07-26 showing $96.2B of revenue up 17.9%, Broadcom's May-quarter revenue up 14.9% to $22.2B, and AMD's June-quarter net income up 66.1% to $2.3B — consistent with AI chip demand staying intact while expectations rotate between names. The strongest argument against is that insiders at all three companies were net open-market sellers in the period ended June 30, 2026 — roughly $375M at Nvidia, $283M at Broadcom, and $153M at AMD (a delayed filing, not a current reading) — and Nvidia's own free cash flow fell 56.0% quarter over quarter to $21.4B while debt-to-equity rose to 0.14. On readiness, this is a watch-list setup, not a signal: the entry needs price above the 20-day EMA, RSI above 50, and a fresh MACD cross above the signal line, and Nvidia and AMD are one crossover away while Broadcom first needs to repair its trend. No robust parameter setup was established because the optimization exceeded its time budget, so the original rules stand as written. A confirmed MACD cross on NVDA or AMD at a daily close would flip this to actionable; a fresh filing showing Nvidia's cash conversion recovering would flip it decisively.

Trade now: three names, one live read — wait for the crossover

The setup is a stacked trend-confirmation entry on daily bars: close above the 20-day EMA, RSI (14) above 50, and a MACD line crossing above its signal line, on NVDA, AVGO, or AMD (priority in that order). At the latest close, NVDA at $230.36 is the closest candidate: it sits $10.92 above its 20-day EMA of $219.44, with RSI at 65.1, well clear of the 50 threshold. AMD at $477.57 also clears its EMA of $475.43, with RSI at 54.4. AVGO at $357.89 is furthest away — price is $15.15 below its EMA of $373.05, and RSI of 40.1 sits roughly 9.9 points below the 50 mark. What is missing everywhere is the final condition: a fresh MACD cross above the signal line. NVDA's MACD is already positive at 3.77, so an entry there would require a pullback and a fresh recross. AMD and AVGO both have negative MACD readings of -8.55 and -9.74, so their crosses are the more plausible near-term triggers if the uptrend resumes. The research author requested bounded optimization because the strict stacking produced no historical entries across three windows, but no robust parameter setup was established, so the original conditions stand as written. This is a watch-list setup, not an active signal. Risk framing once triggered is mechanical: a hard stop at a 2.5% loss on the position and a take-profit at 5.0%, plus a signal exit if the close falls back below the 20-day EMA. Position size is capped at 25% of the book with 2.5% fixed risk per position. "Wait" here means: do not pre-position. Check at each daily close whether a fresh MACD cross has printed on a name that already has price above the EMA and RSI above 50 — on current readings, NVDA and AMD are one crossover away; AVGO needs a trend repair first.

The rotation logic has real fundamental fuel behind it

The idea's core claim — that capital rotates within AI chips rather than leaving the sector — matches what the fundamentals show across all three names. Nvidia's quarter ended 2026-07-26 posted $96.2B of revenue, up 17.9% from the prior quarter, with a 75.0% gross margin and $59.7B of net income. Broadcom's quarter ended 2026-05-03 showed $22.2B of revenue, up 14.9% sequentially, with net income of $9.3B and a 48.6% operating margin. AMD's quarter ended 2026-06-27 delivered $11.5B of revenue, up 12.5% sequentially, with net income jumping 66.1% to $2.3B. These are three accelerating P&Ls, not one strong name propping up weak peers — consistent with the thesis that AI chip demand itself is intact. The cited Yahoo Finance coverage of September 3, 2026 supports the rotation framing directly: Broadcom fell about 6% as soft guidance overshadowed a 221% AI revenue surge, while Nvidia inched higher and AMD slipped. A company tripling AI chip sales and being punished for guidance, rather than demand, is precisely the setup the idea describes — expectations getting ahead of one name, not the business breaking. On quality of the destinations: Nvidia's free cash flow for the full year ended 2026-01-25 was $96.7B, with a 65.0% operating margin in the latest full quarter — among the very top of the Information Technology sector per the peer percentiles (99.5th percentile on free cash flow, 99.0th on operating margin). Broadcom's free cash flow hit $10.3B in the quarter ended 2026-05-03, up 28.1%, while cutting its debt-to-equity ratio to 0.71 from 0.80. AMD's free cash flow of $1.6B still ranks in the 98.6th sector percentile, and its debt-to-equity fell to 0.035. Rotation capital is moving toward names that can fund the AI buildout from their own cash flow. The setup is a watch-list entry, not an active signal: the trend-confirmation rules (price above the 20-day average, RSI above 50, MACD crossover) did not trigger in the last 60 months of evaluated bars across 1,236 daily bars, in any of the three windows tried. That is…

Scores

• Conviction score breakdown: 58
• Thesis support: 74
• Trade readiness: 35
• Risk quality: 48
• Trigger proximity: 60
• Fundamentals trend: 72

Watch items

• NVDA — MACD (12,26,9) fresh cross above signal line
• AMD — MACD (12,26,9) fresh cross above signal line
• AVGO — Price vs 20-day EMA
• AVGO — RSI (14)
• NVDA — Close below 20-day EMA (signal exit)
• NVDA — Debt to equity
• NVDA — Insider net open-market activity
• AVGO — Next quarterly ex-dividend date
• AMD — Free cash flow