The S&P 500 is up more than 12% year-to-date after nine months, and it got there climbing a classic wall of worry: a 10-year yield at 5%, crude near $100 and the year's first rate hike. Rich Ross, Evercore ISI's senior managing director and head of technical analysis, reads the tape as resilient rather than stretched and points to seasonality — October through December is historically the strongest window — as a tailwind into year-end.
A paradox market where bad news steadied the leaders
What Ross calls a market of misdirection showed up last week. Fresh macro data and a high-profile warning that AI could harm us did not sink AI and semiconductor names; they steadied them. A Fed rate increase did not unleash yields; it capped the 10-year around 5%. For a technician with three decades of chart work, the job is to let price tell you when the consensus narrative is being faded.
That lens shapes his playbook: go back to what worked in the first half, not what hurt in the choppy July-August stretch. Semiconductors sat in the eye of that storm as momentum sold off, yet many charts pulled back only to their 200-day moving averages and held. For Ross that moving average is not decoration; it is the line that tells you whether the primary trend is still intact.
Memory versus equipment: two fronts holding the 200-day
Inside semiconductors the split matters. Capital-equipment names lagged for a while and are now bouncing hard off the 200-day, with Lam Research and Applied Materials as textbook examples. Memory is the other front: Micron, Sandisk and Seagate sit at the commodity heart of the AI build-out and look well positioned. In Ross's view, calling the AI theme over or even slowing misreads where the real leverage sits.
The ecosystem stretches beyond fabs. Intel and Dell do not make cutting-edge chips but they live in the same demand chain and have snapped back sharply in recent days, a reminder that AI infrastructure spending ripples outward. Ross's line about the three most expensive words on Wall Street — 'I missed it' — lands here: many memory charts still trade well below their July peaks, so the entry has not closed.
Power and photonics: the hidden bottlenecks of AI
Beyond chips, Ross highlights two bottlenecks that make data centers run: power and photonics. Bloom Energy, newly added to the S&P 500, represents the power lever, while Ciena and Lumentum represent fiber optics. Ciena was recently upgraded by Evercore, Lumentum held up through the summer drawdown while the market wobbled, and both benefit when bandwidth is the scarce input. Nvidia is his Gibraltar rock — about 20% year-to-date, steadier but with less torque than the faster-moving optics and memory names.
Ciena is a case study in how Ross trades drawdowns. After a prior cycle that delivered more than 1,100% from the last major support test, the stock gave back almost half its value in July-August yet held its key supports and kept the uptrend skeleton intact. It is up roughly 60% year-to-date but still about 40% below its 52-week high, a pattern Ross has seen before: hold the support after a 40-50% cut and the next leg can be measured in multiples, not single digits. He is not promising that repeat, but the optics setup still screens as one of the more compelling.
Software selectivity: why cyber separates from apps
Software was the mirror image in the first half: strong semiconductors coincided with weak software in a zero-sum tape. The market priced a world where AI would eat all software, yet Ross sees nuance. Cybersecurity and parts of data infrastructure have ripped back to new highs — Snowflake, Datadog, Palo Alto Networks and CrowdStrike all printed fresh peaks — while application names such as Adobe and Intuit still struggle beneath resistance built during the software slide. That divergence is a stock-picker's signal, not a reason to blankly avoid software.
Why technology has to lead is arithmetic. Semiconductors alone are about 18% of the S&P 500; add the broader technology complex and the AI theme and you are driving more than half the market. Ross argues you do not get to 8,300 on small-caps or on slow compounding alone; you get there if technology leads. That math holds even with Amazon and Tesla classified outside tech — the index's heart still beats in chips and AI.
A broader tape: health care and industrials step up
One of the bigger stories of this market is that leadership has broadened beyond the narrow MAG 7/FAANG era that began in 2014. Small-caps have beaten large-caps at times this year, and the equal-weight S&P has at moments kept pace or better. For diversification, Ross leans toward health care as an offense-and-defense hybrid: Eli Lilly and Johnson & Johnson, often seen as defensive, are up close to 30% year-to-date on a biotech benchmark like XBI, while industrials lever to the same AI economy but face a near-term headwind from higher crude.
On the avoid side, consumer discretionary screens as one of the weakest sectors year-to-date, squeezed by that same 5% yield, $100 crude, rate hikes, inflation and war headlines — the consumer is resilient but the investment case is thin. Energy has moved hard and stays news-driven; Ross would not chase new money there, though holding some as a hedge is different from fading it while wars price in. The invalidation sits in rates: a sustained break above 5% — daily prints of 5.12, 5.18 or 5.25 — would force even a momentum technician off the 8,300 call, while 30-year highs flag tight global rates and 7-8% mortgages already freeze housing, even if Ross still sees higher yields curing themselves through growth cooling.
AI commentary
"What strikes me is how Ross flips the 'wait for the dip' trap: investors who wait rarely buy the dip when it comes, and the real setup lives in memory and optics names holding their 200-day supports."
AI assessment
Steel-manned, Ross's call has internal coherence: if more than half the market is driven by technology and the AI theme, an ambitious target like 8,300 only makes sense led by semiconductors. The 200-day filter holding through a momentum shock such as July-August is also defensible; in many cycles, leaders that held that average powered the next leg. And the seasonality point is data-backed: over the last 30 years, October-December has on average been the strongest slice of the year.
The limits are about numbers and conditionality. A move to 8,300 from here implies roughly 20% upside and is framed as a year-end possibility, which requires three conditions to hold at once: yields staying near 5%, oil rolling over, and earnings not cracking. The 1,100% Ciena anecdote is a single case; not every name that held support after a 40-50% cut produced that multiple, and Ross explicitly says he is not promising it. Memory cyclicality can also flip fast — one quarter of oversupply at Micron or peers can undo the 200-day narrative.
The practical takeaway splits by investor type. For short-term traders, Ross's rapid-fire sheet — Broadcom over Nvidia, Nasdaq over S&P, momentum over breadth, chips over software infrastructure — is a clean momentum filter, and a sustained break above 5% on the 10-year is a usable invalidation stop. For longer-term accumulators, the balanced expression is a basket rather than a single optics bet: memory plus equipment plus power/photonics, hedged with a health-care kicker like XBI. In both cases, managing the 'wait for the dip' psychology matters more than picking a level.
My net judgment is that Ross's framework works as an offense-tilted compass for the fourth quarter, but 8,300 should be read as a momentum scenario, not a sacred price target. If 5.12-5.25 on the 10-year and new highs in the long bond stick, the scenario invalidates quickly; housing at 7%+ mortgages is already the economy's brake. Until those signals print, names in the memory-optics-power triangle that defend the 200-day deserve to stay on the offensive side of a portfolio — with position size tied to that support, not to the story.
Sources
6 links; no other published story cites them. Stories sharing a link do not confirm each other; a source's origin is not inferred from how often it is cited.
- @youtube.com YouTube — Stop Waiting For The Dip — Rich Ross On What To Buy Now
- @evercore.com https://www.evercore.com/our-business-and-capabilities/equities/research/
- @cnbc.com https://www.cnbc.com/2026/07/30/chip-stock-rally-lam-research-micron-amd.html
- @investing.com https://uk.investing.com/news/stock-market-news/sp-500-at-8000-gold-at-5400-tesla-to-double-here-are-top-trades-for-2026-4426815
- @lumentum.com https://investor.lumentum.com/financial-news-releases/news-details/2025/Lumentum-Optical-Circuit-Switch-to-Improve-Next-Generation-AI-Data-Center-Scalability
- @ciena.com https://www.ciena.com/insights/articles/how-optical-networking-powers-ai-data-centers
s&p 500 · semiconductors · evercore isi · optical networking · technical analysis · nodesdaily