월스트리트의 가장 인기 있는 거래가 뒤집힐 이유 – 그리고 이 "죽은" 주식이 2026년의 승자가 될 수 있습니다

Why Wall Street's Hottest Trade Is About to Flip – and These "Dead" Stocks Could Be 2026's Winners THE DAILY --- H.E.A.T. Torching Wall Street’s Obsolete Playbook. Delivered straight to your inbox 5 days a week. Subscribe Home Posts Why Wall Street's Hottest Trade Is About to Flip – and These "Dead" Stocks Could Be 2026's Winners Why Wall Street's Hottest Trade Is About to Flip – and These "Dead" Stocks Could Be 2026's Winners Dec 18, 2025 I’ve been a trader and investor for 44 years. I left Wall Street long ago—-once I understood that their obsolete advice is designed to profit them, not you. Today, my firm manages around $4 billion in ETFs, and I don’t answer to anybody. I tell the truth because trying to fool investors doesn’t help them, or me. In Daily --- H.E.A.T. , I show you how to Hedge against disaster, find your Edge, exploit Asymmetric opportunities, and ride major Themes before Wall Street catches on. ❝ I’m hosting a webinar entitled “Why Covered Call ETFs Suck and What to Do Instead” (More Info Below) January 15 2-3pm. Sign Up Here Table of Contents --- H.E.A.T.--Software If 2023–2025 was the age of worshipping anything w … The “Death of Software” Narrative The Counter‑Narrative: AI’s Next Leg Has to Flow T … Winners, Watchlist, Losers Likely Winners: “Agentic Rails + Data Moats” Watchlist: Great Businesses in “Show Me” Mode Losers / Danger Zone: Where AI Burnout Really Bite … Takeaways --- News vs. Noise: What’s Moving Markets Today The AI Hangover vs. a Quietly Healthy Market --- A Stock I’m Watching Why Covered Call ETFs Suck-And What To Do Instead --- H.E.A.T. If 2023–2025 was the age of worshipping anything with a GPU, 2026 is shaping up to be the year investors ask a much less romantic question: Where does AI actually show up in revenue and margins? Right now, the answer is not where most people have their money. Look at what just happened to the software leaders that were supposed to be the cleanest “AI productivity” winners: $NOW is down ~27% YTD. $CRM is off ~24%. $ADBE is down ~22%. $WDAY, $TEAM, $ASAN, $MNDY have all been smoked double‑digits. This is after solid prints : beats on earnings, mostly stable or even raised guidance, and a constant stream of “AI assistant,” “agent,” and “copilot” announcements. So why are the stocks trading like they missed the AI memo? The “Death of Software” Narrative Talk to the Street and you hear the same line: ❝ “If AI can just generate workflows, documents, code, and dashboards… why do I need expensive software subscriptions at all?” RBC’s Rishi Jaluria summed it up bluntly: investors are treating AI as “the death of software” , especially application software. The fear is that: Foundation models become the new operating system. Horizontal AI “agents” replace entire categories of apps. Pricing power melts as every vendor bolts the same models onto similar features. Layer on competition—Big Tech pushing AI into every corner of the enterprise—and you get what we’re seeing now: multiple compression even with good fundamentals. The numbers tell you how far sentiment has swung: $NOW at ~38× forward earnings vs a 5‑year average over 60×. $CRM at ~20× vs ~35× historically. $ADBE at ~15× vs ~28×. This is what an AI burnout phase looks like for software: The market doesn’t hate the businesses. It just refuses to pay 2021‑style multiples until it sees proof that AI is more than a keynote slide. The Counter‑Narrative: AI’s Next Leg Has to Flow Through Software Now flip to what HSBC’s Stephen Bersey is saying: ❝ “Material productization of AI will mostly come from the software sector … We see 2026 as the year investors begin to shift focus from hardware to software.” Translate that: The infrastructure leg of the AI trade (GPUs, networking, data centers) is maturing. The application leg —where AI agents are embedded into daily workflows—is still early. The winners will be vendors that already control the data and workflows enterprises run on. That short list looks a lot like: $MSFT – owns the office stack, identity, and much of the cloud. Building a full “agentic ecosystem” inside $AZURE and $O365, backed by a $45B model‑access deal with $NVDA and Anthropic. Azure just grew 40% YoY again. $ORCL – yes, the same Oracle everyone is panicking about on the infra side. Underneath the capex drama, it still controls massive amounts of core enterprise data and is rolling out a vectorized database that makes that data searchable and usable by AI agents. Bersey thinks EPS can grow >25% annually from 2025–2030 and has a $364 PT. $CRM – deeply embedded in customer and sales data. Its Agentforce product is all about letting AI agents qualify leads, answer tickets, and automate front‑office workflows. Management is already talking about a return to double‑digit revenue growth in 12–18 months as this gains traction. $NOW – sits at the “workflow control plane” of the enterprise: tickets, approvals, governance. Its “AI Control Tower” is literally about managing other AI agents. If you believe in an agentic future, something has to orchestrate and audit all those bots. In other words: if AI is going to move from an R&D spend line to a productivity P&L line , it’s going to move through the companies that already run the workflows and hold the data . That’s not the death of software. That’s a transfer of power to the right software. Winners, Watchlist, Losers Again, not recommendations—just how I’m organizing the space right now. Likely Winners: “Agentic Rails + Data Moats” These are the names that: Control mission‑critical data and workflows. Have clear AI roadmaps tied to paid products. Are already cheap relative to their own history. $MSFT – The AI Operating System Owns the desktop, the office suite, identity, and a huge chunk of the cloud. Copilot + Azure “agent ecosystem” lets them charge for AI twice: once in the cloud bill, once in the per‑seat license. Balance sheet and FCF mean they can spend on AI without ever talking to the bond market. $CRM – Front‑Office Agents With Receipts Agentforce is built directly on top of the customer data companies already live in every day. If AI can close tickets and qualify leads, the ROI story is simple: fewer humans per dollar of revenue. Stock is down ~24% this year with a forward P/E ~20× vs ~35× historically; RBC and HSBC both flag it as “significantly undervalued.” $NOW – Workflow Control + AI Governance Already the “system of action” in IT, ops, and HR. AI Control Tower = the dashboard executives use to see which agents are doing what, where, and with what permissions. Market hates the Armis acquisition rumor (dilution + distraction), which is exactly the kind of controversy that often gives you entry points into long‑term winners. $ORCL – Two Stories, One Optionality Balance‑sheet drama on the infra side, yes. CDS blowing out, huge lease commitments, and questions about funding the OpenAI build‑out. But on the software/data side, Oracle’s vectorized database and “data dominance” could be a huge edge as enterprises train agents on sensitive business data that’s already sitting in Oracle systems. If management proves they can fund the infra build‑out without torching the rating, the software side becomes a powerful call option. Watchlist: Great Businesses in “Show Me” Mode These are where I’d build a watchlist, not a full position , and wait for either: Clear AI monetization metrics or Panic‑driven entries. $ADBE – Creative Stack vs AI Flood Solid earnings and guidance, but constant fear that cheap AI tools eat the low‑end creative market. Key question: can Adobe make Firefly and the broader AI toolset something enterprises pay more for, not just a defensive bundle? At ~15× forward earnings vs ~28× 5‑year average, you don’t need perfection—but you do need management to stop missing on innovation vs Big Tech. $WDAY, $TEAM, $MNDY, $ASAN and other “work software” All working hard on AI copilots and workflow automation. What I’d watch: are customers actually paying incremental dollars for AI modules, or just getting them for free as part of bundles? Without clear AI upsell, these stay multiple‑compressed even with okay growth. Losers / Danger Zone: Where AI Burnout Really Bites This is where I’d tread very carefully: 1. “AI Tourist” SaaS Companies that jam “AI” into every press release, but don’t own the data, the workflow, or the distribution. You see it in pitches like “we’re the AI of X,” with no evidence of pricing power or net‑retention uplift. 2. Over‑levered roll‑ups chasing AI via M&A The market is already punishing over‑spenders. $NOW’s reaction to the Armis rumor is a warning shot. If management starts buying AI companies because they missed the internal innovation cycle, that’s usually late‑cycle behavior. 3. “Free model” victims Any software business whose core feature can be trivially replicated with off‑the‑shelf models and no proprietary data is vulnerable. Think one‑trick point tools with weak integration and low switching costs. Takeaways AI burnout in software is real—but mostly in the multiples. The businesses themselves are, in many cases, still compounding; investors just refuse to pay “AI fairy‑tale” prices without hard evidence of monetization. 2023–2025 belonged to hardware and infra. 2026–2028 is where we find out who actually gets paid for AI. That almost by definition means the software layer that owns the workflows, governance, and data pipes. From here, the questions to ask about any software name are simple: What specific AI use cases are they shipping that customers can’t get elsewhere? Can they charge for it? (Seat uplift, module pricing, usage expansion.) Does it show up in net retention, deal size, and margin? The opportunity is in the disconnect. You have high‑quality franchises like $MSFT, $CRM, $NOW, $ADBE, $ORCL that are either derated or stuck in “prove it” mode at the exact moment when AI is about to move from experiment to plumbing. In other words: the Street is exhausted with the AI story just as the software leg of that story is about to matter. That’s usually when you want a watch list, not a eulogy. --- News vs. Noise: What’s Moving Markets Today The AI Hangover vs. a Quietly Healthy Market The headline this week isn’t “market crash” – it’s who is crashing. The AI‑CapEx / high‑beta complex is getting taken to the woodshed while the rest of the market quietly rotates and breathes just fine. The epicenter is still $ORCL. Its equity has become the shock collar for the entire AI build‑out, and now its 5‑year CDS is through the 2022 highs – the market’s way of saying, “We’re not sure you can fund this $100B‑ish AI/cloud plan without abusing the balance sheet.” That’s the broader story: the AI‑infrastructure trade has been declared “over” for now. Growth and momentum are underperforming value by about 300 bps MTD; high‑beta has given back the last three months of gains; and anything tied to “OpenAI‑complex” CapEx (AI semis, DC REITs, GPU clouds) is the funding source. But step back from the AI car crash and look under the hood of the indices: About 55% of stocks are still above their 20‑day and 53% above their 50‑day – that’s neutral, not panic. The advance/decline line has barely budged off the highs even as the S&P has dropped ~200 handles – meaning selling is concentrated in AI and mega‑cap momentum, not the whole market. Equal‑weight S&P, small caps, transports, and value are hanging in; this is rotation, not liquidation. Sentiment, though, has flipped hard. The general Fear & Greed index slid back into “fear,” and the Bitcoin fear/greed gauge is now sniffing the ’22 bear‑market lows , which tells you everything about how crowded the AI / high‑beta trade had become. The chart of $ORCL vs. Bitcoin this year looks like the same asset with different tickers – when your cloud stock trades like a leveraged crypto ETF, you know positioning, not fundamentals, is running the show. Macro backdrop? Still “growth > inflation” in Powell’s mind. November payrolls bounced +64k, but the unemployment rate quietly ticked up again to a new cycle high at 4.56% . Powell spent the last FOMC presser downplaying inflation (“tariffs = one‑off; services cooling”) and leaning into labor‑market downside risk. Translation: unless something breaks, the Fed would rather err on the side of being too easy, not too tight. That’s backed up by the data the market actually cares about: control‑group retail sales up +0.8% and the Affirm CFO saying their consumer still looks “quite healthy” with no credit‑stress tells. Growth is slowing, not falling off a cliff. The wild card for 2026 is politics. Trump wants a Fed chair he can call at 2 a.m. and tell, “Rates should be 1% or lower.” Hassett is still the betting favorite; Warsh is the market’s nightmare; Waller is the “adult in the room” compromise. But the big picture is simple: the new administration wants lower rates, more fiscal juice, and a weaker dollar heading into the midterms. That’s not a deflationary mix. If the economy re‑accelerates on top of that, the bond market will start to price a late‑cycle inflation scare sometime next year. Rates and the dollar are already sniffing around that idea. DXY is trying to hold ~98 – a break below opens the door toward the mid‑90s over time; a reclaim of the 100‑ish zone sets up a counter‑trend rally. The 10‑year has been capped under ~4.2% , with a potential inverse‑head‑and‑shoulders that targets ~4.35–4.5% if growth/inflation re‑heat, and 3.8–4.0% if disinflation plus slowdown fears win instead. The bond market is basically saying: “We’ll let you have your cuts… up to the point where inflation and deficits stop cooperating.” And floating over all of this is the “electrify everything” / AI power story, which quietly supports natural‑gas demand regardless of what $ORCL’s chart does. TD Cowen’s work pegs U.S. data‑center growth at 50+ GW by 2030 , adding 6+ Bcf/d of gas demand on top of LNG and industrial needs – enough, in their model, to keep Henry Hub north of $4 long‑term and closer to $5 in 2026, with certain gas E&Ps earning north of 80% of their current EV in free cash flow by 2030 at those prices. What’s News vs. Noise here? News: AI‑CapEx excess is finally being priced; $ORCL’s CDS blowout makes financing risk part of the AI story, not a footnote. High‑beta and momentum have fully round‑tripped their last three months of gains. The Fed is openly more worried about jobs than inflation. Politics point to more stimulus and a softer dollar into 2026, not less. Noise (for now): The idea that “AI is over” or that this unwind equals a bear market. Breadth is still healthy, the selloff is narrowly focused, and we’ve already had two “AI is dead” scares this year that turned out to be great entry points once the leverage came out of the system. How I’d frame it: this is less “end of the AI story” and more “margin call on the weakest balance sheets in the AI ecosystem.” From here, I want to: Stay underweight debt‑funded AI infra tourists and speculative GPU clouds that depend on perfect execution and cheap capital. Keep core exposure in self‑funded AI winners plus the “physical AI” trade – gas, power, and grid names that get paid whether or not the latest model beats the benchmark. Treat this AI unwind as late‑innings positioning cleanup, not a structural break – and watch the S&P 100‑day and Mag‑7 trendline as the real tells on whether it stays a rotation… or becomes something nastier. --- A Stock I’m Watching Today’s stock is Si Bone (SBIN)….. SIBN (SI-BONE) is a “picks-and-shovels” way to play the continued shift of musculoskeletal care toward minimally invasive, outpatient procedures: the company is effectively the category creator in sacroiliac (SI) joint fusion with a large, still-underdiagnosed patient pool and a growing base of clinical evidence that supports broader physician adoption. The asymmetric setup is that once a rep team and surgeon base are in place, incremental procedure growth can translate into meaningful operating leverage—so a modest step-up in utilization (more surgeons doing more cases) can matter disproportionately for earnings power. The bull case is straightforward: improving diagnosis + expanding surgeon penetration + durable reimbursement = a long runway in a niche that’s big enough to compound but focused enough to defend. The bear case is equally clear: reimbursement or coverage tightening, higher-than-expected competition from larger spine players, or a slower “education cycle” (surgeons/payers) that keeps growth choppy. What I’d watch to validate the thesis is less the narrative and more the “plumbing”: procedure volume momentum, salesforce productivity, gross margin trajectory (mix + manufacturing efficiency), and any signals that coverage policies/guidelines are widening rather than narrowing—because for SIBN, adoption velocity is the whole game. How Else I Can Help You Beat Wall Street at Its Own Game Inside --- H.E.A.T. is our monthly webinar series, sign up for this month’s webinar below…. Why Covered Call ETFs Suck-And What To Do Instead Thursday January 15, 2-3PM EST Covered call ETFs are everywhere — and everyone thinks they’ve found a “safe” way to collect yield in a sideways market. The truth? Most of them suck. They cap your upside, mislead investors with “yield” that’s really your own money coming back , and often trail just owning the stock by a mile. Join me for a brutally honest breakdown of how these funds actually work — and what you should be doing instead. What You’ll Learn: 🔥 Why “high yield” covered call ETFs are often just returning your own capital 📉 How most call-writing strategies quietly destroy compounding 🚫 Why owning covered calls in bull markets is like running a marathon in a weighted vest 💡 The simple structure that can fix these problems — and where the real daily income opportunities are hiding Sign Up Here All rights reserved. Keep Reading No posts found View more caret-right 0