Hello Fellow Patriots,
Tech has felt choppy in early February 2026. QQQ is down about 2.8% YTD (with a sharp slide around Feb 5), while SPY is only down about 0.6%, so the pain is showing up most in tech.
That’s why earnings week can be weird. A company can post solid numbers and still drop on guidance worries, big capex headlines (think AI spend), profit taking, or a simple “sell the news” reaction.
I’m treating 2026 like a stock picker market, because tech isn’t one trade anymore. In this post, I’m sharing my process and a watchlist style approach for the earnings-week dip, focusing on essential building blocks (AI infrastructure, connectivity, servers, and real world digital hardware) instead of hype apps.
This isn’t financial advice. I’m just laying out how I screen, what I watch, and what I’d consider buying if the price gets hit for the wrong reasons.
What makes an earnings week dip worth buying (and what makes it a trap)
An earnings-week drop can be a gift or a warning label. I’m not trying to catch every red candle. I’m trying to buy the dip when the stock price is reacting to noise, not when the business story is cracking.
My quick filter is simple: if the company’s future earning power looks intact (or improving), I’ll consider scaling in. If the report hints that demand is getting destroyed, margins are breaking, or a key customer is pulling away, I treat the dip like a trap and move on.
I separate a headline miss from a real business problem
I start by asking one question: is this a one-quarter bump or a broken trend?
A one-quarter issue often looks ugly in the headline but stays clean underneath. Think: revenue slightly light because a few deals slipped, gross margin down because of a product mix shift, or guidance phrased cautiously because management hates being aggressive. These are the kinds of lines that can spark an overreaction:
- “Capex is rising” (common in AI infrastructure). Higher spend can be smart if it supports real demand. TSMC raising 2026 capex (from $52B to $56B) can scare people who only see the bill, not the bookings.
- “Macro uncertainty” or “customers are optimizing spend.” Sometimes that means delayed orders, not lost orders.
- “We’re being prudent” guidance language. The stock can drop even when the midpoint is fine, because investors wanted a beat and raise.
A broken story shows up in the trend lines. I look at:
- Revenue growth trend: Is growth steady, re-accelerating, or fading for multiple quarters? One soft quarter is normal, a clear down-slope is not.
- Margins: A small margin dip is fine if it’s explained (mix, ramp costs). A multi-quarter slide with weak explanations is a problem. ASML is a good example of a stock reacting to margin expectations even while orders stay strong.
- Customer concentration: If one customer drives a huge chunk of sales, a single procurement change can wreck the year. I want to know if the company is diversifying or getting more dependent.
- Order backlog and bookings: Backlog is the “already sold” pile. Strong bookings can make a soft quarter easier to ignore. ASML’s €13.2B in new orders is the kind of detail that matters more than a single quarter’s optics.
- Delayed vs. destroyed demand: Delayed demand comes back (deals slip, builds move). Destroyed demand does not (a product gets displaced, a platform shift, a competitor wins the socket).
If I can explain the dip with timing, language, or investor positioning, I get interested. If I have to explain it with excuses, I pass.
The 5 numbers I check right after the call
Right after the release and call notes, I check five numbers that tell me if I’m looking at a bargain or a value trap.
- Forward growth (revenue or EPS): I want a believable path for the next 2 to 4 quarters, not just last quarter’s beat.
- Free cash flow (FCF): Earnings can be “cleaned up.” Cash is harder to fake. If FCF is strong (ASML has posted strong annual FCF recently), the business usually has real strength.
- Gross margin trend: I’m fine with flat margins if growth is strong. I get cautious when margins compress while growth slows.
- Guidance range and what changed: I compare the new range to the prior guide and to what management implied about demand. A tighter range, a lower midpoint, or a new “uncertainty” phrase can matter more than the quarter that just ended.
- Valuation versus growth (PEG as a shortcut): A high P/E can be normal for a fast grower. The catch is the growth has to be real and durable. If growth is slowing but the multiple is still priced for hypergrowth, dips can keep dipping.
This is my sanity check. If two or three of these flash yellow at once, I slow down.
My dip buying rules so I do not panic buy
When a stock gaps down after earnings, my biggest enemy is speed. My rules are built to keep me from buying because the chart looks scary.
Here’s what I stick to:
- I scale in with 2 to 4 buys. My first buy is a starter, not a statement. If the stock stabilizes and the thesis holds, I add.
- I set a max position size before I click buy. If I don’t know my max size, I’m not investing, I’m reacting.
- I wait for volatility to cool. On earnings days, I usually give it the first 30 to 60 minutes after the open. Early prints are often emotional.
- I don’t buy before I read the release and key call notes. Price action alone is not research. I want the guidance, margin commentary, and demand tone in plain words.
- I use limit orders. Market orders on an earnings gap can fill in the worst spot.
- I pre-plan the next 10% down. Before my first entry, I decide what I’ll do if it drops another 10%:
- If the business is fine, that’s often where my next scale-in sits.
- If new info shows a real business issue, I stop buying and reassess.
Those rules keep me calm. I’m not trying to “win” the first hour after earnings. I’m trying to buy great tech businesses when the market is busy overreacting to the headline.
My top tech stocks to buy on the earnings week dip, focused on the building blocks
When I’m shopping an earnings-week dip, I want “building block” tech, the stuff other tech companies have to buy to grow. That’s why this mix works for me:
- AI bandwidth and optical plumbing (Lumentum)
- Advanced chip manufacturing capacity (Taiwan Semiconductor)
- Enterprise servers and storage (Dell)
- Real-world digital infrastructure (Daktronics)
These are not “one app goes viral” stories. They’re more like toll roads and toolboxes. When the market freaks out over guidance wording or a capex headline, I’m looking for moments where the long-term demand picture still looks healthy.
Lumentum (LITE), a pick and shovel play on AI bandwidth and optical networking
Lumentum sells optical and photonic components that help move data fast. That includes parts used in high-speed networking gear and data center connections, where latency and throughput matter. If you picture AI as a factory, Lumentum makes some of the conveyor belts and sensors that keep materials moving.
The AI push is a bandwidth story as much as it’s a compute story. Training and serving models creates huge east-west traffic inside data centers, plus more demand for data center interconnect links between facilities. When racks get packed with accelerators, the network becomes the bottleneck. That’s where fiber capacity, faster optics, and better packaging start to matter a lot.
Valuation is where people get spooked. Lumentum can screen as expensive on a simple multiple, and I get why. A P/E of 67.51 looks like a red flag if you stop there. But I don’t stop there. When growth is strong, I care more about growth-adjusted value. A PEG of 0.67 can look cheap if the growth holds up, because it hints the market may not be pricing in the full earnings ramp.
On the earnings call, I’m listening for a few very specific signals:
- Cloud capex signals: Are hyperscalers still building, or are they pausing orders quarter to quarter?
- Data center interconnect demand: Are customers pulling forward optical upgrades, or are they “waiting for the next gen”?
- Optical scale-up opportunities: I want updates on co-packaged optics (CPO) momentum and ultra-short reachconnections (the really short, high-throughput links that matter inside dense AI clusters).
Why I’d buy the dip: If the stock sells off on cautious language, but management still talks about strong optical demand and a solid pipeline, I’m interested. This is the kind of name that can look “overvalued” right before it looks “obvious,” because the network buildout tends to happen in waves.
Two risks I keep in front of me:
- Customer spending pauses: Cloud and large networking buyers can freeze budgets fast, even if demand comes back later.
- Cyclical telecom demand: Telecom can swing from upgrade cycles to digestion periods, and optics suppliers feel that whiplash.
Taiwan Semiconductor (TSM), the backbone of advanced chips in a capex heavy world
TSMC is the top contract chipmaker. It manufactures advanced chips for many of the biggest names in tech. That sounds simple, but the moat is brutal to copy because it’s not just one thing. It’s process know-how, supply chain depth, yield learning, equipment relationships, and the ability to scale leading-edge nodes with high reliability.
AI accelerators and high-performance computing are the cleanest demand drivers here. Even if consumer gadgets wobble, the buildout of data center compute keeps pulling advanced wafers. When companies race to train bigger models and deploy them across products, they need more leading-edge silicon, more packaging capacity, and more predictable supply.
The main fear I see people trade around is capex. Semiconductor manufacturing is expensive, and headlines about spending can pressure the stock. The way I think about it is simple: this is a capex-heavy business by design, and the winners are the ones who can invest through cycles without breaking their balance sheet or losing technical ground.
That’s why I pay attention to profitability and cash generation. TSMC produces very large cash from operations, which matters because it helps fund investment when conditions soften. In other words, they can keep building even when the market gets nervous, and that’s often how leaders stay leaders.
I also care about geography and capacity. The plan to produce advanced 3-nanometer chips in Japan matters because it points to diversification and resilience. More locations can help with customer confidence and supply stability, especially when geopolitical worries flare up. It also signals a long-term commitment to expanding advanced capacity closer to key manufacturing ecosystems.
During earnings week, I’m not trying to predict every macro twist. I want a few simple signs that the core story is intact:
- Utilization: Are fabs staying busy at advanced nodes, or is there a meaningful air pocket?
- Pricing power: Are they holding the line on pricing, especially for leading-edge capacity?
- Node leadership: Are they staying ahead on the process roadmap, with clear progress and customer adoption?
Two risks I respect here:
- Geopolitics: This is the big one, and it can reprice the stock fast on headlines.
- Cyclical downturns: Even great chip companies live with cycles, and downturns can compress multiples and push customers to cut orders.
Dell (DELL), a cheaper way to play enterprise AI servers and hybrid cloud builds
I don’t think of Dell as “just PCs.” The story I care about is the enterprise side: servers, storage, networking, and services that keep business IT running. When companies decide they want more AI capability on-site, Dell sits right in the purchase order flow.
On-prem AI is not a fantasy. A lot of real businesses do not want to ship all their sensitive data to the public cloud, and many want predictable cost and performance for specific workloads. That pushes them toward hybrid cloud setups, where some workloads run in the cloud and others run in their own data centers. AI fits that pattern well, especially for companies with privacy needs, regulatory limits, or steady internal demand.
That’s why I like Dell as an earnings-week dip candidate. It’s a more practical AI play: sell the picks, shovels, racks, and storage that enterprises need to run models and manage data.
The valuation setup also catches my eye. Dell shows a P/E of 11.59 and a PEG of 0.78 in the notes I’m working from. That’s the kind of GARP profile I like when earnings volatility shakes out weak hands. If growth holds up, the multiple does not require heroics.
I also keep the Nvidia angle simple: Dell builds and ships systems designed to run AI workloads. Nvidia provides the GPU platform, and Dell is one of the ways enterprises buy complete, supported hardware stacks. For many CIOs, buying “a known system from a known vendor” beats stitching together parts.
On the call, I’m listening for plain-English demand signals:
- Are enterprise customers buying AI servers now, or just “testing”?
- Is storage demand rising with AI data growth?
- Are services and support attaching, or is it a bare-metal price fight?
What could go wrong is pretty straightforward:
- Slower enterprise spending: If IT budgets tighten, big server refreshes can slip.
- Margin pressure: Aggressive pricing, mix shifts, or higher component costs can squeeze profits.
- Supply chain shocks: AI hardware depends on tight supply coordination, and disruptions can delay revenue or raise costs.
Daktronics (DAKT), a small cap way to ride digital displays and connected venues
Daktronics makes large-format digital displays. Think scoreboards and video walls in stadiums, signage in transport hubs, and big screens in venues that want to sell ads and improve the fan experience. It still fits “tech” in my book because it’s digital infrastructure. These are connected systems that blend hardware, software, installation, and long-term service.
What I like about Daktronics is that it can have a very “real economy” demand profile. Teams and venues want new display systems to boost sponsor revenue and modernize the experience, and public spaces want clearer, more flexible signage. It’s not trendy, but it’s durable when the project pipeline is healthy.
The backlog is the anchor here. The notes show a $321M backlog, up 36% year over year, which gives me better visibility than I usually get in a small cap. Backlog does not remove risk, but it helps answer the question, “Is demand real, or is it just talk?”
There’s also specific project momentum. The notes reference MLB-related display projects, and the key point I take from that is not a single contract. It’s that Daktronics keeps winning meaningful, high-visibility installs that can lead to follow-on work and service revenue.
Valuation can be another reason I watch it on dips. With a PEG of 0.74 and EV/Sales of 1.21 in the notes, it can screen like a reasonable growth-at-a-fair-price small cap, especially when the market is punishing anything without a mega-cap label.
Small caps drop hard on earnings weeks for a couple of non-business reasons:
- Liquidity: Fewer buyers and sellers means bigger moves on the same news.
- Guidance tone: If management sounds cautious, the stock can drop even when numbers look fine, because traders shoot first and read later.
Two risks I keep front and center:
- Project timing: Revenue can slip if installs get delayed, permits take longer, or schedules change.
- Customer budgets: Venues and public-sector buyers can pause projects if financing or budgets get tight.
If I’m buying a dip here, I want the backlog to stay solid and commentary to support continued project conversion. When that’s intact, a sharp earnings-week selloff can be more about market mechanics than fundamentals.
How I time my buys during earnings week without trying to be a day trader
Earnings week is when good stocks do stupid things. Big gaps, fast headlines, and one sentence from a CEO can move a chart more than a full year of steady execution. I don’t try to “trade” that noise. I plan a few calm entry windows, size small at first, and let the story, not the candle, tell me what to do next.
My simple calendar plan, before earnings, after earnings, and the week after
I follow a three-window calendar. Each window has a different risk level, and I treat them like different tools.
1) Before the print (highest risk, smallest size)
I only buy before earnings when I already want the stock for the next 1 to 3 years and the price is already in a good spot. I keep it small because I’m accepting “gap risk,” meaning the stock can open down 10% to 25% and I can’t do anything about it.
My rule: starter position only (about 25% of my intended size), and I’m fine if I miss it.
2) After the release and call (more info, still volatile)
This is my most common window. I read the release, scan the key numbers, then listen for the real tell: guidance, margins, and demand tone. Even then, the first hour can be chaos, so I’m patient.
My rule: I wait for the market to react, then I buy one staged entry if the thesis still holds.
3) The following week (often the cleanest setup)
A lot of the drama fades after a few sessions. Analysts update notes, traders move on, and the stock either builds a base or breaks for real. If it’s going to stabilize, I’d rather buy into that than guess the bottom on day one.
My rule: if the story is intact, I’ll place my second or third entry the next week, when price action is less emotional.
I pair a fundamental check with one easy technical signal
I keep this simple: fundamentals tell me what I want to own, and one chart reference helps me choose a decent spot to start.
My fundamental check is quick:
- Did guidance get worse for a real reason, or was it just cautious wording?
- Did margins crack, or did they explain a temporary hit?
- Did they confirm demand (backlog, bookings, pipeline), or did they dodge?
Then I use one technical “reference line,” not a magic trigger. Most of the time I use the 50-day or 100-day moving average as a sanity check for trend. If a stock is in a long-term uptrend, these areas often act like “speed bumps” where selling can slow down.
I also mark simple support zones, meaning prior areas where buyers showed up. If the stock flushes below support on earnings, I don’t panic buy. I wait to see if it can reclaim that area, or at least stop making new lows for a couple of sessions.
Trend lines are context for me, not the reason to buy. The reason is always the business.
My risk controls, when I take a loss, when I add, and when I do nothing
This is the part that keeps me from turning an earnings dip into a portfolio scar.
Position sizing: I decide my max size before I buy. For a single earnings-week idea, I usually cap it at 2% to 5% of my portfolio (smaller if it’s a small cap). I build that position in 3 steps:
- 25% starter
- 35% add
- 40% final add
Max loss per position: I set a line where I’m wrong. For most names, I’m not willing to eat more than 7% to 12% from my average cost unless the fundamentals clearly improved after I bought. If I get stopped out, I step away and re-read the report later with fresh eyes.
If the dip keeps dipping: I don’t “average down” just because the price is lower. I only add when:
- the thesis is intact, and
- the stock starts to stabilize (tight trading, fewer big red days, or a reclaim of a key area)
If it keeps sliding on new bad info, I do nothing. Cash is a position.
Time stop: If management’s story changes, or the next update shows the same weakness getting worse, I cut it even if the chart looks “cheap.”
And I never bet the farm on one earnings reaction. I spread my risk across a few quality names so one ugly gap doesn’t wreck my month.
The big 2026 theme behind these picks, AI capex, connectivity, and real infrastructure
My 2026 through-line is simple: AI is moving from excitement to buildout. That means more spending on data centers, chips, servers, networking, and the physical stuff that makes the whole AI stack run day after day.
This is why my watchlist leans toward “real infrastructure” names. When big tech talks about expanding capex, the market often reacts like it’s bad news. I see it more like a highway expansion project. It’s expensive and messy now, then traffic flows better for years.
Why capex headlines can scare the market even when they are bullish long term
Capex is just money spent today to build capacity for tomorrow. The catch is that spending hits near-term profits and cash flow. Depreciation rises, free cash flow can dip, and management might guide margins down while new builds ramp. If you only trade next quarter, that sounds like trouble.
Early February gave a clean example of how fast this can hit tech. Alphabet highlighted $175 to $185 billion of planned 2026 capex for AI infrastructure, and the big capex talk across mega-cap tech spooked investors. The Nasdaq-100 sold off after the headlines. That’s the “sell first, ask questions later” reflex.
Here’s how I keep it grounded:
- Capex is a bill, not a result: It tells me intent, not ROI. I still need proof demand is real.
- The market hates timing gaps: The spend is immediate, the payoff can take years.
- Bigger capex can strengthen the suppliers: Even if the spender’s margins compress, the companies selling chips, optics, servers, and power related gear can see orders sooner.
When I see capex fear hit my infrastructure picks, I don’t celebrate blindly. I read the call notes and look for one thing: are they building into real demand, or building “just in case”?
Why I prefer enablers over apps in a choppy tape
When price action is jumpy and investors are picky, I’d rather own the building blocks than a single “must-win” application.
Apps can be amazing businesses, but they often depend on one product cycle, one user trend, or one pricing model holding up. Enablers have a different setup. If AI spending stays big, lots of customers need the same core ingredients.
In plain terms:
- Chips: The compute engines that train and run models.
- Optics and connectivity: The fiber and photonics that move data inside and between data centers.
- Servers and storage: The racks, systems, and data gear enterprises and cloud firms buy to deploy AI.
- Displays and physical tech: The real-world hardware that turns digital into revenue (like venue and signage systems).
I like enablers in 2026 because they can sell to many buyers at once. One cloud giant slowing down hurts, but it doesn’t always break the story if others keep ordering. That gives me better visibility than betting on a single app being the winner.
My quick check for when I should just buy the ETF instead
Sometimes the right move is skipping single names and buying the basket. If I’m not willing to read earnings releases, scan guidance, and listen for margin and demand clues, an ETF can be the safer choice.
The real-time context also matters. The notes I’m working from mention QQQ outflows and higher volatility, without giving clean numbers. That still matches what I feel on the tape: fast rotations, sharp one-day drops, and sudden bounces.
My rule of thumb looks like this:
- If I can’t follow 3 to 5 earnings calls this month, I buy the ETF.
- If I can do the work, I concentrate into a few high-conviction names.
- If volatility is spiking and I’m unsure, I split it, part ETF, part my best 1 to 2 setups.
In volatile stretches, stock picking can beat broad exposure, but only when I treat it like a job. If I’m not doing the work, I don’t pretend I’m stock picking, I just buy the index and move on.
Conclusion
Earnings-week dips are normal in tech, even when the business is fine. In 2026, I’m treating this as a stock picker year, and I’m staying focused on the less flashy parts of the stack that still get paid when AI and IT spending stays real.
My four dip-buy targets are Lumentum (LITE) for connectivity and optics, Taiwan Semiconductor (TSM) as the foundry backbone, Dell (DELL) for enterprise servers and storage, and Daktronics (DAKT) for real-world digital displays. Each one sits in infrastructure, not hype, which is why I’m willing to be patient when a headline or guidance tone spooks the tape. The goal is to buy quality when sentiment breaks, not to chase a pop.
My next steps are simple, I build a watchlist, set price alerts, read the release and key call notes, then scale in with small buys if the story holds. I keep risk tight with position limits and clear “I’m wrong” levels. Thanks for reading, if you’re tracking earnings this month, I’d like to hear what name you’re watching for a clean dip.