Latest / Stock Picking with Fexingo: Individual Equities, Research, and Building a Concentrated Portfolio / How to Build a Thesis That Beats the Market
Transcript
- Lucas: Most stock pickers fail not because they can’t read a balance sheet, but because they haven’t actually built a thesis that survives contact with reality. Look at Apple this week; it’s up five point eight percent over the last five days while the S&P 500 is barely moving at seven thousand six hundred forty points. That kind of divergence doesn’t happen by accident, and it certainly doesn’t happen if you’re just trading off headlines. Luna: Right, and half the people buying into that rally probably think it’s just AI hype again. But the cash flow story there is what actually matters for the long game. Lucas: Exactly. Before we dive into how to structure that kind of conviction, I want to say something quickly about the show itself. If our money conversations have helped you think about a decision differently lately, the way these stay ad-free is listener support — buy me a coffee dot com slash fexingo. Luna: Yeah, it really helps keep the research deep when you’re around. Now, back to the table. Lucas: So let’s talk about the anatomy of a good thesis. A lot of investors confuse a narrative with an argument. Saying 'Apple is great' isn’t a thesis. A thesis is a falsifiable statement about future cash flows driven by specific operational levers. Luna: Like what? Giving me an example of a bad thesis versus a good one would help clarify that distinction right now. Lucas: Sure. A bad thesis is: 'Apple will grow because services are big.' That’s vague and easy to dismiss. A good thesis is: 'Apple’s installed base of active devices is growing faster than its unit sales, meaning recurring revenue per user will increase even if hardware volume stays flat, driving operating margins above thirty-five percent next year.' Luna: That’s much tighter. It gives you a specific metric to watch and a clear path to validation or rejection. Lucas: And that specificity is what protects you from confirmation bias. When you have a precise mechanism in mind, you know exactly what data to look for. If the data contradicts the mechanism, you kill the trade. Most people never do that. Luna: But how do you find those mechanisms in the first place? Everyone has access to the same earnings reports and the same news feeds. Lucas: You look for dislocation between price and fundamental reality. Take Microsoft this week; it’s up nearly three percent. The market is pricing in continued cloud dominance, but the real alpha might be in how they’re managing their capex relative to AI demand. Are they spending efficiently, or are they just throwing money at the problem? Luna: I’ve been looking at their capital allocation notes. They seem to be prioritizing internal tools over external expansion right now. Lucas: Precisely. And that’s a signal. If they’re building internal efficiency, it suggests they see a bottleneck elsewhere, maybe in energy or cooling infrastructure. That changes the valuation model entirely. You stop looking at top-line growth and start looking at margin expansion through cost control. Luna: So the thesis shifts from 'cloud growth driver' to 'margin improver via operational leverage.' That’s a completely different bet. Lucas: It is. And it’s one you can actually act on. The problem with generic 'growth at any cost' themes is that they don’t tell you when to sell. If the margin improvement stalls, you exit. Simple. Luna: But doesn’t that require a lot more time? Reading footnotes and cross-referencing capex statements takes hours most people don’t have. Lucas: It does. Which is why concentrated portfolios work better for individual investors than diversified ones. You only need ten strong ideas, not two hundred weak ones. If you spend twenty hours researching ten stocks, you’re in a better position than someone who spends ten minutes checking ten ticker symbols. Luna: There’s truth to that. Diversification is often just a shield for ignorance. If you don’t know what you own, you shouldn’t own it. Lucas: Exactly. Let’s bring it back to Apple for a second. Why did it move five point eight percent this week? Was it earnings? No. Was it guidance? No. It was likely a shift in sentiment regarding their share buyback program. Luna: Oh, I saw that. Their quarterly buyback authorization was larger than expected. That’s a direct vote of confidence in their own balance sheet. Lucas: And that’s a concrete piece of evidence for your thesis. They aren’t just talking about growth; they’re using excess cash to reduce share count. That boosts EPS without needing more revenue. It’s a pure financial engineering play that works when interest rates are stable. Luna: But what if rates drop further? Then debt becomes cheaper, and they might borrow instead of using cash reserves. Does that change the thesis? Lucas: It refines it. Your thesis needs to account for macro variables. If rates drop, the cost of capital decreases, which validates higher valuations. So the thesis becomes: 'Apple is undervalued relative to its risk-adjusted cost of capital, and buybacks accelerate under both low-rate and high-cash scenarios.' Luna: That feels robust. It covers multiple bases. Most people just buy because the chart looks pretty. Lucas: Charts are lagging indicators. Fundamentals are leading. But here’s the trap: everyone knows fundamentals lead. So why does the market ignore them sometimes? Luna: Because liquidity trumps logic in the short term. Or because people are reacting to fear, not facts. Lucas: Fear is the biggest killer of thesis integrity. When NVIDIA dropped six percent this week, a lot of traders panic-sold based on sentiment rather than checking if the underlying demand for AI chips had actually changed. Did it? Not really. The order books are still full. Luna: So the opportunity was created by noise. If you had a solid thesis about long-term AI infrastructure demand, you could have bought the dip. Lucas: Yes. But you had to be willing to sit on your hands while everyone else was selling. That’s the hardest part of stock picking. It’s not the math; it’s the discipline. Luna: Discipline requires a plan written down before you enter the trade. Otherwise, you’re just guessing. Lucas: Agreed. Write down your entry price, your target price, and your invalidation point. If the stock hits the invalidation point, you sell. No debate. No hope. Luna: That sounds brutal, but it keeps emotions out of the equation. Hope is not a strategy. Lucas: Hope is the enemy of returns. Let’s look at a counter-example. Consider Intel this week. They were in the headlines for premarket moves. A lot of people thought it was a turnaround story. Luna: It’s always a turnaround story until it isn’t. The manufacturing delays kept dragging on. Lucas: Right. If your thesis was 'Intel will regain market share in foundry services,' you needed to see specific customer wins. Did they get any major design wins this quarter? Not really. So the thesis was broken, but the price didn’t reflect that immediately. Luna: Because the market was distracted by other tech news. Distraction creates mispricing. Lucas: Exactly. Your job is to find those moments of distraction. Where is the market looking away? Right now, it’s looking at AI models. But what about the companies enabling the physical infrastructure? Energy, cooling, silicon manufacturing equipment. Luna: Marvell was mentioned in the headlines too. They make the custom chips for cloud providers. Is that a better angle? Lucas: Potentially. Custom silicon is becoming a moat for big cloud providers. If Marvell is designing chips exclusively for Amazon or Google, they have sticky revenue. That’s a durable thesis. Less volatile than consumer software. Luna: Sticky revenue means predictable cash flows. That’s easier to model and therefore easier to hold through volatility. Lucas: Which brings us back to concentration. If you have ten stocks with sticky revenue and clear catalysts, you don’t need to watch the VIX every hour. You monitor the fundamentals. Luna: And if the fundamentals hold, the price follows eventually. Patience is just another form of alpha. Lucas: Absolutely. The market is a voting machine in the short run, but a weighing machine in the long run. Weighing requires effort. Voting is easy. Luna: So the key takeaway is to do the weighing. Build the thesis, test it, and stick to it unless the data breaks. Lucas: Yes. Don’t just pick stocks. Pick reasons. And make sure those reasons are stronger than the noise. Luna: Stronger than the noise, but flexible enough to adapt if the world changes. It’s a balance. Lucas: It is. And that’s what separates professional investors from gamblers. Gamblers chase the noise. Professionals wait for the weigh-in. Luna: I’ll take the weigh-in any day. It’s slower, but it gets you home. Lucas: Well said. Next time you’re looking at a chart, ask yourself what the balance sheet is saying instead. Luna: And if the two disagree, trust the balance sheet. The numbers don’t lie, even if the prices do. Lucas: Let’s see if that holds up over the next quarter. For now, keep your thesis tight and your portfolio focused. Luna: Tight thesis, focused portfolio. Sounds like a recipe for sleep at night. Lucas: Sleep is underrated. It helps you think clearly tomorrow.