The Inflection Point Theory: How to Read Stocks the Way You Read People What If Stock Selection Is Not About Numbers? Most quant models are built on one assumption: the future looks like the past. You find patterns in price history, assume they repeat, and trade accordingly. But what if you're watching the wrong thing entirely? A mentor I know doesn't pick stocks the way Wall Street teaches. She picks people. She watches for the moment a person decides to change—and that moment, she says, is where all the signal lives. She calls it "翻篇" (fān piān): turning a new page. A point where the old rules stop applying and something new begins. I spent a morning listening to her explain this to a student. The conversation was about stocks, but it wasn't about numbers. And as I transcribed it, I realized this framework might be the most practical thing I've encountered about reading markets. This article is my attempt to translate her logic into a framework you can actually work with—not because I think I've nailed it, but because the attempt reveals how different this thinking is from what most of us were taught. In trading, everyone talks about trends. "Trade with the trend." "The trend is your friend." But here's what's rarely asked: what makes a trend start in the first place? Her answer: an inflection point. A moment of "当下生灭" (dāngxià shēngmiè)—instantaneous birth and death. The old pattern dies; a new one is born. You can't predict it from the old pattern. You can only recognize it when it happens. This is not the same as "breakout" in the technical analysis sense. A breakout is a price crossing a resistance level. An inflection point is deeper—it's a fundamental shift in behavior. Someone who was stuck for months suddenly takes action A person who always deferred decisions starts making them decisively Someone goes from passive to active without external pressure A company that struggled for years suddenly finds product-market fit Leadership that was reactive becomes proactive An industry that was declining attracts a new kind of player who changes the game The key difference: you can't tell from the historical chart alone. You have to know what's happening inside the company—and that requires knowing the people. Here's a practical insight she shared: watch three related stocks together instead of one alone. Why? Coarse reading is clearer. Three stocks in a cluster show you the map. One stock in isolation shows you noise. Interaction reveals signal. When three stocks influence each other, their relative movement tells you something none of them could tell alone. It's lazy—and lazy works. She explicitly said she learned this "lazy" approach from her partner. Not doing the micro-analysis on every tick. Just watching the gross structure. This maps directly to portfolio construction: Find 3 companies in the same ecosystem (same supply chain, same customer base, same industry cycle) Watch them move together and against each other The one that's out of sync with the other two—that's your signal Most investors have a return-maximizing algorithm: find undervalued assets, buy them, wait for the market to agree. Her algorithm is different. She said explicitly: "I don't buy stocks because they'll go up. I buy stocks that let me see clearly. Whether they go up or down, I'm learning something." This is a fundamentally different optimization target: Traditional Investor Her Approach Maximize return Maximize insight Diversify to reduce risk Concentrate to increase clarity Cut losses Watch the person, not the price Build position over time Move when the inflection is clear And here's the hard part: you can't fake this. You need to actually know the people in the companies you're watching. That's the prerequisite no quant model accounts for. So what would a quantitative system built on this logic actually look like? Here's my attempt to translate the philosophy into process: An inflection point has three characteristics: Behavioral shift: The company stops doing what it was doing and starts doing something recognizably different Temporal clustering: The shift happens in a compressed time window—not gradually over years, but noticeably within weeks or months Leadership involvement: The change is driven by people, not just market conditions Rather than one indicator, layer multiple: Price-volume divergence: The stock breaks out with volume that doesn't match prior patterns Fundamental catalyst detection: New management, new product line, new customer segment—any qualitative change that precedes the quantitative change Social signal for public companies: How does the company's communication style change? Earnings call language? Employee reviews? Glassdoor trajectory? Pick 3 companies in the same value chain or competitive set. Track: Their relative performance over 90-day windows When one diverges from the other two, investigate why When all three move together, watch for the next divergence Here's the twist: position size is not determined by confidence in return. It's determined by how much the position teaches you. High teaching value, high conviction → larger position Still learning → smaller position or watch-only Can't learn anymore → exit I have to be direct: I don't have a production-ready quant model sitting here. What I have is a philosophical framework that changes what you're optimizing for. Most of quant finance is built on the assumption that prices contain all information. This framework says: no, the interesting information is in the people, and people are not in the price. You can spend your career building increasingly complex factor models on historical prices. Or you can accept that the real edge is in understanding inflection points—and accept that understanding requires work that can't be automated. That's not a comfortable conclusion for a programmer. We're trained to externalize cognition into systems. This framework asks you to internalize it. One line from that morning's conversation has stayed with me: "You can only turn a page when you're willing to let the previous page end." In markets, most people can't let the previous page end. They hold onto the losing position because "it'll come back." They hold onto the old thesis because changing it feels like admitting failure. The people who are good at inflection points—they've already ended the previous page in their own minds. They've done the inner work of letting go. That, more than any indicator, is what I keep thinking about. This piece is an interpretation of a conversation I transcribed. The framework is mine, built from listening. If I've missed something or gotten it wrong, that's on me.