Correlation and diversification — why ten stocks can be one bet — trading basics, chapter 9
Owning ten stocks feels diversified. If they all move together, it's one position with extra steps. What correlation is, why narrative-driven clusters move as one, and how to diversify the risk that actually matters.
The standard advice is "diversify — don't put all your eggs in one basket." The story is half-right and quietly dangerous, because most beginners diversify the names without diversifying the risk. They buy ten different stocks, feel protected, and then watch all ten fall together on a single bad day — because the ten stocks were never ten separate bets. They were one bet wearing ten tickers.
A more accurate frame: diversification only works if your positions are uncorrelated — if they move independently. When positions are correlated, owning more of them doesn't reduce risk; it concentrates it while feeling safer, which is the worst combination. This chapter defines correlation, explains why narrative-driven clusters move as one, and shows how to diversify the risk that actually matters.
The TL;DR. Correlation measures how much two stocks move together, from +1 (lockstep) to −1 (opposite) to 0 (independent). Real diversification needs low correlation between holdings. Stocks in the same theme — the same bubble — often correlate near +0.8, so holding eight of them is barely more diversified than holding one. The risk you must watch is your correlated exposure, not your number of tickers.
What correlation actually measures
Correlation is a single number describing how two stocks' price moves relate:
- +1.0 — perfect lockstep. When one rises 2%, the other rises 2%. Identical bets.
- 0 — independent. One's move tells you nothing about the other's. True diversification.
- −1.0 — perfect opposition. When one rises, the other falls. A natural hedge.
Most stocks sit somewhere between 0 and +1. Two random companies in unrelated industries might correlate around +0.3 — loosely linked by the overall market. Two semiconductor names riding the same AI narrative might correlate +0.85 — nearly the same stock for trading purposes.
The number isn't fixed. Correlations rise in crashes: in a panic, forced selling hits everything at once, and stocks that normally move independently suddenly all fall together. This is the cruelest property of correlation — your diversification works in calm markets and evaporates exactly when you need it, in the crash. Plan for the crash correlation, not the calm one.
Why clusters move as one — the narrative link
Chapter 6 covered narrative as a price force. Correlation is its measurable fingerprint. When a narrative dominates — AI, EVs, nuclear, crypto — money floods toward every stock associated with it, and the whole group rises and falls on the narrative's headlines rather than each company's results.
That's why a single AI headline can move a dozen chip names the same direction on the same day, profitable and unprofitable alike. They're not trading on their own fundamentals; they're trading as a cluster on a shared story. Their correlation spikes toward +1 because the dominant driver is identical.
This is the entire reason QA is built around bubbles, not just stocks. A QA bubble is a cluster of stocks that measurably move together — defined by correlation, not just by sector labels or vibes. The semiconductors bubble groups names that trade as one because the same narrative drives them all. If you hold five names from one bubble, you don't have five positions — you have one bubble-sized bet. QA's correlation tool shows the actual pairwise numbers so you can see your real exposure instead of guessing.
The diversification illusion
Picture a beginner who "diversifies" by buying eight AI-adjacent stocks: a chipmaker, a cloud name, a data-center play, an AI software company, and so on. Eight tickers, eight charts, feels like a spread portfolio.
It isn't. All eight ride the AI narrative, so all eight correlate near +0.8. On a good AI day, all eight rise — exhilarating. On a bad one, all eight fall together — and the sizing math from chapter 7 turns brutal: eight positions each "risking 1%" aren't eight independent 1% bets. Because they move together, a bad day for the narrative is closer to a single 8% hit. The trader took on the concentrated risk of one big bet while believing they'd spread it across eight.
This is why chapter 7 insisted on capping aggregate risk, especially within a cluster. Counting positions is meaningless; counting correlated exposure is everything.
How to diversify the risk that matters
Real diversification means holding positions with genuinely low correlation to each other:
- Across themes, not within one. A semiconductor name and a consumer staple and a utility are far less correlated than three chip names. Spreading across bubbles — or into stocks that belong to none — is real diversification.
- Across direction. Holding some longs and some shorts, or a hedge, can push portfolio correlation toward zero. (Advanced — but the concept matters now.)
- Across asset type. Stocks, bonds, cash, commodities have historically lower correlations with each other than stocks do among themselves.
The practical beginner move is simpler than it sounds: before adding a position, ask whether it's correlated with what you already hold. If you own three chip stocks and want a fourth, you're not diversifying — you're adding leverage to a bet you already have. A genuinely uncorrelated fourth position is worth more to your risk profile than a fourth chip name, even a "better" one.
Correlation cuts both ways
Correlation isn't only a risk — it's information. QA's research thesis is that correlation often beats narrative for understanding what's really happening: when a stock's correlation with its cluster breaks down, something genuine has changed about that specific company, independent of the story. A name that stops moving with its bubble is sending a signal the headline doesn't carry. That's an edge — and it's why measuring correlation is a tool, not just a warning. The full argument is in Why correlation > narrative in thematic investing.
What to watch as you start
- Your real exposure, not your ticker count. Ten correlated names is one bet. Map your holdings to their bubbles and you'll usually find you're far more concentrated than you felt.
- Cluster concentration. If most of your positions live in one bubble, a single narrative shift moves your whole account. Cap how much correlated risk you'll carry at once.
- Rising correlation in stress. In a sell-off, expect normally independent positions to fall together. Size for the crash correlation, not the calm-market one.
- Correlation breakdowns. When a stock decouples from its cluster, that's a genuine signal about that name — sometimes the cleanest one available.
See the live numbers on QA's correlation tool and map your names to clusters on /bubbles. The next chapter confronts the hardest truth in trading: why most strategies that look profitable are actually curve-fit illusions — and how to tell.
Next in this series: Why most strategies fail — curve-fitting, out-of-sample testing, and the validation bar that kills 46% of backtests.
Go deeper: Why correlation > narrative in thematic investing.
See it live: /correlation for the matrix, /bubbles for the clusters. Live correlation shifts are part of /pro.
QuantAbundancia is educational research. Nothing here is investment advice. See /disclosures.
Related bubbles
Get the daily digest.
One email a day · alerts + bubble shifts + new research. Free during beta.
No spam. One email per day max. Telegram alerts coming with the paid tier.