Sector Rotation: Reading Where the Money Is Going

You read sector rotation by comparing each sector's performance against the broad market, over more than one time horizon, and watching which groups are gaining relative strength and which are losing it. Money rarely leaves the market outright — it moves between sectors — and the direction of that movement tells you whether risk is being put on or taken off.

That's the whole skill in one sentence. The rest of this page is how to do it by hand: what to measure, what the common patterns tend to mean, and — just as important — what a rotation read cannot tell you. I use this read every morning before the open. Further down I'll show you a dated example where it worked and a dated example where it didn't, both from our own logged calls, because both happen and you deserve to see both.

What is sector rotation, actually?

The US market breaks into eleven sectors: technology, consumer discretionary, communication services, financials, industrials, materials, energy, consumer staples, utilities, health care, and real estate. Each one has a widely traded fund that tracks it, so each one can be measured against the index with a single division.

On any given day, money is being added to some of those groups and pulled from others. The index nets all of it out, which is exactly why the index alone misleads. The market can print a flat, boring day while a violent rotation happens underneath it — one sector bid hard, another sold just as hard, the two canceling each other at the index level. If you only watch the index, you never see it. That's the same blind spot market breadth covers from a different angle: the index tells you the score, not the game.

Rotation matters for a second reason: individual stocks mostly move with their group. A mediocre stock in a sector being accumulated gets carried. A good stock in a sector being distributed gets dragged. Reading the group first isn't optional homework — it's most of the context around any single trade you'll ever consider.

How do you measure which sector is leading?

One ratio. Divide the sector's price by the benchmark's price and plot the result as a line. When the line rises, the sector is outperforming the market. When it falls, it's underperforming. That's relative strength — not the RSI oscillator, just the ratio. Note what this means: a sector's price can be falling while its relative strength rises, because it's falling less than everything else. That's still information. Money hides in the strongest house on a bad street.

Two habits make the ratio useful instead of noisy:

1. Watch it on more than one horizon. A sector can lead on the week and lag on the quarter. Look at roughly a week, a month, and a quarter. When the horizons agree, the read is strong. When they argue, the rotation is either young or failing — you don't know which yet, and it's fine to say so out loud.
2. Rank, don't admire. The question is never "is technology up?" It's "where does technology sit relative to the other ten?" Sort all eleven sectors by relative strength and read the top three and the bottom three. The middle five are usually noise.

Full disclosure on where I stand: our desk runs a computed version of this for its own internal use — it isn't published, it isn't for sale, and I'm not going to publish the arithmetic either; the computed output is the desk's own work. But here's the honest part: you don't need it to do the read. The ratio line plus two or three horizons, checked once a day, gets you most of the picture by hand. What the computed version buys is consistency and speed, not access to a secret.

What do the common leadership patterns mean?

Every row in this table is a tendency, not a law. The market is under no obligation to respect any of them on a given day.

One addition sharpens every row: volume. A sector move on heavy participation is a different animal from a drift, and relative volume is how you tell them apart. The sector rolling over on high volume — like the dated example further down — is the version worth respecting.

What are the four phases of a rotation?

A sector doesn't jump from leader to laggard in one step. It travels, roughly in order, through four phases:

1. Improving. Still underperforming, but the relative strength line has stopped falling and started to climb. Nobody is talking about it yet.
2. Leading. Outperforming and still strengthening. The crowded, comfortable phase.
3. Weakening. Still ahead of the market, but the ratio line is flattening and starting to roll. The headlines are still bullish. The ratio isn't.
4. Lagging. Underperforming outright. Now it's obvious — and mostly priced in.

Then the loop restarts. The practical use is blunt: attention belongs in phases one and two, caution in phase three, and phase four is a place to wait for phase one to begin again — or, in a heavy tape, where short-side candidates concentrate. The mistake this framing prevents is the most common one in retail rotation-watching: discovering a sector in phase three, when its story is loudest and its ratio is already bending.

How do you use rotation before trading a single stock?

With an alignment check. Before considering any name, ask three questions in a stack: is the stock stronger than its sector? Is the sector stronger than the market? Is the market itself pointed in the trade's direction? When all three line up, the trade has its context in order. When they don't, it's fighting a current you can't see on its own chart. That first question has a sharper version, too: comparing when a stock's own cycle turn arrives against when the index's turn arrives often shows relative strength or weakness before the sector-level read catches up to it.

That's the concept. Here's what it looks like in real, dated, scored use — the miss first:

July 9, 2026 — logged on our desk, scored a miss. AMD set up long on the morning work, and the alignment stack was complete: the stock qualified, the sector supported it, and the broad market was positive. It worked for exactly one session — up about 2% the next day — then gave all of it back and more. By July 17 it sat roughly 9% below the call. Full alignment, clean setup, miss. It's on our scoreboard as a miss.

And the other side, from the same session:

July 9, 2026 — logged on our desk, scored directionally right. The other read that morning was group-first rather than name-first: financials had broken down the prior session and then reversed higher, which is a different animal from a breakdown that keeps going, and strong financials were called the tell for the broader market. The group did hold and extend — JPM finished about 3.7% above the call-day close two weeks later. The record grades it directionally right rather than a clean win, and the reason matters: it was flagged as an observation with no entry, no target and no timeframe attached, so there was nothing to score precisely against. A group-level read that works is still not a trade unless someone wrote down what would make it wrong.

Two dated calls from one session, one miss and one directional hit. That's what this tool actually looks like in use. If someone shows you only one of those two kinds of outcomes, they're marketing, not teaching.

What doesn't a rotation read do?

The honest limits, stated out loud:

• It doesn't time entries. Rotation is slow. It tells you which way the wind is blowing, not which minute to step outside.
• It degrades in chop. When leadership churns week to week, the read turns to noise. Recognizing that it's currently noise is itself the read — and often the argument for standing down.
• It knows nothing about single-company news. The strongest sector in the market will not save a name with its own problem, and a lagging sector sometimes contains the day's best stock-in-play.
• It will sometimes simply be wrong, as the first dated miss above shows. Nothing on this page, or this site, is a promise about results — we never guarantee outcomes.
• It runs on a different clock than the macro backdrop. This page is a days-to-weeks read. Where the broader economy sits in its growth cycle is a quarters-to-years read that has, historically, tended to favor similar leadership groups for similar reasons — worth knowing as a slow-moving prior, never a substitute for this page's daily read.

Where it fits: for me, the rotation read is one fixed slot in the pre-open block — part of a structured morning, not something I stare at all session. On our desk it then sits next to a second, independent question — where price stands inside its larger cycles — and that pairing of money flow with timing is the core of the method.

Questions traders ask

How often should I check sector rotation?

Once a day, before the open, is enough. The signal moves in days and weeks, so checking it intraday mostly adds noise and temptation. If you swing trade or hold positions for weeks, a weekly review of the eleven ratio lines covers you. The discipline that matters isn't frequency — it's consistency: the same check, at the same time, so changes stand out.

Can sector rotation predict market direction?

Not by itself, and I'd be careful with anyone who says otherwise. Rotation describes where risk appetite is right now. Shifts in the pattern — defensives quietly improving while leaders fade — can front-run trouble in the index, but "can" is the honest verb. Treat it as one piece of evidence, weigh it alongside breadth and volume, and let price settle the argument.

Do I need paid software to read rotation?

No. A ratio chart of a sector fund against the index is free on essentially every charting platform, and eleven of them take a few minutes a day to scan by hand. A computed ranking buys consistency, multiple horizons blended cleanly, and time saved — which is why our desk computes one for its own internal use — but it doesn't buy access to a secret. The concept on this page is the load-bearing part, and it's free.

Keep reading