Utilities Flipped Direction 5 Days Straight. That's Not a Trend.
The average NASDAQ-listed utility stock reversed sign every session for a week while the 10-year yield rose 11bp. Here's why that's noise, not rotation.
Between September 28 and October 2, 2026, the average NASDAQ-listed utility stock did something a sector "trending" in one direction is not supposed to do: it reversed sign on every single trading session. Down 0.10% on Monday. Up 1.50% on Tuesday. Down 2.18% on Wednesday. Up 0.27% on Thursday. Down 1.54% on Friday. Five sessions, five flips. Over the same week the 10-year Treasury yield climbed from 5.17% to 5.28% — a real, directional move of 11 basis points. The sector most textbook-sensitive to that move couldn't hold a direction for even two days in a row.
| Session | Avg. NASDAQ utility stock | 10-year yield |
|---|---|---|
| Mon Sep 28 | -0.10% | 5.24% |
| Tue Sep 29 | +1.50% | 5.26% |
| Wed Sep 30 | -2.18% | 5.29% |
| Thu Oct 1 | +0.27% | 5.24% |
| Fri Oct 2 | -1.54% | 5.28% |
This is worth sitting with, because it's a clean test of a mechanism covered here before: utilities are rate-sensitive because their valuations run through a discount rate and their balance sheets are debt-heavy, so a rising 10-year yield should pressure the group. Three of the four day-over-day yield moves in this table do line up that way — yield up, utilities down (Wed, Fri), yield down, utilities up (Thu). But Tuesday breaks it: the yield rose 2bp and utilities still gained 1.50%. The mechanism is real; it just doesn't win every single day, and a reader who only saw Tuesday would have concluded the opposite relationship.
What this number actually measures
It matters, too, that this isn't the utilities sector ETF (XLU) return — it's the average return of individual NASDAQ-listed stocks FMP tags as Utilities, each counted once regardless of size. That's a different construction than a cap-weighted sector ETF, where one or two large-cap names can carry the whole sector's reported move, and it's a narrower universe than "every US utility" since several large utilities list on the NYSE rather than NASDAQ. Both of those are reasons this figure can diverge from the "utilities were up/down today" headline you'd get from a sector ETF quote — not because one number is wrong, but because "the utilities sector's return" is at least two different calculations depending on which stocks get counted and whether size matters. Checking a sector by its full constituent list rather than a single ETF print — which a screener pass filtered to one sector makes easy — is the only way to know which version of "the sector" a headline is actually describing.
Healthcare's quieter three-day slide
Utilities' flip-flopping isn't the only pattern in the same data. Healthcare moved in the opposite style: up 1.05% Monday, up 0.71% Tuesday, then three straight negative sessions — down 0.26% Wednesday, down 2.25% Thursday, down 1.06% Friday. That's a real short streak, not noise in the alternating sense Utilities showed. But it's harder to pin to a single clean mechanism the way the Utilities-and-yields relationship is; healthcare's average stock return can move on drug-pricing headlines, a single large biotech readout, or an unrelated clinical-trial result just as easily as on anything macro. The honest read is that a three-day streak is a pattern worth watching, not yet a rotation call, and not one with an obvious one-line explanation the way a rate-sensitive sector's move usually has.
Noise, streaks, and what a rotation call actually requires
None of this is a criticism of the rate-sensitivity mechanism or a claim that healthcare is quietly rotating out of favor. It's a reminder of the gap between a mechanism being real and a mechanism being the dominant driver on any given day. Dozens of forces hit an equal-weighted sector average in the same session — earnings from one over-represented subindustry, a single stock's specific news, index-fund rebalancing flows, option-expiration mechanics — and on most days several of them are pulling in different directions at once. A sector that's genuinely losing share of the market's capital does it over weeks and months, visible as a sustained drift in its index weight, not as a coin-flip pattern across five sessions.
That distinction is exactly what a live sector rotation model is built to isolate: not whether Wednesday was red or Thursday was green, but whether a sector's actual share of invested capital is climbing or falling once the day-to-day direction flips wash out. Utilities flipping sign five days running and Healthcare stringing together three red sessions are both real, checkable data points — and both are also exactly the kind of short-run pattern that looks meaningful in the moment and tells you almost nothing about where the next quarter's capital is actually headed. The way to tell the difference isn't to stare harder at one week's table; it's to watch whether the pattern survives past the week that produced it.