Energy Is 2026's Best Sector and Institutions Were Still Selling It in Q2

Written August 27, 2026. Prices below are as of that date; crude has moved substantially since. The positioning data is drawn from Q2 filings and does not change.

Energy was the best-performing sector in the US market in 2026. As of late August, XLE was up roughly 30% year-to-date while the S&P had ground out a fraction of that, WTI was trading around $81 and Brent near $87, and the EIA's August outlook had Brent averaging about $85 through the third quarter. Occidental posted its largest quarterly profit since 2022 on August 5.

Now look at what institutions were doing about it. Aggregate shares of XLE held by 13F filers at the end of June were down about 6.4% from the end of March. Sixty-three filers opened a new position in the fund during the quarter; sixty-four closed one. That is not a sector being chased. That is a sector being traded around by people who mostly do not own enough of it.

How the underweight got so extreme

This did not start in 2026. Going into the year, fund manager surveys had energy as the single most underweight sector in the market, with oil and gas allocations sitting close to two standard deviations below their twenty-year average. Three years of AI-led index returns had made every dollar in a cyclical commodity producer an active bet against the thing that was working. Funds that held energy through 2023 and 2024 underperformed, explained themselves to allocators, and eventually stopped.

Positioning that stretched does not need good news to reverse. It needs the absence of bad news. What it got instead was a genuine supply disruption — constrained Strait of Hormuz transits, roughly 0.6 million barrels a day of production the EIA does not expect back until well into next year — arriving on top of a shareholder base that had already been cleared out.

The lag problem, again

The Q2 filings that landed on the August 14 deadline describe portfolios as of June 30. Much of the crude move and all of the second-quarter earnings confirmation came after that snapshot. So the honest read of the data is not "smart money is selling energy." It is "smart money had not yet bought energy as of two months ago, at prices that no longer exist."

That distinction matters more in energy than almost anywhere else, because commodity equities re-rate fast. A fund that added Oxy or EOG in July shows up in filings on November 14. By then the trade is a quarter old. Our flow tracker is built for exactly this problem — reading the direction and pace of institutional accumulation rather than treating any single quarter's holdings table as a live recommendation.

Where the exposure actually was

Underweight in aggregate does not mean absent everywhere. Point72 added to Occidental and EOG Resources. Baupost has carried Pioneer. Richard Perry's firm built a sizable Williams position. Carl Icahn's energy concentration, which we wrote about in July, has been one of the better-aged bets in his book.

The pattern in the filings is that energy conviction clusters in a small number of managers who hold it in size, rather than being spread thinly across the field. That is the opposite of how a crowded trade looks. When you check the overlap tool, the AI and hyperscaler names show dozens of top funds sharing the same tickers; the energy names show a handful of funds with unusually large weights. Concentrated conviction and crowding are different risks, and the second one is the one that unwinds violently.

Reading it as a macro signal

The more interesting question is what an energy underweight implies about everything else in these portfolios. Being short energy exposure is, functionally, a bet on disinflation and on supply staying boring. That was a reasonable bet for three years. It is the same bet embedded in long-duration growth positioning, in rate-cut assumptions, and in the multiple attached to anything valued off cash flows a decade out.

If the oil move persists, the pain is not confined to the energy sleeve. It shows up as an input-cost problem in transports, industrials and consumer discretionary, and as an inflation problem for the rate path that the rest of the book depends on. That correlation is why we track sector positioning alongside stated macro views in the macro consensus view — the sector a fund refuses to own often tells you more about its worldview than the sector it talks about.

What to watch

Two things. First, whether the November filings show broad new energy positions or just existing holders adding — broad accumulation into a 30% move is late-cycle behavior, incremental adds by long-time holders is not. Second, whether the funds adding energy are trimming AI exposure to fund it, which would make this a rotation rather than a sleeve adjustment.

The Q2 data cannot answer either question. It can only tell you the starting point: institutions entered the best sector of 2026 owning very little of it, and spent the quarter owning slightly less.

---

Data sourced from public SEC 13F filings. Educational research only — not investment advice.