If the EIA cannot hold a price call for a month, no operator should be running a capital plan that requires one to be right.
On July 7, the Energy Information Administration published its Short-Term Energy Outlook. The forecast had been completed six days earlier, on July 1. It cut the 2026 Brent projection to $82 per barrel from $95 — a 14 percent reduction in a single month — and took the 2027 projection down to $65 from $79, an 18 percent cut.
The reasoning was explicit and, on the information available, entirely sound. The United States and Iran had signed a memorandum of understanding on June 18 to end the conflict and reopen the Strait of Hormuz. Traffic through the strait was increasing. Shut-in production was expected to return. The agency raised world production expectations by 3.5 percent, lifted its OECD inventory projection by nearly 15 percent, reduced the third-quarter Brent forecast by $27 per barrel, and concluded that the market would revert to its pre-conflict state of oversupply.
The forecast was three weeks old.
This is not a story about the EIA getting it wrong. It is a story about what forecasting can and cannot do in the current environment — and what operators should build instead.
It is worth being precise about the point, because the easy conclusion is the wrong one. The easy conclusion is that the EIA is unreliable and operators should trust their own analysis instead. That is exactly backwards. The EIA has more data, better models, and fewer incentives to be wrong than any private forecaster in energy. If their number had a three-week shelf life, yours has a shorter one.
The correct conclusion is structural. When the interval between a material price signal and its complete reversal is measured in weeks, the binding constraint on capturing value stops being analytical accuracy and becomes execution speed. Being right matters less than being able to act on something you did not predict.
Look at what the year has actually done. WTI opened in an unremarkable range, pushed above $110 in March as the conflict escalated, collapsed into the mid-$60s in early July as the strait reopened, and recovered to roughly $90 by the end of the month — up about 26 percent in thirty days and 39 percent against a year ago. That is a swing of more than $45 inside four months, in both directions, with the reversal in each case arriving faster than a conventional financing process can complete.
Submitted mid-March. Funded late May. Completed into a falling market with the window already closed.
Funded inside the window. Completed into the strength. Booked production at the top of the tape.
This is why the standard framing of financing speed as a convenience is so badly miscalibrated in energy right now. A fifty-day differential between a conventional close and one with a capital provider who evaluates oilfield assets internally is not an inconvenience when the price signal reverses in three weeks. It is the entire difference between participating and reading about it afterward.
And the cost does not show up as an interest rate. It shows up as the completion that happened at $67 instead of $95, or the lift program that was funded after the wells had already declined, or the gathering line that was not built in time to move production while the differential was favorable. None of that appears in a rate comparison. All of it dwarfs the rate.
There is a version of capital planning that acknowledges all this and still fails, and it deserves mention because it is common. An operator concludes that volatility is the environment, builds wider price bands into the model, hedges more aggressively, and stress-tests the program across a range. All sensible. None of it addresses the actual problem, which is that when the favorable end of that range materializes, the operator still cannot fund fast enough to act on it.
Wider bands improve the analysis. They do nothing for the execution. And in a year where the range has been $45 wide and traversed twice, the execution is the whole game.
The second half will not be more predictable than the first. The conflict is active, the memorandum that briefly ended it has already failed once, and the EIA continues to project a return to oversupply — which may well be right on a two-year view and tells an operator nothing useful about the next eight weeks.
The operators who capture the next window will not be the ones with the best price view. Nobody has a reliable price view this year — including the people whose job it is.
First National Capital Corporation funds completion capital, production equipment, and field infrastructure on timelines measured in weeks — no covenants, $500K to $250M+, more than $4.5 billion funded. Start a conversation.