How to Read the EIA Short-Term Energy Outlook

Read the Short-Term Energy Outlook in three passes: the revision table first, then the world balance in Table 3a, then the U.S. detail in Tables 2 and 4a. The October 6, 2026 edition is a clean example. EIA raised its 2026 Brent forecast to $96 a barrel from $91 a month earlier, and the reason sits in the balance tables, not in the headline.

The STEO, published by the U.S. Energy Information Administration, is the most widely quoted public forecast of oil, natural gas, electricity and coal markets for the next one to two years. It is free, monthly, and written by a statistical agency rather than a producer group, which is why journalists quote it and why readers tend to over-read it. A single edition contains hundreds of forecast series. Most people only ever see one of them: the Brent price. This guide works through the rest, using figures from the October 2026 edition checked on eia.gov on October 22, 2026.

Release timing and what "forecast completed" means

EIA posts its release calendar a year ahead. For 2026 the dates are January 13, February 10, March 10, April 7, May 12, June 9, July 7, August 11, September 9, October 6, November 10 and December 8. September 9 fell on a Wednesday, and the October edition doubled as the Winter Fuels Outlook. The published schedule for 2027 already runs from January 12 to December 7, with the Winter Fuels Outlook edition on October 13, 2027. EIA states that the report normally appears between 12:00 pm and 12:15 pm Eastern time.

Every edition also prints a second date that matters more than the release date: the day the forecast was completed. For the October 2026 edition that was October 1, five days before publication. Anything that happened in markets after October 1, a price spike, a pipeline attack, an OPEC+ meeting, is not in the numbers even though the report is dated October 6. When a news story says "EIA ignored" an event from the week of release, the completion date usually explains it.

The STEO sits in the monthly layer of the EIA calendar, below the weekly reports. Our guide to the EIA release schedule shows how the weekly petroleum and natural gas storage reports, the Petroleum Supply Monthly and the STEO fit together. The short version: weekly data tells you what just happened, the monthly surveys correct it, and the STEO projects it forward.

What one edition contains

The web version of the October 2026 edition opens with a forecast overview, a short "Notable Forecast Changes" table, and an overview summary table. Below that sit the report sections: global oil markets, petroleum products, natural gas, electricity with coal and renewables, and a section on the economy, weather and carbon dioxide emissions. Each section has its own page with charts. Then come the data tables, the interactive data viewers, and the archive of previous editions.

Not every month carries the same depth. An EIA explainer from April 2020 describes the January and July editions as the comprehensive ones, covering the whole energy system in full, and the other months as updates with forecast highlights and a review of near-term crude, product and natural gas markets. The April edition is paired with a Summer Fuels Outlook and the October edition with a Winter Fuels Outlook. If a month's narrative feels thin on a topic, it may simply not be a comprehensive month; the tables are still updated.

The same explainer says the forecast rests on economic models whose inputs include EIA's own monthly surveys (the Petroleum Supply Monthly, Natural Gas Monthly and Electric Power Monthly, among others), weather data from NOAA, spot and futures prices, and macroeconomic indicators. Analysts then adjust the model output for events the equations cannot see: regulatory changes, weather disruptions, supply shocks, pipeline constraints. That judgment layer is why a STEO forecast can jump when the world changes, and why the text around the numbers matters.

The 25 tables, and the six oil readers use most

The STEO data page lists 25 tables. Oil readers spend most of their time in six of them. The table below gives the official titles as listed on eia.gov and what each one is good for.

The remaining tables cover hydrocarbon gas liquids and refinery balances (4b), biofuels (4d), coal (6), five electricity tables (7a to 7e), renewables (8a), and three macro and weather tables (9a to 9c). They matter for oil readers mostly as assumptions: Table 9a carries the GDP path that drives demand, and Table 9c the heating and cooling degree days that drive seasonal fuel use.

Table 10a deserves a note. Readers who know EIA's Drilling Productivity Report will recognize its metrics here. The background on how those metrics are built is in our walkthrough of the drilling productivity data, and it applies to the STEO version too.

Start with the revision table, not the headline

Each edition now opens with a small table of notable forecast changes: the current forecast, the previous month's forecast for the same period, and the percentage change. It is the most efficient minute a reader can spend on STEO day, because a forecast level on its own says little; the direction and size of the revision say what EIA learned in the past month. Here is the October 2026 version, as published.

Source: EIA, Short-Term Energy Outlook, October 6, 2026, Notable Forecast Changes, checked October 22, 2026.

Two things stand out before any analysis. First, the later year moved more than the current one: the 2027 Brent forecast rose by $10 while 2026 rose by $5. That is a statement about persistence. EIA did not just mark up the present; it now expects tight conditions to last longer. Second, retail fuel forecasts moved far less than crude in percentage terms, which is normal: taxes and distribution costs are fixed cents per gallon and dilute any crude move, a pattern laid out in our gasoline price breakdown.

Worked example: what changed between September and October 2026

A revision is only useful if it can be decomposed. The October edition gives enough pieces to do that for Brent, using only numbers EIA printed.

Step 1: check the printed percentages. Dividing the rounded figures gives $96 / $91 = 1.055, a 5.5 percent increase, while the table prints 6 percent. For 2027, $84 / $74 = 1.135, or 13.5 percent, printed as 14 percent. Gasoline for 2026 works out to $3.91 / $3.84 = 1.8 percent but is printed as 2.0 percent. The differences are not errors: EIA computes the change from unrounded forecasts and then rounds the displayed prices to whole dollars or cents. The lesson for anyone building a revision history is to download the table values, not copy the rounded headline numbers, or small revisions will look larger or smaller than they were.

Step 2: find where in the year the change landed. The global oil page of the same edition says EIA now expects Brent to average $105 a barrel in the fourth quarter of 2026, $14 above the previous month's forecast, and that Brent averaged $114 in September 2026 with a daily high of $131 on September 15. So the September edition carried roughly $91 for the fourth quarter, the same as its full-year figure.

Step 3: back out the first eight months. Treat the annual average as a simple average of twelve monthly averages (EIA's own calculation is day-weighted, so this is an approximation). Twelve months at $96 sum to $1,152. September contributes $114 and October to December contribute three months at $105, or $315. That leaves $723 for January through August, an average of about $90 a barrel. Those eight months are history, so they barely change from one edition to the next. Nearly all of the $5 upgrade to 2026 therefore sits in the last four months: September came in far above what the September edition expected, and the fourth-quarter path was raised by $14.

PetroEyes arithmetic on EIA-published figures. Approximate, because it uses rounded values and a simple monthly average.

Step 4: read the reasons EIA gives. The edition names them: attacks on Saudi Arabia's East-West pipeline, continuing withdrawals from global crude inventories, and tight diesel markets that raise refiners' demand for crude. Each one is checkable elsewhere. Inventory withdrawals show up in Table 3a, diesel tightness in the distillate stock data covered in our guide to the weekly petroleum status report, and the pipeline risk in the supply narrative. A revision whose stated reasons do not show up in the tables deserves skepticism; this one is internally consistent.

The world balance behind the price call

The STEO's price forecast is not an independent guess. It is tied to a supply and demand balance, and the balance is where a reader can test whether the price path makes sense. The October 2026 global oil page gives these annual levels, in million barrels per day.

Source: EIA, Short-Term Energy Outlook, October 2026, global oil markets section, checked October 22, 2026. The difference row is a simple subtraction and ignores the statistical adjustments EIA carries in Table 3a.

Read across the rows and the price path explains itself. In 2025 the world produced more than it consumed, and Brent averaged $69. In 2026 OPEC output falls by about 5.4 million barrels a day in EIA's numbers, consumption falls too but by less, and the world draws on its stocks: EIA says global inventories fell by an average of 1.9 million barrels a day in the third quarter of 2026 and expects a further 0.7 million a day decline in the fourth. Brent averages $96. In 2027 production recovers faster than consumption, the simple balance swings to a surplus of about 5 million barrels a day, and the price forecast falls through the year to $74 in the fourth quarter.

The production collapse in 2026 is EIA's estimate of shut-in output in the Middle East: the edition puts it at a peak of 10.9 million barrels a day in May 2026, 5.8 million in August and 4.8 million in September, and assumes 4.5 million on average in the fourth quarter, with flows constrained around the Strait of Hormuz and partly restored through convoys, bypass routes and ship-to-ship transfers. That single assumption carries much of the forecast. If shut-ins fall faster than 4.5 million a day, the 2026 draw shrinks and the price path should come down; if they rise, the reverse. When a forecast hinges on one input, track that input more closely than the forecast itself.

A further check: a large 2027 surplus with a still-elevated $84 average is not a contradiction. Stocks drawn down in 2026 need to be refilled before a surplus pushes prices back to where they were in 2025, and the average for 2027 includes a first half that EIA still prices near $87 (its second-quarter 2027 figure). The surplus shows up as a falling path within the year, not as a low annual average.

The U.S. lines: Tables 2 and 4a

For U.S. readers the most quoted numbers come from the summary table, which draws on Tables 2, 4a and 5a. The October 2026 edition prints these annual figures.

Source: EIA, Short-Term Energy Outlook, October 2026, overview summary table, checked October 22, 2026.

The contrast is the lesson. A 39 percent jump in the Brent forecast between 2025 and 2026 comes with only a 0.2 million barrel a day rise in U.S. crude output. Higher prices do not turn into U.S. barrels within the same year: rigs have to be added, wells drilled and completed, and companies set budgets on multi-year price views. The 2027 rise to 14.3 million is where EIA lets the price response show. Our rig count explainer covers the lag between price, rigs and production that this table compresses into two numbers.

Table 4a is also where to read the U.S. as a trading nation: crude imports and exports, product exports, and the stock lines for commercial crude and the Strategic Petroleum Reserve. The weekly report gives the same categories with a few days' lag; the STEO projects them by month. A reader comparing the two should remember that the weekly figures will be revised by the Petroleum Supply Monthly, and the STEO history columns already incorporate those revisions.

Where the STEO sits next to OPEC and the IEA

Three monthly forecasts dominate the oil market: EIA's STEO, the International Energy Agency's Oil Market Report, and OPEC's Monthly Oil Market Report. The International Energy Forum publishes a side-by-side comparison each month. Its editions through 2026 show how far apart the three were on 2026 demand.

Source: International Energy Forum, Comparative Analysis of Monthly Reports on the Oil Market, June, August, September and October 2026 editions, checked October 22, 2026.

Two readings follow. All three agencies revised the same way for four months running, toward weaker 2026 demand, which tells you the direction of the data even though the levels differ. And the gap in levels is large: in October, EIA's 2026 world demand of 102.4 million barrels a day sat 3.3 million below OPEC's 105.7 million. That gap is bigger than most annual growth figures, which is why quoting any one agency's demand number as "the" demand figure misleads. The companion guide on reading OPEC's Monthly Oil Market Report explains the producer group's side of the same comparison.

Horizon and assumptions: what the model is told to take as given

Each STEO forecasts through the end of the following calendar year. That makes the October 2026 edition a 15-month forecast and the January 2027 edition, when it comes, a 24-month one. The near months are anchored by data already collected; the far months are anchored by assumptions. A sensible reading weights them that way.

The assumptions that matter most for oil are:

• Disruption paths. In 2026 the dominant assumption is the size of Middle East shut-ins, stated in the text with a fourth-quarter figure.
• OPEC+ policy. EIA forecasts what it expects producers to do, not what they announce. When those differ, the text usually says so.
• Macroeconomic growth. The GDP path in Table 9a drives the demand side. The October 2026 summary table shows U.S. GDP growth of 2.1 percent in 2026 and 2.6 percent in 2027.
• Weather. Heating and cooling degree days in Table 9c are built from NOAA weather data, and winter fuel forecasts swing with them.
• Prices as an input and an output. The STEO forecasts prices, and the same edition's demand figures have to be consistent with them. A reader should not assume demand is held fixed while the price path moves.

EIA has, in some years, published a separate market prices and uncertainty supplement that draws on options prices to show the range of outcomes the market itself was pricing. A reader who wants that perspective should look for it on the current STEO page rather than relying on older editions; this guide could not locate a current version and does not quote one.

Common misreads

• Treating the annual average as a year-end price. A $96 annual Brent average is consistent with a path that starts near $90 and ends above $100. The quarterly and monthly columns in Table 2 show the shape; the annual figure hides it.
• Comparing two editions on different completion dates as if one were wrong. A forecast completed on October 1 cannot reflect an event on October 3. The revision lands next month.
• Reading the history columns as fixed. Past months in the STEO are updated as EIA's monthly surveys come in. Last month's "history" can change.
• Mixing crude-only and total-liquids figures. World production in Table 3a is liquid fuels, which include natural gas liquids, biofuels and refinery processing gains. U.S. crude production in Table 4a is crude oil only. Adding or comparing them directly double counts or understates.
• Quoting a percentage change from rounded figures. As the worked example showed, $96 over $91 computes to 5.5 percent while EIA prints 6 percent. Use the downloaded table.
• Treating the STEO as a market forecast. It is a balance forecast with prices attached. Futures prices on the same day can sit well above or below it, and neither is obliged to converge to the other.

A routine for STEO day

1. Note the forecast completion date and list any major events between it and the release.
2. Read the notable forecast changes table. Write down the direction and size of each revision.
3. Open the global oil page and find the reason given for each price revision.
4. In Table 3a, compare production with consumption by quarter. Confirm that the implied stock change matches the direction of the price path.
5. Find the one assumption the forecast leans on most (in 2026, Middle East shut-ins) and record its value for next month's comparison.
6. Check Table 4a for U.S. production and net trade, and Table 2 for the monthly price shape.
7. Compare the demand level with the latest IEA and OPEC reports, using the IEF comparison for a quick side-by-side.
8. Save the downloaded tables. A revision history built from your own saved files is more useful than any single edition.

The PetroEyes forecasts page tracks agency outlooks alongside each other, and the oil price forecast topic hub collects the related pages on inventories, rigs and supply that feed into any outlook.

Questions readers ask about the STEO

The natural next step is the other side of the comparison: how to read OPEC's Monthly Oil Market Report, which shows how the producer group builds a demand figure 3.3 million barrels a day above EIA's and why its production tables come in two versions.

Sources, all checked October 22, 2026: U.S. Energy Information Administration, Short-Term Energy Outlook, October 6, 2026 (forecast completed October 1, 2026; notable forecast changes and overview table); STEO global oil markets section (world production, consumption, OPEC and non-OPEC supply, inventory changes, Middle East shut-ins, quarterly Brent path); STEO release schedule; STEO data tables list; EIA Today in Energy, April 8, 2020 (how the STEO is produced, comprehensive editions, seasonal supplements, forecast horizon); September 2026 STEO archive PDF. International Energy Forum, Comparative Analysis of Monthly Reports on the Oil Market: June 2026, August 2026, September 2026, October 2026. Differences, implied balances and the January to August Brent average are PetroEyes arithmetic on the published figures.