Last month this site argued that anyone being pushed to sign a fixed natural gas contract before EIA’s October outlook was being asked to decide with less information than would shortly exist. That outlook has now published, and so has the Winter Fuels Outlook that comes with it. The information exists. It does not say what the people selling contracts are likely to tell you, and it does not say what you might assume either.
The three most-quoted forecasts for this winter disagree about which direction household gas bills are going.
| forecast | published | natural gas this winter |
|---|---|---|
| NEADA (National Energy Assistance Directors Association) | 14 September 2026 | up 5.8% |
| EIA Winter Fuels Outlook 2026–27 | 6 October 2026 | roughly flat; Northeast down 1% |
| American Gas Association | Winter Heating Outlook, September 2026 | down 14% on space heating |
Three credible organisations, a spread of twenty percentage points. If a salesperson quotes you one of these as *the* forecast, they have chosen it.
Why the forecasts disagree
They are measuring different things, over different periods, with different methods.
EIA models consumption and residential price for homes whose main heating fuel is gas. This winter it expects households heating with natural gas to use about 2% less on slightly milder weather, against residential prices about 1% higher, which nets out close to flat. Averaged across Northeast households whose main heating fuel is gas, it puts the base case at $868, about 1% below last winter. That is a regional average, not an estimate for any one house.
NEADA reports a household’s total heating cost rather than one fuel’s expenditure, and it published on 14 September, three weeks and one STEO edition before EIA’s figure. Its stated driver is rate cases: it counts more than 275 electric and natural gas utilities that have raised rates, had increases approved, or proposed them since 2025, which it estimates could add more than $101 billion to customer bills through 2028. That is NEADA’s explanation of its own number, and it is worth knowing which of the three forecasts rests on it.
The AGA figure is the narrowest of the three: it covers space heating only, not the water heating, cooking and drying that the same meter also serves.
So all three can be internally honest and still point different ways. Ask a salesperson which forecast they are quoting and what it covers.
The Northeast number the September piece said to wait for
That piece made a point of distinguishing a printed date from a pattern: EIA prints its STEO schedule, but the Winter Fuels Outlook arriving in October was a two-year habit rather than a commitment. The habit held. It published 6 October 2026, and it carries the regional detail that makes it worth waiting for.
EIA’s forecast for this winter, averaged across Northeast households that heat with natural gas:
| forecast average, gas-heated households | vs last winter | |
|---|---|---|
| base case | $868 | −1% |
| 10% colder than base | $931 | +6% |
| 10% warmer than base | $828 | −6% |
That band between the colder and warmer cases is what a fixed rate is actually sold against. It is a band rather than a crisis. Whether paying a premium to remove that much variability is worth it is a judgment about your own tolerance, not a market call, and it is a smaller band than most pitches imply.
The storage cushion, a month later
The September piece leaned on a regional storage argument: that the East was entering winter without the inventory cushion other regions had. That argument has been overtaken, and the honest thing is to say so rather than quietly carry it forward.
EIA’s October chapter gives a national inventory figure, and the regional picture has since improved. As of the week ending 9 October 2026, working gas stocks in the East, Midwest, Mountain and Pacific regions were all above their five-year averages. Lower 48 inventories stood at 3,567 Bcf, and EIA’s October outlook forecasts about 3,850 Bcf at the end of October, roughly 2% above the five-year average entering the heating season.
Two caveats keep this from being an all-clear. The national surplus is narrowing, from 2.0% to 1.3% in a single week. And one region is genuinely short: South Central inventories sit 3.1% below their five-year average, with its salt-cavern storage 15.4% below. That is not New England’s problem directly, but South Central is where the flexible, fast-cycling storage lives.
On gas specifically: the cushion is adequate and thinning, not depleted. Anyone telling you otherwise should be asked for the week and the region.
Where the reserves are depleted, and why it is not gas
There is a depleted-reserve story this winter, it is serious, and it belongs to a different fuel.
The Strategic Petroleum Reserve holds crude oil, not natural gas. EIA’s weekly stocks series, released 15 October 2026, puts it at 282.983 million barrels in the weeks ending 2 and 9 October, unchanged between the two and its lowest since 1982. Reporting on the drawdown has raised questions about the condition of the salt caverns the oil is stored in, and about how far the reserve can be drawn before refilling it becomes harder.
This matters to Northeast households for one reason: heating oil. EIA’s October report has East Coast distillate inventories 32% below their five-year average, and forecasts them staying 20–30% below through the forecast period. NEADA projects heating oil costs up 31.3% this winter, by far the sharpest increase of any fuel, attributing it to crude that was trading near $72 a barrel before this year’s conflict and later moved toward $100 amid concern about shipping through the Strait of Hormuz.
Put a number on it. EIA’s weekly survey priced New York residential heating oil at $6.335 a gallon in the week of 12 October 2026, against a U.S. average of $6.138. Both exclude taxes, so a delivery ticket reads higher than either. EIA’s own Winter Fuels Outlook, published 6 October, forecasts the national price at $5.79 a gallon for November, the highest month in that winter forecast, and $5.26 across the season, which it puts at a 34% increase over last winter. Those forecasts include taxes, and the 34% is a change in price per gallon, not in what a household spends. The same table puts heating oil spending at $2,115 for the season, up 21%, which is the kind of number NEADA’s 31.3% is. The first two figures are surveyed averages, one for New York and one national; the last three are forecasts, and none of them is a quote for your house, which comes down to your supplier and your contract.
If your instinct is that something is badly depleted this winter, that instinct is correct and it is pointed at oil. Conflating it with natural gas storage is the mistake a gas contract pitch would be glad to have you make.
What the market has done since
Annual averages hide the thing that actually moves a winter bill. EIA’s October outlook puts Henry Hub at $3.48/MMBtu for 2026 and $3.16 for 2027, against $3.53 actual in 2025. That looks like mild decline.
Most of that decline is one month. EIA attributes it largely to January 2026, when Henry Hub averaged $7.72/MMBtu in the aftermath of winter storm Fern. Strip that month out and the agency forecasts $3.09/MMBtu from February through December 2027, unchanged from the same stretch of 2026. A forecast that falls because one past month was extraordinary is not a forecast of falling prices.
Near-term movement has been mixed rather than one-directional. The November contract settled at $3.24/MMBtu on 15 October, up 7.6% from its $3.01 settlement on 29 September, but below its $3.37 settlement on 24 September and 8.4% below its June level. AGA reads the month’s pricing as suggesting “the winter supply outlook remains strong”. Regional movement diverged further: eight regional day-ahead indices fell over the same period, with the Northeast among the most pronounced at 27.4% lower month-to-date.
Check the unit before any of this gets used on you. Henry Hub is a benchmark in Louisiana and an annual average smooths twelve months into one number. Neither describes what a New England household is charged on a cold Tuesday in February, and neither is the figure on the contract in front of you.
A second question for PJM readers in New Jersey, Pennsylvania and Delaware
Readers in New Jersey, Pennsylvania and Delaware sit inside the PJM grid region, where there is a second question this winter: whether the electricity supply holds in a deep cold snap. It is separate from household gas service and worth keeping separate.
NERC, the federally designated reliability organisation, classifies PJM as Elevated Risk for 2026 through 2028, moving to High Risk from 2029. In its January 2026 long-term assessment it is specific about when the trouble would come:
“the greatest risk of resource shortfalls leading to unserved energy or load loss in PJM occurs in the winter months during the early morning and evening hours. The risk is associated with generator availability and performance issues that can arise from equipment freezing and fuel supply issues during extreme winter conditions.”
Its base case for PJM puts expected unserved energy at 3,251 MWh in 2027, with loss-of-load hours at 0.61 a year. NERC footnotes the 2026 column of that table as a comparison drawn from an earlier study, so the two years are not a like-for-like series.
The Energy Department has been acting on it. Secretary Chris Wright has issued a series of emergency orders keeping Units 3 and 4 of the Eddystone Generating Station in Pennsylvania running past a retirement originally set for 31 May 2025. The orders ran in May, August and November 2025, and again for 24 February to 24 May 2026, which is the sequence DOE’s own announcement sets out as of October 2026. DOE’s stated reason is blunt: during Winter Storm Fern, between 26 and 29 January 2026, those two units ran for more than 124 hours in the middle of a declared energy emergency.
Wright’s own framing, in DOE’s announcement:
“The energy sources that perform when you need them most are inherently the most valuable — that’s why natural gas and oil were valuable during recent winter storms.”
What this does and does not mean for a household. The exposure is the electricity grid that depends on gas-fired generation. Residential gas service is delivered under different arrangements from the gas that fuels power plants, and a residential customer in Trenton or Wilmington is not the subject of these warnings. What the regulator describes is a deep, sustained cold snap in which the plants that make electricity may not all get gas when they need it. That is the mechanism NERC names when it points at fuel supply as a winter risk.
It is also an argument for the building envelope. A house that needs less heat is less exposed to an electricity event in a cold snap, whatever fuel it burns.
What actually reduces the bill
A fixed rate changes the shape of what you pay, not the amount of energy your house needs. The house is the part you own.
If you have not had a professional look at where the heat is going, start there. Many utilities in this region offer a home energy assessment at no cost to the customer, though eligibility, what it covers and whether it is free for you vary by utility and programme, so confirm all three with yours before booking. Our rebate finder works from a ZIP code and may ask which utility serves you, then lists the efficiency programmes that may apply. Insulation and air sealing reduce the energy the house needs under all three of the forecasts above, and they go on doing it after a supply contract has expired.
Frequently Asked Questions
Who should I talk to about fixed price energy rates?
Ask the person offering you a rate how they are paid, and ask before you discuss price. Compensation arrangements differ and are not something to assume:
- Who pays you for this, and how much?
- Is any part of what you earn tied to the rate I sign or the length of the term?
- Will you put the answers to those two questions in writing?
- What happens to your compensation if I decide to stay on utility default supply?
Whoever you are speaking to should be able to answer all four, and the answers tell you whose interests sit where. Rules on who may sell retail supply, and what they must disclose, are set state by state, so your state’s public utility commission is the place to check anyone’s standing.
Nobody can tell you what gas will cost this winter, and the three forecasts above disagree with each other. What a good adviser of either kind can do is explain what the contract does when prices move in each direction, which is a question about the document in front of you rather than about the market. Read the contract’s term, its cancellation terms and how it ends, and confirm them with the supplier and your utility before you sign.
And the same question about this page. GetEnergyRebates.com has no commercial relationship with any energy supplier, broker or retail supply marketplace, and earns nothing whether you sign a supply contract or stay on utility default supply. It does earn from other things: affiliated entities provide fee-based energy consulting and receive referral fees for connecting customers with contractors, and this site earns from its rebate finder. None of that income depends on your supply decision, and if any of it ever does, this page will say so.
This page is general information about a market, not financial advice about your situation.
Why do the forecasts disagree so much?
Because they measure different things. EIA forecasts one fuel’s expenditure for homes that heat with it. NEADA reports a household’s total heating cost and published three weeks earlier, against a different STEO edition. AGA’s figure covers space heating alone. A twenty-point spread between them is mostly a definitional gap, not a factual dispute.
Is natural gas storage low going into this winter?
No, on the most recent published data. As of the week ending 9 October 2026 the East, Midwest, Mountain and Pacific regions were all above their five-year averages, and national inventories were forecast about 2% above average entering the heating season. The national surplus is narrowing, and South Central is the exception at 3.1% below.
I am in PJM. Should I be worried about the grid this winter?
Worth understanding rather than worried. NERC rates PJM Elevated Risk for 2026–2028 and says PJM’s greatest shortfall risk falls in winter, driven by equipment freezing and fuel supply problems at generators. DOE has issued repeated emergency orders keeping a Pennsylvania plant online for exactly this reason. It concerns electricity supply in extreme cold, not your household gas service. NERC’s modelled loss-of-load hours for 2027 are 0.61 a year, and the same table rises to 9.97 hours by 2029, when it moves PJM to High Risk.
I keep hearing reserves are depleted. Is that wrong?
It is right about the wrong fuel for this page. The Strategic Petroleum Reserve holds crude oil and is near its lowest level since 1982. East Coast heating oil inventories are well below average, and EIA’s weekly survey put New York residential heating oil at $6.335 a gallon in the week of 12 October 2026. That is a statewide survey average and it excludes taxes, so it is neither a quote for any one house nor what a delivery ticket shows. Natural gas storage is a separate system and is not in that condition.
Figures in this article are quoted from EIA’s October 2026 Short-Term Energy Outlook and Winter Fuels Outlook 2026–27, NEADA’s September 2026 winter heating report, and AGA’s Winter Heating Outlook and market indicators for 16 October 2026. Each is identified in the text so you can check it. Energy markets move; where a figure carries a date, that date is the point it was accurate.