Natural Gas Forecast: Henry Hub Slides to $2.86 While Brent Rips 10% — the Widow-Maker at a Record 18 Cents Prices Zero Winter Risk

Natural gas fell to $2.86/MMBtu on July 16, down 2.31% from the previous session. The contract has dropped 9.17% over the past month and sits 19.35% below where it traded a year ago.

Brent is at $84.54, up 10.09% in five days. WTI holds above $80. US natural gas is at a two-month low.

That divergence is the entire article. Henry Hub is the only major energy benchmark in the world falling while a shooting war closes the most important chokepoint in the hydrocarbon trade.

The path has been relentless. August NYMEX futures fell 13.9 cents, or 4.3%, to $3.073 on Thursday July 9 — a six-week low and the lowest close since May 27. They broke below $3.00 on July 10. By Tuesday July 14 the front month fell 2.7 cents, or 0.9%, to $2.87, marking the lowest close since May 13, the fifth consecutive down session, and a second straight day in technically oversold territory — the first time both had happened since April. Wednesday saw the contract edge to $2.91 before sliding back below $2.90. It traded sideways through midday Wednesday, remaining below $3.00 as LNG maintenance curbed feedgas demand and production kept pressure on prices.

Six sessions. From $3.20 to $2.86. A 10.6% decline into a war.

The thesis is mechanical and it explains everything. US gas has decoupled from global gas because Freeport's 2.4 Bcf/d outage is trapping domestic supply at precisely the moment Qatari cargoes cannot sail. Ample domestic supply is shielding the US from the export pressures coming out of the Middle East. The front month is pricing an outage. The back month is pricing the export demand that follows it. That spread is the trade.

Today's storage print was the first crack in the bear case. Futures treaded lightly early Thursday as market participants girded for the government inventory data and the potential for a modestly bullish print relative to historical norms.

They got one. It has not moved the price yet.

The structural counterweight is building. Rapid demand growth and a trend toward higher-cost supply could lead to structurally higher Henry Hub prices over the next decade. An unprecedented surge in electricity demand has pushed the cost of new US gas-fired power generation to its highest level in nearly two decades.

None of that matters in July. All of it matters by February.

The 41 Bcf Print Was the First Bullish Number of the Summer

The EIA reported a net injection of 41 Bcf for the week ended July 10, lifting working gas in storage to 3,024 Bcf. Stocks sit 21 Bcf below last year at this time and 181 Bcf above the five-year average of 2,843 Bcf. Total working gas is within the five-year historical range. All regions posted increases.

Read the comparison, because it is the first genuinely constructive data point of the injection season.

NGI projected a 44 Bcf build. The year-ago injection was 47 Bcf. The five-year average for the week is 45 Bcf. The actual came in at 41 Bcf.

That print undershot the projection by 3 Bcf, the year-ago by 6 Bcf and the five-year norm by 4 Bcf. It is the first sub-average build of the summer.

Now stack the sequence and the deceleration becomes obvious. The week ended May 29 delivered 95 Bcf. June 5 produced 108 Bcf. June 12 came in at 73 Bcf. June 19 built 76 Bcf. June 26 injected 87 Bcf. July 3 added 61 Bcf. July 10 landed at 41 Bcf.

From 108 Bcf to 41 Bcf in five weeks. That is a 62% collapse in the weekly injection rate.

The context matters because the market has been trained to expect the opposite. The June 26 print of 87 Bcf topped market expectations and pressured futures lower — a decidedly bearish government print that had the market betting Lower 48 supply would prove sufficient to meet both summer demand and storage needs. The July 3 print of 61 Bcf exceeded expectations for 49 Bcf, topped the year-ago 53 Bcf and beat the five-year average of 51 Bcf, sending futures 4.3% lower to a six-week low.

Two consecutive bearish surprises. Then a bullish one.

The regional detail from the prior week is the tell nobody read. South Central salt storage posted a 1 Bcf withdrawal, and South Central nonsalt remained 5.6% below last year. Salt-dome facilities are the fast-cycling storage that responds to price signals within days. A withdrawal from salt in early July, with the national headline reading comfortable, is the physical market saying something different from the aggregate.

Buyers should pair the national headline with weather, LNG feedgas, production, regional basis and power-sector demand. The national number is the least informative line in the report.

3,024 Bcf and a Surplus That Just Contracted

The surplus to the five-year average has been the bear's entire argument, and it just moved the wrong way for the first time.

Track it. May 29: 138 Bcf above the five-year average of 2,440 Bcf. June 12: 151 Bcf above 2,608 Bcf. June 19: 152 Bcf above 2,683 Bcf. June 26: 175 Bcf above 2,747 Bcf. July 3: 185 Bcf above 2,798 Bcf. July 10: 181 Bcf above 2,843 Bcf.

The surplus expanded from 138 to 185 Bcf across five weeks, then contracted to 181 Bcf.

Four Bcf. It is nothing. It is also the first negative print in the series that has justified the entire short position.

The year-over-year comparison is doing something similar. Stocks ran 3 Bcf below last year on May 29, then 29 Bcf below on June 12, then 49 Bcf below on June 19, then 23 Bcf below on June 26, then 15 Bcf below on July 3, and now 21 Bcf below on July 10.

At 3,024 Bcf, inventories sit 0.7% below year-ago levels and 6.4% above the five-year norm. That is not a glut. That is a market with a normal cushion.

The projection is where the bear case still holds. EIA forecasts working gas inventories reaching 3,966 Bcf by the end of October — 5% above the five-year average. Inventories remain relatively high because record natural gas production, led by growth in the Permian region, is meeting rising demand. Above-average inventories heading into winter drive a 4Q26 Henry Hub forecast of $3.57/MMBtu, 5% less than the same quarter last year.

Run the arithmetic on that path. Getting from 3,024 Bcf on July 10 to 3,966 Bcf by October 31 requires 942 Bcf across roughly 16 weeks — an average build of 59 Bcf per week.

The last print was 41 Bcf. The prior was 61 Bcf.

If injections keep decelerating from 41 Bcf, the EIA's end-October target is not reachable, and the 5% surplus that anchors the $3.57 fourth-quarter forecast does not exist. That is the single most important unpriced variable in this market.

Storage was on track to surpass 3 Tcf despite lingering heat. It did. What happens next is the question.

Freeport's 2.4 Bcf/d Is Trapping the Gas That Cannot Sail

This is the mechanism that explains why American gas is falling in a war, and it is entirely artificial.

Maintenance at the 2.4-bcfd Freeport LNG terminal began July 10 and runs until late August. Scheduled outages at the Texas facility prevent gas flows from being readied for export, increasing the available supply for domestic use. The plant is only expected to resume operations in late August.

Two point four billion cubic feet a day, for roughly six weeks, redirected into the domestic market.

Do the math. 2.4 Bcf/d over 42 days is 100.8 Bcf of gas that would have been liquefied and shipped, now landing in storage instead. That is 2.4 weeks of injections at the current 41 Bcf pace, arriving in the middle of what should be peak cooling demand.

The market caught it late. Natural gas futures extended their sell-off on July 10 as an unexpected drop in Freeport LNG demand added fresh pressure to a market already seeing disappointing upside from summer cooling demand and stout production. Futures dug deeper into the red as robust supply intersected with the unexpected maintenance work.

Unexpected is the word that cost holders 10%.

The offset is partial. Golden Pass LNG has steadied natural gas deliveries after a late June swoon, offering a bright spot for feedgas demand as the Freeport turnaround pulls US export volumes lower. Average gas flows to the nine big US LNG export plants rose to 17.6 bcfd so far in July, up from 17.4 bcfd in June — but they remain below the monthly record high of 18.8 bcfd set in April.

That increase in average feedgas came despite the Freeport reduction. Read that correctly: the rest of the complex is running hot enough to offset a 2.4 Bcf/d outage and still grow month over month.

BloombergNEF estimated LNG feedgas deliveries at 19.2 Bcf/d on one Monday, near record territory.

The expiry is the catalyst. Late August. When Freeport returns, 2.4 Bcf/d leaves the domestic balance in a single step — heading into shoulder season with storage that may not have reached 3,966 Bcf.

The August contract does not survive that. The February contract is already priced for it.

Production at 110.2 and the Permian Problem

Supply is the counterweight and it is the one variable that does not respond to price.

Average gas production in the Lower 48 states increased to 110.2 billion cubic feet per day so far in July from 110.0 bcfd in June. Output climbed to nearly 114 Bcf/d over a recent weekend — the highest level in more than two and a half months. BloombergNEF pegged one Monday's dry gas production at 111.2 Bcf/d, up 2.3% year over year.

Record production, led by growth in the Permian region.

Here is the trap, and it is the most important structural fact in this market. Strong crude encourages more Permian drilling. More Permian drilling produces more associated gas whether gas prices justify it or not. Higher oil prices can actually increase natural gas supply even while gas-specific economics deteriorate.

Read what that means today. Brent at $84.54, up 28.83% year to date, is bearish for Henry Hub. The war that lifted crude 10.09% in five days is funding the drilling that produces the associated gas that is pinning Henry Hub at $2.86.

Associated gas has no price elasticity. It comes out of the ground because the oil is worth $84. The operator does not care what gas fetches — it is a byproduct, and at Waha it has historically been a liability.

That dynamic is now improving in a way that makes it worse for Henry Hub. Prices at the Waha hub in West Texas could average $3/MMBtu next year — a remarkable turnaround for a market long plagued by negative pricing and pipeline bottlenecks. Waha at $3 means Permian gas that used to be flared or given away now has a path to market.

More Permian gas reaching a hub is more Permian gas competing at Henry.

The demand side is real but delayed. EIA forecasts US natural gas consumption in the electric power sector increasing 2% in 2026 and another 4% in 2027 to a record 38.1 Bcf/d, with monthly consumption reaching 50.6 Bcf/d in July 2027.

Record power burn arrives next year. Record production is here now.

Mexico's imports of US pipeline natural gas hit an all-time record in June at 8.35 Bcf/d — the one demand line that is already working.

The August-February Spread at $1.09 Is the Whole Trade

The curve is where the disagreement lives and it is enormous.

August futures sit at $2.86. February 2027 futures held close to $3.95.

That is $1.09 of spread — a 38.1% premium for winter gas over summer gas.

That spread between August and February tells you where the real conviction is. The summer contract sees the supply. The winter contract sees the export demand eating into it month after month.

That is the cleanest framing of this market available and it deserves to be taken literally. Two contracts on the same commodity, at the same delivery point, six months apart, disagree by 38%. One of them is wrong.

The structural demand from exports is the reason the spread stays wide even when front-month production runs at highs. Approximately 15% of all US dry gas production is now committed to overseas buyers. EIA expects LNG exports to average roughly 17.0 Bcf/d this year before climbing another 9% next year.

Run that forward. 17.0 Bcf/d growing 9% is 18.5 Bcf/d in 2027 — against Lower 48 production of 110.2 Bcf/d. Export demand alone would consume 16.8% of total output. Add power-sector burn rising to a record 38.1 Bcf/d and the demand stack tightens materially before the supply stack responds.

February is pricing that. August is pricing Freeport being closed.

The historical precedent for how violently this contract can reprice is recent and it is on the record. In January 2026, the Henry Hub spot price rose $1.86 in a single report week — from $3.12/MMBtu to $4.98/MMBtu. The February 2026 NYMEX contract climbed $1.76, from $3.120 to $4.875. The 12-month strip rose 65 cents to $3.970.

Spot went up 59.6% in seven days. The 12-month strip moved 19.6%. The increase was mostly a reaction to anticipated changes in 2026 storage balances and had no material influence on the longer-dated curve.

That is what a storage scare does to this asset. It happened six months ago, at a starting price 26 cents above today's.

The Widow-Maker at 18 Cents — a Record Low

There is one number in this market that measures fear directly, and it just printed the lowest reading in its history.

The premium of futures for March over April 2027 fell to a record low of around 18 cents per MMBtu.

The industry calls the March-April spread the widow-maker, because rapid price moves resulting from changing weather forecasts have forced some speculators out of business.

That spread is the market's price on winter risk. March is the last month of withdrawal season. April is the first month of injection season. The gap between them is what the market will pay to insure against storage running out before spring.

Eighteen cents is the market saying there is no risk at all.

In a sign the market is not too worried about gas supplies in coming months, the widow-maker collapsed to a record low. That framing is correct and it is exactly why the number is dangerous.

A record low on a spread whose entire purpose is to price tail risk means positioning is uniformly on one side. There is no premium embedded for a cold winter, for a supply disruption, for Europe's storage failing to rebuild, or for Freeport returning into a deficit.

Every one of those is live.

Europe's storage is below seasonal norms and needs to rebuild before winter. Qatar's Ras Laffan damage keeps the pull on US cargoes strong. Iran and the US have resumed blockading tankers from leaving the Persian Gulf, which limits LNG flows for major European and Asian consumers. Ample US supply contrasted with those constrained global flows.

Read the chain. European storage is short. Qatari LNG cannot sail. The only marginal cargo available is American. And the March-April spread — the instrument that would price the consequence — is at a record low.

The market has priced abundance into the one contract designed to price scarcity.

That is not a prediction that it breaks. Record lows persist for months. It is an observation that the asymmetry is now extreme, and that the widow-maker earned its name by punishing exactly this configuration.

Eighteen cents is what complacency costs.

LNG Feedgas at 19.2 and 15% of Production Spoken For

The export franchise is the structural story and the numbers are already large.

Approximately 15% of all US dry gas production is now committed to overseas buyers. BloombergNEF estimated LNG feedgas deliveries at 19.2 Bcf/d on one Monday, near record territory. Average flows to the nine big US LNG export plants rose to 17.6 bcfd so far in July from 17.4 bcfd in June, though below the 18.8 bcfd monthly record from April.

The EIA expects exports to average roughly 17.0 Bcf/d this year before climbing another 9% next year. Analysts expect US LNG exports to grow about 10% annually through 2030.

Compound that. 17.0 Bcf/d growing 10% annually reaches 27.4 Bcf/d by 2030 — a quarter of today's entire Lower 48 output, contracted away.

The pipeline of new demand is filling. Argent LNG signed a tentative agreement with Ukraine's Naftogaz Group aimed at shipping US LNG supply as the developer ramps up commercialization of its proposed Louisiana export project. LNG Canada's backers agreed to give First Nations neighboring the British Columbia facility an opportunity to invest up to $1 billion in part of its second phase.

Every one of those projects converts domestic molecules into export molecules permanently.

The near-term picture is the opposite and it is why the front month is at $2.86. Natural gas cash prices trended lower at most locations on Wednesday amid mixed weather and subdued LNG feedgas demand. Golden Pass has steadied after a late June swoon. Freeport is down until late August.

Feedgas at 17.6 Bcf/d with Freeport offline means the complex is running at 20.2 Bcf/d of effective capacity utilization once the plant returns. The April record of 18.8 Bcf/d gets broken in September.

The demand is contracted. The supply is committed. The only variable is timing.

That is why February holds $3.95 while August prints $2.86, and it is why the front-month weakness is a maintenance calendar rather than a market verdict.

Cooler Through July 23 and Renewables at Record Highs

Weather killed the summer trade and the second driver is structural.

The Commodity Weather Group said forecasts had turned cooler, with below-average temperatures anticipated in the Southwest through July 23, likely limiting cooling demand.

That is the peak of the cooling season, forecast below normal, in the region that drives the most air-conditioning load in the country.

The market was already disappointed. Futures sold off on disappointing upside from summer cooling demand and stout production. Heat forecasts kept the selling from accelerating but they did not reverse it.

The mid-summer heat wave of 2026 served as a case study in regional market insulation. East Coast markets experienced dramatic moves while the national balance barely registered them. Summer heat drove up Midwest cash prices, but one of the nation's largest storage regions continued rebuilding inventories above both last year's level and the five-year average.

Read that. A heat wave arrived and storage kept building. That is the definition of a market where supply overwhelms weather.

The second driver is permanent and it is doing more damage than anyone models. Solar and wind power generation stateside rose to near record highs in July, taking market share from gas-powered power plants.

Renewables at record output during peak cooling season is gas demand destruction that does not reverse when the weather turns. Every gigawatt of solar installed in the Southwest permanently removes summer afternoon gas burn from the balance — which is precisely the load Henry Hub relies on in July.

EIA forecasts wholesale electricity prices lower this summer than last, primarily because of lower costs of natural gas delivered to power plants, though heatwaves could still cause price spikes.

Cheaper gas producing cheaper power producing less incentive to conserve. The loop is not tightening.

Cooling degree days were higher during the July 10 reporting period, which is what produced the 41 Bcf build against a 45 Bcf norm. Heat is working. It is just working against a 110.2 Bcf/d supply wall and a 2.4 Bcf/d export outage.

NGPL, Hugh Brinson and the Pipes That Make It Worse

Infrastructure is quietly restructuring the Henry Hub balance and every project points the same direction.

Kinder Morgan's Natural Gas Pipeline Company of America has stepped up natural gas deliveries toward Louisiana after bringing its Texas-Louisiana Expansion Project to full service, adding supply pressure to Henry Hub spot prices.

That is molecules arriving at the delivery point of the front-month contract, on a pipeline that just came into full service, during the month the contract is being priced.

Energy Transfer's Hugh Brinson Pipeline has taken its first formal step with federal regulators, filing a petition with FERC to set rates and operating terms for the first phase of its 1.5 Bcf/d Permian-to-North Texas line.

Another 1.5 Bcf/d of Permian egress in the queue. That is 1.5 Bcf/d of associated gas that currently has nowhere to go, gaining a route to market.

The Waha consequence is direct. Prices at the Waha hub in West Texas could average $3/MMBtu next year, a remarkable turnaround for a market long plagued by negative pricing and pipeline bottlenecks. Waha going from negative to $3 is a triumph for Permian producers and a headwind for Henry Hub, because it means stranded gas stops being stranded.

The bottleneck was the bull case for Henry Hub and it is being engineered away.

The demand-side infrastructure is real too. Mexico's imports of US pipeline natural gas hit an all-time record in June at 8.35 Bcf/d. That is the single largest pipeline export market growing to a record while the domestic price falls — which tells you the constraint was never demand.

Argent LNG's Naftogaz agreement and LNG Canada's second phase are both years out.

The asymmetry in timing is the whole problem. Supply infrastructure — NGPL's expansion, Hugh Brinson's 1.5 Bcf/d, Permian debottlenecking — arrives in 2026 and 2027. Demand infrastructure — new liquefaction trains, data center power, the 38.1 Bcf/d record power burn — arrives in 2027 and beyond.

That gap is exactly the $1.09 between August and February, extended across a decade.

Rapid demand growth and a trend toward higher-cost supply could lead to structurally higher Henry Hub prices over the next decade. Every pipeline coming online delays that arrival by a quarter.

EIA Says $3.60 and the Tape Says $2.86

The gap between the official forecast and the market is the widest in the energy complex, and one of them is going to look foolish.

EIA expects the Henry Hub spot price to average close to $3.60/MMBtu over 2026 and 2027. Adjusted for inflation, that price sits about 10% below the average Henry Hub price from 2016 through 2025. A separate framing has the spot price averaging close to $3.70/MMBtu in 2026 before declining below $3.50/MMBtu next year.

The front month trades at $2.86.

That is 74 cents — 20.6% — below the annual average the government expects, with five and a half months left in the year.

Run what that requires. If Henry Hub averages $3.60 for 2026 and the first six and a half months have averaged materially below it, the back half has to run substantially above $3.60 to hit the annual number. That is not a forecast of the current price. It is a forecast that the current price is wrong.

The quarterly detail sharpens it. 4Q26 is forecast at $3.57/MMBtu, 5% less than the same quarter last year. 4Q27 averages $3.78/MMBtu, up 6% from 4Q26. For 2027 as a whole, the price averages just under $3.50/MMBtu.

February 2027 futures at $3.95 sit above the entire 2027 EIA forecast. The market is more bullish on winter than the government is on the year.

The inventory assumption underneath the forecast is the load-bearing wall. EIA expects the surplus to the five-year average to narrow to 1% at the end of October 2027, on the back of strong demand growth. That narrowing from 5% to 1% across twelve months is what produces the 6% price increase into 4Q27.

The forecast was completed July 1 — before the 41 Bcf print, before Freeport's outage was fully absorbed, before the front month broke $3.00. The next release lands August 11.

The honest read: EIA is modeling a market that normalizes. The tape is pricing a market that is drowning in a six-week maintenance window. Both can be right, sequentially.

Oversold for the First Time Since April

The technical setup is the most stretched it has been this year and the levels are close.

The August contract fell for five consecutive sessions into July 14 and stayed in technically oversold territory for a second straight day — both firsts since April. The contract is holding below the $3.00 mark. Earlier in the month it was holding above the 50-day moving average, but it took a fight to get there.

That fight was lost. $3.00 broke on July 10 and has not been reclaimed.

Map the levels. $2.86 is spot. $2.87 was Tuesday's close — the lowest since May 13. Below that, the market has no reference until the spring lows, because the entire structure from May onward was built above $3.00. $3.00 is now resistance rather than support. $3.073 was the July 9 close and the six-week low that is now overhead. $3.20 marked the early-July trade and the level where the 50-day sat.

The upside reference is $3.95 — February 2027 — which is not a level the August contract can reach but is the number the curve says the commodity is worth in six months.

Five consecutive down days, two days oversold, first time since April, and a market that just received a bullish storage print and did not rally.

That last part is the tell and it cuts against the bulls. A 41 Bcf build against a 45 Bcf five-year average, a 44 Bcf projection and a 47 Bcf year-ago comparison is a bullish surprise on three measures. The contract is down 2.31%.

When a market cannot rally on good news, it has a positioning problem rather than a fundamental one — and positioning problems resolve violently in both directions.

The historical analog is instructive. September futures once plunged to a nine-month low, dropping 4.5% amid near-record output, ample storage and cooler weather reducing demand, falling for five straight weeks and down 24% overall — with record production averaging 108.4 bcfd and storage 6% above normal.

Production is now 110.2 bcfd. Storage is 6.4% above normal. The setup rhymes.

What Has to Break

Map the bull case. The 41 Bcf print is the start of a trend rather than a weather artifact — injections keep decelerating from 41 Bcf and the path to EIA's 3,966 Bcf end-October target breaks, because it requires a 59 Bcf weekly average from here. Freeport returns in late August and 2.4 Bcf/d leaves the domestic balance in a single step, into shoulder season with storage short of target. The widow-maker at a record 18 cents reprices as Europe's below-normal storage collides with Qatari cargoes that cannot clear Hormuz. LNG feedgas breaks the 18.8 Bcf/d April record in September. The 181 Bcf surplus keeps contracting.

On that path, the January precedent applies — a $1.86 move in seven days on a storage scare — and $3.95 February is the anchor rather than the outlier.

Map the bear case. Production holds 110.2 Bcf/d and grinds toward the 114 Bcf/d weekend print, because Brent at $84.54 funds Permian drilling that produces associated gas regardless of what gas fetches. NGPL's expansion and Hugh Brinson's 1.5 Bcf/d keep pushing molecules toward Louisiana. Waha at $3 unstrands West Texas gas. Renewables at record output permanently erode summer burn. Below-normal Southwest temperatures through July 23 kill the last of cooling demand. The 181 Bcf surplus rebuilds and 3,966 Bcf arrives on schedule.

On that path, $2.86 is not the floor. It is the way station.

The base case sits on a calendar. The August contract chops between $2.80 and $3.00 until Freeport returns in late August, then the balance loses 2.4 Bcf/d overnight and the market discovers whether 3,024 Bcf was enough.

What makes this market unlike any other energy benchmark right now: it is the only one where a war is bearish, where higher oil prices increase supply, and where the instrument designed to price winter risk sits at a record low while European storage runs below normal and Qatari LNG sits behind a blockade.

The August contract sees the supply. The February contract sees what happens when the outage ends.

$1.09 is the distance between them. One of them is wrong.

That's TradingNEWS