Henry Hub Ignores $96 Brent as Record 111.2 Bcf/d Production Caps the Rally — 18.62% Upside to $3.44 Needs Feedgas Above 19 Bcf/d

Front-month Henry Hub natural gas futures traded $2.900 per MMBtu on Wednesday, September 2, up $0.012 or 0.42% from a $2.888 prior settlement, inside a $2.832 to $2.933 session band. Volume ran 77,333 contracts. Recent prints have ranged as low as $2.772 against a $2.794 settle for a 0.79% decline, which frames the working band for this week between $2.77 and $2.93.

Gas has held around $2.90 — the highest level in five weeks — supported by a smaller-than-usual storage injection and forecasts for continued warm weather. The September NYMEX contract settled $2.842 in late August after gaining $0.03 on the session, so the front of the curve has added roughly six cents in the intervening sessions.

That six-cent move deserves context, because it happened during the most violent week in energy markets this year.

Brent crude for November delivery ran as high as $96.59 on Wednesday, up over 2%, and has gained roughly 7% on the week. West Texas Intermediate for October reached $91.78. Iran's Revolutionary Guards said two oil tankers struck naval mines in the Strait of Hormuz. U.S. forces hit IRGC targets near Bandar Abbas and Chabahar. Iran retaliated against Jordan, the UAE, Kuwait, Bahrain and Iraq.

Henry Hub moved 0.42%.

That divergence is the thesis of this forecast. U.S. natural gas is the one energy market not trading the Iran war, because domestic gas is landlocked and the only transmission channel to global pricing is liquefaction capacity — which is currently running 12.8% below its record. Gas is trading two variables and two variables only: a record production base and a storage surplus that is finally compressing.

The surplus is the more interesting of the two. Working gas stood 198 Bcf above the five-year average on August 7. Three weeks later that cushion measured 167 Bcf. A 31 Bcf compression in three weeks, during the shoulder period when injections should be rebuilding, is the first genuinely constructive supply signal Henry Hub has produced since spring.

The 15 Bcf Injection Was Less Than Half The Five-Year Average

The most recent storage print was the tightest of the season relative to normal, and the sequence behind it matters more than the single number.

Net injections into storage totaled 15 Bcf for the week ending August 21, against a five-year average of 33 Bcf for the equivalent week and 17 Bcf during the same week last year. Consensus had looked for 20 Bcf.

A 15 Bcf build against a 33 Bcf normal is 55% below average. That is a substantial single-week tightening, and it came alongside falling LNG feedgas demand — meaning the tightness originated on the domestic side rather than from export pull.

The prior weeks trace the same direction. The week ending August 14 delivered a 16 Bcf injection, nearly matching expectations as a heavy South Central withdrawal limited the overall build. The week ending August 7 registered a larger build that lifted working gas to 3,153 Bcf. Before that, the week ending July 31 produced 33 Bcf and the week ending July 24 produced 28 Bcf.

The July 24 week's regional detail is instructive about where the tightness lives. Mountain withdrew 2 Bcf, Pacific withdrew 7 Bcf, and South Central withdrew 9 Bcf — with South Central salt alone withdrawing 14 Bcf while nonsalt stocks ran 6.0% below the prior year. Salt-dome facilities are the fast-cycling storage that responds to power burn, and they were being drained through the peak of summer.

That is a heat-driven withdrawal inside an injection season, and it is what has been eating the surplus.

The weekly report for the week ending August 28 publishes Thursday at 10:30 a.m. ET. Another sub-20 Bcf print would compress the five-year surplus below 150 Bcf and mark the fourth consecutive week of below-normal builds — a pattern that historically precedes a repricing of the winter strip.

The counterweight: a build back above 30 Bcf as cooling demand fades would rebuild the cushion just as fast.

3,184 Bcf Is Comfortable But No Longer Generous

The absolute inventory level is the anchor for every price scenario, and it currently sits in a genuinely balanced position.

Working natural gas stocks totaled 3,184 Bcf as of Friday, August 21. Stocks were 167 Bcf, or 6%, above the five-year average, but 30 Bcf, or 1%, below the prior year's level at the same point.

That combination — above the five-year average, below last year — describes a market that is well supplied but no longer carrying the exceptional cushion it held earlier in 2026.

The trajectory tells the story. Working gas measured 3,056 Bcf on July 17 with a 183 Bcf five-year surplus and a 16 Bcf year-over-year deficit. It reached 3,084 Bcf on July 24 with a 185 Bcf surplus and a 32 Bcf deficit. It hit 3,117 Bcf on July 31 with a 195 Bcf surplus. It reached 3,153 Bcf on August 7 with a 198 Bcf surplus. Then 3,169 Bcf on August 14 with a 185 Bcf surplus. Then 3,184 Bcf on August 21 with a 167 Bcf surplus.

The surplus peaked at 198 Bcf on August 7 and has fallen 31 Bcf in two weeks. Absolute inventories have added only 31 Bcf across the same span.

Periods with higher-than-average inventories are generally associated with lower prices, while lower storage levels correspond with higher prices and tighter conditions. As inventories move closer to or below the five-year average, the forecast Henry Hub price rises. Storage remains the single key indicator of market balance and price formation.

The refill season runs through October 31. At the current pace of roughly 15 to 16 Bcf per week, ten remaining weeks would add approximately 155 Bcf, putting end-of-season inventories near 3,340 Bcf. At the five-year normal pace of 33 Bcf, the total would reach 3,514 Bcf.

The difference between those two outcomes — 174 Bcf — is the entire winter risk premium. It gets decided by September and October weather.

Production Is Setting Records And That Caps Everything

The supply side is the structural reason gas cannot rally the way crude has, and the numbers are unambiguous.

Marketed natural gas production is on track for a new annual record in 2026, expected to average nearly 123 Bcf per day for the year. Dry gas production is also on track for a record, forecast to average 111.2 Bcf per day in 2026 — a gain of 3.3% over the 2025 record.

The gains are being driven by two sources. Increased associated gas production in the Permian basin arrives as a byproduct of oil drilling, which means it is insensitive to gas prices. With WTI at $89.58 and Permian economics strong, operators drill for crude and the gas comes out regardless of whether Henry Hub trades $2.50 or $3.50.

The second source is stronger Haynesville output responding to Gulf Coast feedgas demand. That production is price-sensitive and has been switched on by the LNG buildout.

The balance arithmetic explains the price path. For 2026, demand including exports is forecast to increase by less than 1%, or 0.6 Bcf/d, while supply including imports increases by nearly 1%, or 1.1 Bcf/d. Supply growth outpaces demand growth by 0.5 Bcf/d.

Half a billion cubic feet per day of surplus supply, compounding across a year, is 182 Bcf of incremental inventory. That is why the storage cushion persists despite record heat and why every rally toward $3.00 has been sold.

The associated-gas dynamic creates a perverse feedback loop with the current oil situation. Brent at $96.59 and WTI at $91.78 incentivize maximum Permian drilling. Maximum Permian drilling produces maximum associated gas. So the Iran war, which should be inflationary for energy broadly, is arguably bearish for Henry Hub through the associated-gas channel.

That is the single most counterintuitive fact in this market right now, and it is why gas moved 0.42% on a day crude moved over 2%.

LNG Feedgas At 17.1 Bcf/d Is 12.8% Below The Record

The export channel is the only mechanism connecting Henry Hub to global gas pricing, and it has been running below capacity for months.

For the week ending August 19, feedgas flows averaged 17.1 Bcf per day — 12.8% lower than the early-year daily record of 19.6 Bcf per day. That 2.5 Bcf/d gap represents demand that would otherwise be drawing on domestic supply.

Freeport LNG alone removed roughly 2 Bcf per day of nominal capacity during its maintenance period. The reduction in LNG feedgas demand is significant for the domestic storage balance because export facilities represent one of the largest single sources of natural gas demand in the U.S. market.

The maintenance cycle is now ending. Cheniere Energy's Corpus Christi facility and Freeport LNG in Texas have resumed operations after maintenance, and gas demand from LNG export plants has improved.

That return is the near-term bullish catalyst. Restoring 2 to 2.5 Bcf/d of feedgas demand removes roughly 14 to 17 Bcf per week from the storage build — which alone would take a 30 Bcf normal injection down to the 13 to 16 Bcf range currently being printed.

The August Short-Term Energy Outlook forecasts U.S. LNG exports to average 16.5 Bcf/d in the third quarter of 2026, marked slightly below the prior forecast because of the Freeport maintenance. Exports are expected to continue increasing through 2027.

The structural growth path is substantial. LNG exports grow by a forecast 9%, or 1.3 Bcf/d, in 2026 and 11%, or 1.7 Bcf/d, in 2027, driven by the ramp-up of three new facilities: Plaquemines LNG, Corpus Christi Stage 3, and Golden Pass LNG. Plaquemines and Corpus Christi Stage 3 continue ramping toward full operations, with Golden Pass expected to begin operations during 2026.

Three billion cubic feet per day of incremental export demand over two years against 0.5 Bcf/d of annual surplus supply is what flips the balance. It arrives in 2027, not 2026.

Heat Through September 11 Is Buying Time

Weather is doing the work that fundamentals are not, and the forecast window extends further than usual.

Above-average temperatures are expected across the eastern two-thirds of the United States from September 2 through September 11, keeping cooling demand elevated. That is a ten-day window of sustained power burn arriving after the calendar summer would normally have ended.

A persistent heat dome across the South has been supporting strong regional cooling demand and slowing the pace of storage injections, though the effect on total demand and Henry Hub prices has been muted relative to the intensity of the heat.

That muting is itself the diagnostic. When record heat produces only a 15 Bcf injection instead of a draw, and the price response is six cents, the market is telling you that supply is abundant enough to absorb the demand shock.

The regional price data shows where the heat actually bit. During the most recent report week, the largest reported increase across hubs was $1.19/MMBtu at SoCal Border Average, reflecting the combination of extreme heat, stronger gas-fired power generation and pipeline capacity constraints affecting westbound flows. The largest decline was $0.31/MMBtu at Texas Eastern M-3.

A $1.50 spread between the strongest and weakest regional moves in a single week describes a market where the constraint is transportation rather than molecules. There is enough gas. There is not always enough pipe.

For Henry Hub specifically, that distinction caps upside. Louisiana delivery sits at the intersection of the Gulf Coast production complex and the LNG terminals — the most liquid, best-supplied point on the network. Regional scarcity in Southern California does not lift the benchmark.

The September 11 forecast expiry is the near-term event. Cooling demand fading into normal shoulder-season weather would push weekly injections back toward the 30 Bcf range and rebuild the five-year surplus that has just compressed 31 Bcf.

The bulls need the heat to persist into late September, or they need the LNG restart to arrive before it fades.

The Price Forecast Has Been Cut Three Times This Year

The revision history of the official outlook is the clearest possible statement of how this market has deteriorated.

The August 2026 Short-Term Energy Outlook lowered the 2026 Henry Hub spot price forecast by more than 6% from the July edition, moving from $3.67 to $3.44 per MMBtu. That annual forecast has now fallen more than 20% since the February 2026 estimate of $4.31 per MMBtu.

The February number was elevated by real conditions — sustained heating demand, record storage withdrawals, and temporary price spikes during Winter Storm Fern. At the peak of that event, the Henry Hub spot price rose $1.86 in a single week from $3.12/MMBtu to $4.98/MMBtu, with the February NYMEX contract climbing $1.76 from $3.120 to $4.875 and the 12-month strip adding 65 cents to $3.970.

Every dollar of that winter premium has been given back and then some.

The January outlook had projected the 2026 annual average at just under $3.50 per MMBtu, a 2% decline from 2025, with 2027 rising sharply to just under $4.60 — a 33% year-over-year increase. The $3.44 August revision keeps the 2026 figure roughly on that January path while the front of the curve trades $2.900, well beneath it.

That gap between spot at $2.900 and the 2026 annual forecast at $3.44 is 18.6%, and it exists because the annual average includes the January and February winter spike that has already occurred. The remaining months are being marked considerably lower.

Independent modelling puts September 2026 between $2.752 and $3.320 with a $3.004 average and a $3.058 month-end close, implying a 6.0% monthly gain. October is projected between $2.725 and $3.315 with a $2.992 average and a $2.868 close, a 6.2% decline. November runs $2.696 to $2.980 averaging $2.846.

That shape — up in September, down in October, flat in November — describes a market where the late-summer heat premium fades and winter demand has not yet arrived. It is the classic shoulder-season trough, and it is where the front month currently sits.

Why Gas Is Not Trading The Iran War

The disconnect between crude and Henry Hub this week is structural, and understanding it is the key to positioning.

Brent has gained roughly 7% on the week and traded as high as $96.59. Henry Hub has gained about 2% over the same span. The ratio of Brent to Henry Hub sits near 32.5 to 1 on a nominal basis, against an energy-equivalent parity ratio closer to 6 to 1.

U.S. natural gas is a landlocked commodity priced by North American supply and demand. There is no mechanism for a Hormuz disruption to lift Henry Hub directly, because the gas cannot leave without passing through a liquefaction terminal.

The one channel that could change it is global LNG substitution. Qatari LNG — a substantial share of global seaborne supply — transits the Strait of Hormuz. Crude and petroleum liquids through Hormuz have collapsed to an average of 4.9 million barrels per day in the second quarter from 21.6 million in the fourth quarter of 2025, a 77% reduction, and LNG cargoes face the same mines, the same insurance markets and the same convoy constraints.

If Qatari LNG loadings are meaningfully disrupted, Asian and European buyers bid for U.S. cargoes. Higher international netbacks pull maximum feedgas through Gulf Coast terminals, and Henry Hub tightens through the export channel.

That transmission has not shown up in the data. Feedgas at 17.1 Bcf/d against a 19.6 Bcf/d record is running below capacity for maintenance reasons, not because terminals are being bid to maximum output.

The offsetting bearish channel is real and immediate. Brent at $96.59 and WTI at $89.58 maximize Permian drilling economics, and Permian associated gas is the largest single source of production growth pushing 2026 marketed output toward a record 123 Bcf/d.

Higher oil produces more gas. That is the mechanism, and until LNG feedgas exceeds 19 Bcf/d on a sustained basis, it dominates.

The 2027 Setup Is The Real Trade

Everything constructive about natural gas sits twelve to eighteen months out, and the forecast balance makes that explicit.

Forecast supply growth outpaces demand growth by 0.5 Bcf/d in 2026 but then falls behind by 1.6 Bcf/d in 2027, putting upward pressure on prices. For 2027, demand growth of 2.5 Bcf/d exceeds supply growth of 0.9 Bcf/d.

That is a 2.1 Bcf/d swing in the annual balance across a single calendar turn. Annualized, a 1.6 Bcf/d deficit removes 584 Bcf from inventories over a year — enough to eliminate the entire current 167 Bcf surplus and push stocks well below the five-year average.

The driver is LNG. Feed gas demand from export facilities rises 1.7 Bcf/d in 2027 on top of 1.3 Bcf/d in 2026, as Plaquemines LNG, Corpus Christi Stage 3 and Golden Pass LNG reach full operations. Three facilities, roughly 3 Bcf/d of incremental demand, against supply growth that decelerates from 1.1 Bcf/d to 0.9 Bcf/d.

The price consequence is a forecast 33% increase in the annual average Henry Hub spot price, from just under $3.50 in 2026 to just under $4.60 in 2027. Longer-dated modelling puts the December 2027 futures price near $3.53 with a spot estimate of $4.19.

The forward curve has not fully priced that. Front-month gas at $2.900 against a 2027 forecast near $4.60 implies 58.6% of upside if the balance forecast proves correct.

That gap exists for a reason. LNG project timelines slip routinely, production forecasts have consistently been revised higher, and the current market has spent 2026 watching the annual price forecast get cut from $4.31 to $3.67 to $3.44.

For a September 2026 trade, the 2027 setup is context rather than catalyst. It establishes that the multi-year risk is skewed upward and that positioning short at $2.90 carries asymmetric downside on any weather or supply shock.

What it does not do is give the front month a reason to rally before the heating season begins.

Winter Memory: Fern, $4.98 Spot, And What Cold Does

The market's willingness to price winter risk depends on how recently it has been burned, and this market was burned badly nine months ago.

Winter Storm Fern produced sustained heating demand, record storage withdrawals and temporary price spikes that took the Henry Hub spot price from $3.12/MMBtu to $4.98/MMBtu in a single week — a $1.86 move, or 60%. The February NYMEX contract climbed $1.76 to $4.875. The 12-month strip added 65 cents to $3.970.

At the peak of that episode, working gas stocks stood at 3,065 Bcf, 177 Bcf above the five-year average and 141 Bcf above the prior year. The market entered that winter with a comparable cushion to the one it carries now — 167 Bcf above the five-year average — and the cushion did not prevent a 60% weekly spike.

That is the case for owning optionality rather than being short. Storage surpluses cap price during normal weather and provide essentially no protection against a genuine cold event, because the constraint during a deep freeze is deliverability rather than inventory. Frozen wellheads reduce production precisely when demand peaks.

The 2026 record production base cuts both ways here. A 111.2 Bcf/d dry gas run rate provides more absolute supply than any prior winter, but it also means a higher percentage of that supply sits in freeze-prone basins. Permian and Appalachian production both suffer meaningful freeze-offs during Arctic events.

Earlier in 2026, forward prices demonstrated how fast the curve can reprice on a forecast alone. The 2026 Henry Hub forward strip rose to $3.90/MMBtu on January 21, 21% higher than where it had settled five days earlier, purely on the emergence of sub-freezing temperature forecasts across the Central and Eastern U.S.

A 21% move in five days on a weather model. At $2.900, the equivalent move is $3.509.

That is the risk a short position at these levels carries into October, and it is why the winter strip typically trades at a premium the front month does not.

Technicals And Where The Curve Sits

The price structure is narrow and the levels are close together, which suits the shoulder season.

Front-month gas at $2.900 has established $2.933 as the session high and $2.832 as the session low, with $2.888 as the prior settlement. Recent trade has extended as low as $2.772 against a $2.794 settle. That gives a working two-week band of roughly $2.77 to $2.93 — a 5.8% range.

The five-week high near $2.90 is the immediate resistance to clear. Above it, $3.000 sits 3.45% higher and represents the psychological and modelled pivot. The September month-end projection at $3.058 requires 5.45%. The upper end of the September range at $3.320 requires 14.48%. The 2026 annual forecast at $3.44 sits 18.62% above spot.

Downside references: $2.888 (prior settle, -0.41%), $2.832 (session low, -2.34%), $2.772 (recent low, -4.41%), $2.752 (September range floor, -5.10%), $2.725 (October range floor, -6.03%) and $2.696 (November range floor, -7.03%). Below that band, $2.50 sits 13.79% down and represents the level a full surplus rebuild would justify.

The curve shape carries the real information. The September contract settled $2.842 in late August. Front-month October trades $2.900. Projections put October averaging $2.992 and closing $2.868, with November averaging $2.846. That is a curve in mild backwardation across the shoulder before the winter strip takes over.

Volume at 77,333 contracts on the front month is unremarkable, consistent with a market waiting for a catalyst rather than positioning aggressively.

The event calendar is dense. The weekly storage report publishes Thursday at 10:30 a.m. ET for the week ending August 28. The September Short-Term Energy Outlook releases September 9. The heat forecast window expires September 11. U.S. nonfarm payrolls land Friday and the FOMC decides September 15-16 — macro events that move the entire commodity complex through the dollar.

What Actually Breaks Gas Out Of This Range

The bull case has three requirements and only one is currently satisfied.

The first is LNG feedgas returning above 19 Bcf/d. Corpus Christi and Freeport resuming operations after maintenance is the necessary precondition, and it has happened. Confirmation would be feedgas prints above the 17.1 Bcf/d recorded for the week ending August 19, moving back toward the 19.6 Bcf/d record. Each incremental billion cubic feet per day of feedgas removes roughly 7 Bcf per week from the storage build.

The second is storage injections staying below 20 Bcf into late September. Four consecutive sub-20 Bcf builds would compress the five-year surplus from 167 Bcf toward 100 Bcf and force a repricing of the winter strip. The trajectory is already running that direction — 198 Bcf of surplus on August 7 became 167 Bcf on August 21.

The third is weather. Above-average temperatures across the eastern two-thirds of the country through September 11 provide a ten-day bridge. Extending that into a warm late September, or transitioning directly into an early cold snap in October, is what carries price through $3.00.

Against those, the bearish structure is formidable. Marketed production heading for a record near 123 Bcf/d with dry gas at 111.2 Bcf/d, up 3.3% on the prior record. Permian associated gas rising with $89.58 WTI regardless of gas prices. Supply growth exceeding demand growth by 0.5 Bcf/d through 2026. Storage 167 Bcf above the five-year average with ten weeks of refill season remaining. An official price forecast cut from $4.31 to $3.44 across seven months.

The balanced read is that natural gas at $2.900 is fairly priced for the current balance and cheap for the 2027 balance. The front month has limited downside because production growth is already known and inventories are already discounted, and limited upside because nothing tightens materially before the LNG ramp completes.

That is a range, not a trend — and ranges in natural gas end violently when weather intervenes.

Natural Gas Price Forecast: Levels Into The Winter Strip

Front-month Henry Hub natural gas futures trade $2.900 per MMBtu, up $0.012 or 0.42%, inside a $2.832 to $2.933 session band against a $2.888 prior settlement on 77,333 contracts. Price sits at a five-week high after the September contract settled $2.842 in late August.

The near-term bias is neutral with a mild upward tilt. Storage injections have run 15 and 16 Bcf across the last two reported weeks against a 33 Bcf five-year average, compressing the surplus from 198 Bcf on August 7 to 167 Bcf on August 21 — a 31 Bcf tightening in two weeks. Working gas stands at 3,184 Bcf, 6% above the five-year average but 1% below last year. Corpus Christi and Freeport have resumed operations after maintenance, restoring roughly 2 Bcf/d of feedgas demand. Above-average temperatures cover the eastern two-thirds of the U.S. through September 11.

The cap is production. Marketed output is heading for a record near 123 Bcf/d with dry gas at 111.2 Bcf/d, up 3.3% on the 2025 record, driven by Permian associated gas that rises with $89.58 WTI regardless of Henry Hub. Supply growth outpaces demand growth by 0.5 Bcf/d in 2026. The 2026 price forecast has been cut from $4.31 to $3.67 to $3.44.

Upside targets: $2.933 (session high, +1.14%), $3.000 (+3.45%), $3.058 (September month-end projection, +5.45%), $3.320 (September range ceiling, +14.48%) and $3.44 (2026 annual forecast, +18.62%). The 2027 forecast near $4.60 sits 58.62% above spot.

Downside targets: $2.888 (-0.41%), $2.832 (session low, -2.34%), $2.772 (-4.41%), $2.752 (September floor, -5.10%), $2.725 (October floor, -6.03%) and $2.696 (November floor, -7.03%).

The base case into October is a $2.75 to $3.10 range. Thursday's storage report for the week ending August 28 is the immediate catalyst — another sub-20 Bcf print compresses the surplus below 150 Bcf and puts $3.00 in play. A build above 30 Bcf rebuilds the cushion and sends the front month back toward $2.75.

The verdict is neutral on the front month and constructive on the winter strip. Gas at $2.900 is priced correctly for record production and a 167 Bcf surplus. It is not priced for a repeat of the event that took spot from $3.12 to $4.98 in a single week nine months ago, and it is not priced for the 2027 balance where demand growth of 2.5 Bcf/d overwhelms supply growth of 0.9 Bcf/d. Own the optionality, not the spot.

That's TradingNEWS