Natural Gas Stalls Under $3 As Record 111.5 Bcf/d Production Absorbs Every Bullish Signal

Front-month NYMEX natural gas is holding near $2.90 per MMBtu after the Energy Information Administration reported a 30 Bcf net injection into storage for the week ended August 28. The result met expectations and initially deflated futures, knocking the prompt month off Wednesday's $2.94 print — its highest level in nearly eight weeks.

The setup into the report was constructive. Futures pushed higher Wednesday as increasingly bullish September heat forecasts gave traders another reason to test the $3 mark, with renewed U.S.-Iran fighting adding a jolt from overseas. The October contract had settled at $2.888 on Friday, August 28, up 2.7% on that week, after opening at $2.844 and posting an intraday high of $2.907. Monday's session took it back to $2.85 before the recovery.

The 30 Bcf build takes working gas in storage to 3,214 Bcf, following the 3,184 Bcf reported for the week ended August 21. That prior week's 15 Bcf injection matched the market estimate exactly and left inventories 0.9% below the same week a year ago and 5.5% above the five-year average.

The number that matters is that 5.5%. On July 31 the surplus to the five-year average stood at 6.7%. On August 7 it was 6.7%. By August 14 it had narrowed to 6.2%, and by August 21 to 5.5%. Four consecutive weeks of the surplus compressing, driven by injections running consistently below seasonal norms.

Against that, Lower 48 production averaged a record 111.5 Bcf/d in August, up from July's record of 110.7 Bcf/d. LNG feedgas to the nine major export plants climbed to 18.3 Bcf/d in early September from 17.2 Bcf/d in August as Cheniere's Corpus Christi facility and Freeport LNG returned to full operations after maintenance.

The thesis for this forecast is that natural gas is the one energy market where the Iran conflict is unambiguously bullish and the price refuses to reflect it, because record domestic production is absorbing everything the demand side can generate. The $3.00 handle has capped this market all summer, and it is a production ceiling rather than a demand failure.

Above $3.00 the market has room. Below $2.85 it has an inventory problem that has not gone away.

Inventories At 3,214 Bcf And A Surplus Shrinking Four Weeks Running

The storage trajectory is the strongest bullish argument available, and it has been building quietly for a month.

Working gas stands at 3,214 Bcf after today's 30 Bcf addition. The surplus to the five-year average has narrowed from 6.7% at the end of July to 5.5% as of the August 21 report, and inventories sit 0.9% below the same week last year — an unusual configuration for a market described all summer as oversupplied.

The injection series tells the story better than the level does. For the week ended July 3, operators added 61 Bcf against estimates of 49 to 51 Bcf, a bearish surprise that outpaced expectations despite record heat and sank the prompt month from $3.101 to $3.059. Since then: 28 Bcf for the week ended July 24, 33 Bcf for July 31 against a 31 Bcf expectation, 36 Bcf for August 7 against 31 expected, 16 Bcf for August 14 against 19 expected, 15 Bcf for August 21 in line with 15 expected, and 30 Bcf today.

The four-week average through today runs 24.25 Bcf. The comparable five-year averages for those weeks ran meaningfully higher — 29 Bcf for the week matching August 14 alone.

That gap is what compressed the surplus, and it happened while production was setting monthly records. Demand, not supply constraint, did the work.

The forward arithmetic is where this gets interesting. The federal forecast expects inventories to reach a record 3,985 Bcf at the end of October 2026, 5% above the five-year average and the highest level heading into winter since 2016. From 3,214 Bcf, reaching 3,985 requires 771 Bcf across the nine weeks to October 31 — an average of 85.7 Bcf per week.

Recent builds have averaged 24.25 Bcf. Shoulder-season injections do rise sharply as cooling demand fades, and 70 to 90 Bcf weekly builds are normal for late September and October. But at 60 Bcf a week the season ends near 3,754 Bcf — 231 Bcf below the official target.

A miss of that size on the end-of-October number is the single most bullish scenario available to this market.

Record 111.5 Bcf/d Of Lower 48 Production Is The Ceiling

Every rally attempt this year has died on the same wall, and it is supply.

Lower 48 production averaged a record 111.5 Bcf/d in August, surpassing July's record of 110.7 Bcf/d. That is 0.8 Bcf/d of month-over-month growth on an already record base, and it has been the consistent counterweight to every bullish demand signal the market has produced.

The mechanism is straightforward. Associated gas from oil-directed drilling scales with crude production, and crude at $91.98 for West Texas Intermediate is a price that keeps every basin running flat out. Permian Basin natural gas takeaway is reaching the point where the market can clear, which unlocks more oil production and the associated gas that comes with it.

That is a structural problem for gas bulls. Higher oil prices increase gas supply without increasing gas demand, which means the Iran conflict simultaneously tightens the crude market and loosens the gas market on the supply side.

The federal outlook has been marking down its price forecast in response. The August update cut the third-quarter Henry Hub spot forecast to $2.87 per MMBtu, down 50 cents from the prior month, citing reduced LNG feedgas demand and record natural gas production. It expects prices to remain below $3.00 until November and average $3.03 over the final five months of the year — nearly 50 cents lower than the previous estimate.

Contracts through September 2026 have been trading below $3.00, exactly as that forecast implies.

The comparison to January's outlook shows how much has changed. That forecast had Henry Hub averaging just under $3.50 for 2026 with supply growth outpacing demand growth by 0.5 Bcf/d. Production has run substantially ahead of that assumption, and the price has followed.

Production is the variable that does not respond to weather, war or storage. It responds to drilling economics, and at current oil prices those economics are excellent. Until a rig-count decline shows up in the data, $3.00 remains a ceiling rather than a waypoint.

LNG Feedgas At 18.3 Bcf/d Is The Floor

The demand side has finally started delivering, and the numbers have moved fast.

Average gas flows to the nine major LNG export plants rose to 18.3 Bcf/d in early September, up from 17.2 Bcf/d in August, as Cheniere Energy's Corpus Christi facility and Freeport LNG in Texas returned to full operations following maintenance. That is 1.1 Bcf/d of incremental demand appearing inside two weeks.

Freeport's maintenance began July 10 and was expected to complete in late August, affecting 2.0 Bcf/d of nominal export capacity. Its return alone accounts for the bulk of the improvement, and feedgas flows had already reached their highest level since late June on completion.

The daily numbers remain volatile. Wednesday's nominations put U.S. LNG feedgas at roughly 17.37 Bcf/d, a decline of about 892 MMcf/d from Tuesday, which is the kind of day-to-day swing that makes single readings unreliable and multi-week averages meaningful.

The official third-quarter forecast has LNG exports averaging 16.5 Bcf/d, down 0.2 Bcf/d from the prior estimate. Early September flows at 18.3 Bcf/d run 1.8 Bcf/d above that. If the fourth quarter sustains anything near the current rate, the export assumption underpinning the bearish price forecast is too low.

Pipeline exports add a second leg. Total U.S. pipeline exports are forecast to average 9.6 Bcf/d in 2026 and rise to 10.0 Bcf/d in 2027, up from 9.5 Bcf/d in 2025. The Energia Costa Azul terminal on Mexico's Pacific coast, supplied from the Permian Basin, shipped its first cargo on July 8, bringing 0.4 Bcf/d of nominal export capacity online. Mexican demand has also increased as new gas-fired power plants ramp.

One caveat on the Mexican leg: President Claudia Sheinbaum said in her state of the union address Tuesday that Mexico is targeting a reduction in U.S. natural gas imports by bolstering domestic infrastructure and unconventional development. That is a multi-year risk to 9.6 Bcf/d of guaranteed demand.

Combined export demand at roughly 28 Bcf/d against 111.5 Bcf/d of production is a quarter of the market. It is the only demand line growing structurally.

The Hormuz Premium That Has Not Reached Henry Hub

The Iran conflict has repriced crude, diesel and international gas. It has done almost nothing to American gas, and that gap is the trade.

Escalating U.S.-Iran hostilities this week raised fresh concerns over prolonged disruptions to energy shipments through the Strait of Hormuz. The flare-up added to already simmering concerns about global LNG supply and could add demand for American exports — and affect pricing — should the war drag into the winter months.

The mechanism is documented. In July, international prices rose to levels last reached in early April as LNG vessel traffic through the Strait slowed considerably after strikes on vessels resumed on July 7. Qatar, the world's largest LNG exporter after the United States, ships essentially all of its cargoes through Hormuz.

Brent trades at $96.20, up more than 7% on the week after breaking $97 intraday. The ICE gasoil crack hit a record $79 per barrel and the U.S. diesel crack sits above $100. Every energy product except American natural gas is carrying a substantial war premium.

The reason is that the United States is structurally long gas and short nothing. Record production at 111.5 Bcf/d, storage at 3,214 Bcf, and export capacity constrained by liquefaction terminals rather than by molecules means a global supply shock lifts the international price and the U.S. price barely responds. American gas trades in an island market with a one-way valve to the world.

That said, the valve is opening. Export capacity is the binding constraint, and it grows every quarter. Even with Freeport fully operational, exports remain limited by slow growth in additional capacity despite elevated U.S. price spreads to Europe and Asia caused by the ongoing disruptions.

The asymmetry for traders: if the war escalates further and Qatari cargoes stop entirely, global buyers bid for every available U.S. cargo, terminals run at maximum utilization, and feedgas demand goes from 18.3 Bcf/d toward capacity. That is the scenario where Henry Hub breaks $3.00 decisively.

If Hormuz reopens, international prices fall, the arbitrage narrows, and nothing about the domestic balance changes.

September Heat Could Be The Hottest On Record

Weather is doing more for this market right now than any structural factor, and the forecasts are unusually aggressive.

September is potentially becoming the hottest on record, and increasingly bullish heat forecasts gave traders their reason to test the $3 mark on Wednesday. Temperatures are expected to remain mostly above normal across the central United States through mid-September, keeping demand for gas-fired power generation elevated. Hot conditions across parts of the South, East and East Coast are projected to sustain air-conditioning demand.

That matters more in September than it does in July. Late-summer heat suppresses storage injections at exactly the point in the calendar when the market is calculating its end-of-October inventory number. Every Bcf that goes into a power plant in September is a Bcf that does not go into a salt cavern, and the difference compounds across the nine weeks remaining in the injection season.

The recent data already reflects it. Injections of 16 Bcf and 15 Bcf for the weeks ended August 14 and August 21 came in below both estimates and five-year averages, and they are the reason the surplus compressed from 6.7% to 5.5%.

One forecast has shown a reduction in cooling-degree days pointing to some moderation, which is the standard risk with any weather-driven thesis — models revise, and gas prices revise with them.

The seasonal transition is the harder problem. Cooling demand collapses in October regardless of how hot September runs, and heating demand does not begin in earnest until November. That shoulder window is when injections normally run 70 to 90 Bcf weekly, and it is the period that determines whether the market approaches the 3,985 Bcf record target.

The bull case needs September heat to persist long enough that the shoulder-season catch-up cannot close the gap. At 60 Bcf per week for the remaining nine weeks, the season ends near 3,754 Bcf against a 3,985 Bcf forecast.

That 231 Bcf shortfall would move the winter strip meaningfully, because it changes the starting inventory for a heating season already facing a war-disrupted global LNG market.

Tropical Storm Edouard And The Hurricane Risk Nobody Priced

The supply side carries a seasonal risk that has been dormant and is now live.

Tropical Storm Edouard made landfall near the Texas-Louisiana border, with no significant operational disruptions reported at LNG facilities in the region so far. That is the good outcome, and it is also the reminder that the U.S. hurricane season is in full swing and that the entire American liquefaction complex sits on a stretch of Gulf Coast measured in a few hundred miles.

Sabine Pass, Plaquemines, Corpus Christi, Freeport, Cameron and Calcasieu Pass all sit in the path of any storm tracking into the western Gulf. A direct hit that forces a multi-week shutdown removes several Bcf/d of demand instantly and is bearish for Henry Hub — feedgas that cannot be liquefied gets rerouted into storage.

The opposite risk is offshore and onshore production disruption in the same region, which is bullish. Historically the demand-side hit from terminal outages has outweighed the supply-side hit from production shut-ins, because Gulf of Mexico production is a small share of the 111.5 Bcf/d Lower 48 total while the LNG terminals represent nearly all export capacity.

That asymmetry means hurricane headlines in this market are usually bearish for the prompt month, which is the opposite of how crude reacts.

Lower storage injections could provide bullish fuel if a storm disrupts production without touching terminals, but the base case for any Gulf landfall is feedgas down, injections up, price lower.

The seasonal window runs through November. With inventories at 3,214 Bcf and the market debating whether the end-of-October number lands at 3,754 or 3,985, a single storm that takes 2 Bcf/d of liquefaction offline for three weeks adds roughly 42 Bcf to storage and settles the argument in the bears' favour.

Edouard passed without damage. The next one is the variable that cannot be modelled, and it deserves a place in position sizing rather than in a forecast.

The counterweight: LNG terminals have become far more resilient operationally, and Freeport's July-to-August maintenance already demonstrated the market absorbs 2.0 Bcf/d of lost capacity without breaking.

The Official Model: $2.87 In Q3 And A Record 3,985 Bcf In October

The federal forecast is the reference case, and it is bearish in a way that has been consistently correct this year.

The August outlook, released August 11 with the next update due September 9, expects the Henry Hub spot price to average $2.87 per MMBtu in the third quarter — down 50 cents from the July estimate — citing reduced LNG feedgas demand and record production. It projects prices remaining below $3.00 until November and averaging $3.03 across the final five months of 2026, nearly 50 cents below the prior forecast.

It expects inventories to reach a record 3,985 Bcf at the end of October, 19 Bcf higher than the July estimate and 5% above the five-year average, leaving storage at its highest level entering winter since 2016.

The forecast has been directionally right and repeatedly revised lower. January's outlook had Henry Hub averaging just under $3.50 in 2026. The August version has the third quarter at $2.87. That is a 63-cent markdown across seven months, driven almost entirely by production exceeding assumptions.

The September 9 update is a genuine catalyst. Two things will move the market inside it. If the end-of-October inventory projection gets revised downward from 3,985 Bcf to reflect the below-average injections of the past month, the bearish anchor weakens materially. And if the LNG export assumption of 16.5 Bcf/d for the third quarter gets raised toward the 18.3 Bcf/d flows now running, the demand side of the balance improves.

Both revisions are plausible on the current data. Neither is guaranteed.

The historical relationship worth holding onto: inventories ending October above 3.9 Tcf sit at a level historically correlated with sub-$3.00 pricing. That is the mechanical reason the market cannot break the handle while the official forecast points at 3,985 Bcf.

Break that forecast and the correlation breaks with it.

At $2.90 the prompt month trades 1% above the $2.87 third-quarter estimate and 4.3% below the $3.03 five-month average, which means the curve already prices a modest recovery into the fourth quarter.

The 2027 Setup: Supply Falls Behind Demand By 1.6 Bcf/d

The forward curve tells a completely different story from the prompt month, and it is the strongest argument for owning this market on any weakness.

The official forecast has annual average Henry Hub spot prices decreasing 2% in 2026 and then increasing 33% in 2027 to just under $4.60 per MMBtu. The mechanism is a supply-demand inversion: supply growth outpaces demand growth by 0.5 Bcf/d in 2026 but falls behind by 1.6 Bcf/d in 2027.

The driver is LNG. Feed gas demand from U.S. liquefaction facilities grows faster than production in 2027, reducing storage and putting sustained upward pressure on price. That is a capacity story with construction schedules attached — new trains come online on timelines that are known years in advance, unlike weather or wars.

The arithmetic on a 2 Bcf/d swing in the balance is substantial. Against a market that consumes and exports roughly 115 Bcf/d, a 1.6 Bcf/d deficit is 1.4% — which sounds small until you compound it across a year and translate it into storage. Persistent deficits of that size draw inventories below the five-year average, and the price relationship between storage and Henry Hub is close to mechanical.

Periods with higher-than-average inventories are generally associated with lower prices, and lower storage levels correspond with higher prices and tighter conditions. As inventories move toward or below the five-year average, the forecast price rises.

That is exactly the transition 2027 is expected to deliver, and the current storage surplus compressing from 6.7% to 5.5% over four weeks is the first evidence it may arrive early.

Residential and commercial consumption is forecast to decrease 4% in 2026 to 22.1 Bcf/d on closer-to-normal temperatures against a colder 2025. That is a demand headwind for this year that reverses on any cold winter.

For traders, the implication is calendar structure rather than flat price. The 2027 strip at just under $4.60 against a prompt month near $2.90 is a 59% contango, and the mechanism for closing it is documented in construction schedules rather than in weather models.

International Spreads And Why American Gas Stays Cheap

The gap between U.S. and international gas prices is the single largest arbitrage in global energy, and it explains everything about Henry Hub's ceiling.

Front-month LNG cargo prices in East Asia have run in the $9.61 to $10.73 per MMBtu range in recent reference periods, with Dutch Title Transfer Facility futures between $9.72 and $12.40. Henry Hub sits near $2.90. That is a spread of $7 to $9 per MMBtu, and it persists because the physical link between the two markets is liquefaction capacity that cannot expand on a trading horizon.

The war has widened it. In July, international prices rose to levels last reached in early April as LNG vessel traffic through the Strait of Hormuz slowed considerably following the resumption of strikes on vessels on July 7. U.S. price spreads to Europe and Asia remain elevated due to ongoing disruptions.

That should be maximally bullish for American producers, and to an extent it is — every terminal is running at maximum utilization and feedgas has climbed to 18.3 Bcf/d. But maximum utilization is a fixed number, and once it is reached additional international demand cannot pull another molecule out of the United States.

Which is why Henry Hub does not respond to Hormuz. The constraint is steel, not gas.

European storage dynamics add the winter variable. Reduced EU inventories have historically been a key driver of rising TTF prices, and a European heating season that begins with disrupted Qatari supply and depleted storage would produce international prices far above current levels. That does not lift Henry Hub directly, but it guarantees U.S. terminals run flat out through the winter — which locks in 18-plus Bcf/d of demand regardless of American weather.

That is the quiet floor under this market. Export demand at maximum capacity is price-insensitive and weather-insensitive, and it removes the scenario where a mild winter collapses Henry Hub the way it did in prior cycles.

Thirty-eight vessels with a combined 143 Bcf of carrying capacity departed U.S. ports in a single recent week. That is roughly 20 Bcf/d of loaded cargo leaving the country.

Technical Structure: The $3.00 Ceiling And The $2.85 Floor

The chart is a range that has held since midsummer, and both edges are well tested.

The prompt month holding near $2.90 sits between two levels that have defined the last two months. Wednesday's $2.94 print was the highest in nearly eight weeks and represents the top of the current attempt. The $3.00 handle above it is the level the market has been described as testing repeatedly without clearing — a psychological round number that also coincides with where the July contract broke down.

Below, $2.85 is the level Monday's session found, and the October contract's $2.888 settlement on August 28 sits between the two. The April reference at $2.626 marks the low end of the year's trading.

The pattern of failed attempts is instructive. Futures rallied 14.0 cents on one recent Tuesday amid a strengthening weather and LNG outlook before plunging 16.5 cents the next day as technical resistance stymied momentum. That is a market where every rally into resistance gets sold by producers hedging record output.

Today's price action followed the template. The prompt month advanced Wednesday on heat forecasts and the Iran jolt, the 30 Bcf print landed in line, and futures deflated on the release.

Measured from $2.90: $2.94 is 1.4% higher, $3.00 is 3.4%, $3.03 is 4.5% and $3.20 is 10.3%. Downside: $2.85 is 1.7% below, $2.75 is 5.2% and $2.626 is 9.4%.

Those distances are small in absolute terms, which is characteristic of a compressed market. Natural gas volatility expands violently once a range breaks, and a move through $3.00 with the storage surplus continuing to compress would likely travel to $3.20 quickly rather than grinding.

The confirmation signal is a weekly close above $3.00 paired with an injection below 50 Bcf in the shoulder season. One without the other is noise.

The invalidation is a weekly close below $2.85 combined with a build above 80 Bcf, which would put the 3,985 Bcf October target back on track and take the prompt month toward the mid-$2.70s.

Equity Expression: The Producers Have Outrun The Commodity

The equity market has been pricing the 2027 setup rather than the 2026 tape, which is where the sector's opportunity and its risk both sit.

Energy has led the S&P 500 in 2026 with a 43% gain and led the third quarter at 22%, with the sector fund up 47.3% year to date against 27.7% for technology. That performance has come overwhelmingly from crude exposure, with Brent at $96.20 and refining margins at records.

Gas-weighted producers have a different driver. Companies levered to Henry Hub have been trading the forward curve — a 2027 strip at just under $4.60 against a prompt month near $2.90 — rather than the spot price. That is rational: a producer's value is the discounted present value of a decade of production, not next month's settlement.

The risk in that framing is that the same record production driving 111.5 Bcf/d is coming from these companies. Every producer benefits from a higher price and every producer is contributing to the supply that suppresses it. Permian associated gas, in particular, comes out of the ground as a byproduct of oil economics that remain excellent at $92 crude — meaning gas-directed producers are competing against supply that does not care what gas costs.

The infrastructure and export names carry cleaner exposure. Liquefaction operators earn tolling fees on capacity utilization rather than on the commodity spread, and with feedgas at 18.3 Bcf/d and international prices $7 to $9 above Henry Hub, every terminal is running at maximum. That is a volume business operating at capacity in a market where capacity is the binding constraint.

The leveraged ETF products deserve a warning. Daily-rebalanced instruments tracking natural gas futures decay in a rangebound market, and this market has been rangebound between $2.85 and $3.00 for weeks. Holding period matters more than direction in those vehicles.

Coal provided a useful cross-check today. Met coal names fell hard — Alpha Metallurgical down 5.31%, Peabody 5.12%, Warrior Met 4.78% — on the same dovish Fed repricing. The inflation trade is unwinding, and gas equities levered to it face the same rotation.

Verdict And Forecast: $3.20 Above $3.00, $2.75 Below $2.85

Natural gas near $2.90 after an in-line 30 Bcf injection deserves a constructive stance with the range respected and the September 9 forecast update as the near-term catalyst.

The bull case has been building quietly for a month. The surplus to the five-year average has compressed from 6.7% on July 31 to 5.5% on August 21 across four consecutive weeks of below-normal injections, and inventories at 3,214 Bcf sit 0.9% below the same week a year ago. Reaching the official 3,985 Bcf end-of-October target requires 85.7 Bcf per week for nine weeks against a recent four-week average of 24.25 Bcf; at 60 Bcf weekly the season ends near 3,754 Bcf, a 231 Bcf miss. LNG feedgas has jumped to 18.3 Bcf/d in early September from 17.2 Bcf/d in August as Corpus Christi and Freeport returned to full operations, running 1.8 Bcf/d above the official third-quarter assumption. September may be the hottest on record with above-normal temperatures across the central United States through mid-month. And Hormuz disruption has widened the U.S. spread to Europe and Asia to $7 to $9 per MMBtu, guaranteeing terminals run flat out through winter. The 2027 strip at just under $4.60 prices supply falling behind demand by 1.6 Bcf/d.

The bear case is one number and it is decisive. Lower 48 production hit a record 111.5 Bcf/d in August after July's record 110.7 Bcf/d, and at $91.98 crude every basin runs flat out while Permian takeaway constraints clear. That supply does not respond to weather or war. The official forecast has third-quarter Henry Hub at $2.87 and inventories entering winter at their highest since 2016, a level historically correlated with sub-$3.00 pricing. Hurricane season through November is net bearish because terminal outages remove more demand than production shut-ins remove supply. And Mexico has stated it intends to reduce U.S. gas imports.

The forecast: rangebound between $2.85 and $3.00 into the September 9 update. A weekly close above $3.00 paired with a shoulder-season injection below 50 Bcf opens $3.20, worth 10.3%, and would confirm the end-of-October target is being missed. A weekly close below $2.85 with a build above 80 Bcf targets $2.75 and puts $2.626 back in view. Watch three things: the weekly injection against the 85.7 Bcf/week pace required, LNG feedgas holding above 18 Bcf/d, and whether the September 9 forecast cuts its 3,985 Bcf projection. Verdict: constructive on weakness, no chasing into $3.00.

That's TradingNEWS