NG Breaks $2.659 Support To $2.642 As Production Hits 111 Bcf/d And Hugh Brinson Adds 1.5 Bcf/d September 1

September natural gas futures traded $2.642 per MMBtu Thursday morning, down 4.6 cents or 1.71%, after printing a fresh multi-month low earlier in the week. A separate benchmark read put the contract at $2.67, down 0.84% — the lowest level in more than three months. The front-month sat at $2.751 against a $2.781 prior settlement, down 1.08% on 6,105 contracts.

The decline is not a wobble. Over the past month the price has fallen 18.36%. Against the same point a year ago it is down 13.09%. Against the two-year high near $4.55 tested in May, the contract has lost 41.9%. Against the January spike to $4.875 on the February contract, it is down 45.8%.

The session structure told the story before the data arrived. The market opened in a downtrend on both the swing chart and the 50-day moving average, then sellers took out Tuesday's low at $2.659 and reaffirmed the trend. That break happened ahead of the 10:30 a.m. ET storage report, which is the signature of a market that had already decided what the number would say.

Then the number landed and confirmed it. A net injection of 33 Bcf for the week ended July 31 came in on the heavier side of consensus expectations for 31 Bcf, and a hefty storage build depressed bullish sentiment and weighed on futures at midday.

Every input is aligned against the bid. Production near 111 Bcf per day. Storage running 6.4% above the five-year average. Cooler forecasts across the Midwest and Northeast. Liquefaction feedgas at roughly 18 Bcf per day, below the spring peak. New Permian pipeline capacity arriving September 1. A market where the heat is in the wrong regions.

The rally attempts have all failed at the same place. Short-covering pushed futures to $2.991 ahead of a report in mid-July and the number killed the move right at the top of the range. A 28 Bcf build below estimates produced a rally that did not survive the week. One moderate number forces shorts to cover for a session; it does not fix a surplus built over months.

Buyers had reasons to rally this week and could not hold any of them. That is the whole forecast in a sentence.

The 33 Bcf Injection Landed Heavy And Killed The Bid

The weekly storage number was the immediate catalyst and it broke the wrong way. Working gas rose 33 Bcf for the week ended July 31, against a consensus for 31 Bcf. That two-Bcf overshoot sounds trivial and it is not, because the direction of the surprise determines whether shorts have to cover.

The threshold was clearly defined going in. A build above 31 Bcf tells sellers the balance is still loose and the downtrend continues. A number near the estimate does nothing for buyers because inventories are already comfortable and the trend is working against them. Buyers needed repeated tight builds paired with hotter weather and rising feedgas. They got a heavy print in isolation.

The context that makes 33 Bcf bearish rather than neutral is the calendar. That build is smaller than the prior week's injection but still above the five-year average for early August. A market injecting above seasonal norms in the peak cooling month has no mechanism to close a surplus before the withdrawal season begins.

The run of recent builds shows how inconsistent the tightening has been. The week ended June 26 delivered 87 Bcf against expectations clustered between 79 and 83. The week ended July 10 produced 41 Bcf, short of expectations and historical norms, and futures retreated anyway. The week ended July 17 injected 32 Bcf against a 29 Bcf estimate. An earlier July week printed 61 Bcf. The most recent prior report came in at 28 Bcf, below estimates, and the rally it produced lasted days.

That sequence averages roughly 46 Bcf a week across five reports in a period when power burn should be consuming everything production can deliver. Summer demand is doing some work — the 32 Bcf print was less than half the 87 Bcf figure earlier in the season — and the surplus is not shrinking fast enough to change the direction of the market.

The next report covers the week ending August 7 and lands next Thursday. For it to matter, it has to surprise materially to the tight side, not marginally. A single lean number against a five-year surplus this wide does not give buyers enough ammunition to push through resistance.

The pattern is now established across six weeks: sellers use every hot-weather rally to reload.

Storage At 6.4% Above The Five-Year Average

The surplus is the structural fact that overrides everything else on this board. Working gas inventories are running approximately 6.4% above the five-year average, with a separate estimate putting stocks around 6.6% above normal for the week ended July 31.

That cushion has been in place since March, built by strong output and mild spring conditions, and it has not eroded through the hottest weeks of the year. Inventories stood near 3,056 Bcf as of the week ending July 17, roughly 6.4% above the five-year average and slightly below year-ago levels. At the end of June, stocks were 6% above the five-year average.

June at 6% and late July at 6.4% is the number that kills the bull case. Peak cooling demand across an entire month failed to reduce the surplus — it widened it by four-tenths of a percentage point. Power burn is the only demand lever capable of drawing inventories in summer, and it has not been sufficient against production near 111 Bcf per day.

The forward projection compounds the problem. Working inventories are forecast to reach 3,966 Bcf by the end of October, 5% above the five-year average, with above-average stocks heading into winter. Storage remaining above the five-year average through much of the forecast horizon limits upward price pressure structurally, because inventories stay relatively high while record production, led by Permian growth, meets rising demand.

That is the definition of a market with no scarcity premium available. Entering the withdrawal season at 3,966 Bcf and 5% above normal means winter has to deliver a genuine cold shock before storage becomes a bullish variable, and the market will not pay for that possibility in August.

The counterweight to a 5% end-October surplus is that it is narrower than the current 6.4%. The projection embeds some tightening across August, September and October, driven by liquefaction demand rather than weather. If that tightening runs faster than modelled — through feedgas recovering from maintenance and export capacity ramping — the surplus compresses into a winter where a cold month produces violent upside.

That is the trade for the fourth quarter. It is not the trade for August, where the surplus is a wall.

Production Near 111 Bcf/d Is The Ceiling

Supply is the reason none of the demand improvements have mattered. Output is running near 111 Bcf per day, and record production continues to meet rising demand while putting moderate downward pressure on prices.

Put that number against the demand stack. Liquefaction feedgas at roughly 18 Bcf per day represents 16.2% of production leaving the domestic system. Even at the spring peak near 19 to 20 Bcf per day, exports absorb under a fifth of output. The remainder has to be consumed domestically or injected into storage, and domestic consumption is capped by weather.

The trajectory has been relentless. Output was estimated to average 109.5 Bcf per day over a seven-day window in June, topping the monthly pace by half a Bcf, with production running strong specifically to meet liquefaction needs. From 109.5 Bcf to 111 Bcf in six weeks is another 1.5 Bcf per day of supply added into a market that already carried a 6% surplus.

Permian growth is the identified driver, and that growth is not price-responsive at these levels because the gas is associated production from oil wells. Producers targeting crude economics will keep bringing gas to market regardless of whether Henry Hub trades $2.64 or $3.50, which removes the supply-response mechanism that normally puts a floor under a commodity.

The rig count and hedging behaviour matter less than they did in prior cycles for the same reason. Associated gas from oil-directed drilling does not shut in when gas prices fall. It shuts in when crude prices fall — and September crude at $76.13 is nowhere near a level that curtails Permian activity.

That structural feature is why the surplus has been durable rather than cyclical. Production has not slowed down through a 41.9% price decline from the May high at $4.55, and there is no mechanism visible that would make it slow before the end of the injection season.

The one offsetting factor is that record supply is also what makes the market vulnerable to a demand shock. A system running 111 Bcf per day with 6.4% surplus storage and a 41.9% drawdown from its high has almost no length in it — which means a genuine cold snap or a feedgas surge produces a squeeze rather than an orderly repricing.

The Heat Keeps Missing The Regions That Matter

Weather is doing exactly the wrong thing, and the geography of it is the entire problem.

The West and South are hot. That is not the trade. Natural gas needs sustained heat across the Midwest and Northeast, where population density and air-conditioning load can move power burn fast enough to change the storage math. Recent forecasts shifted cooler across both regions and the market reacted immediately.

That distinction gets lost in headlines about record temperatures. Texas running hot with grid operators hitting load records generates real gas burn, and it does not generate enough to offset 111 Bcf per day of production plus above-average storage. The Southeast and Southwest were already priced for summer heat. The marginal demand that moves the national balance lives in the corridor from Chicago to Boston.

Forecasts calling for cooler temperatures across the central and eastern United States have weighed on prices by reducing expected air-conditioning demand. Moderating temperatures across much of the country in the coming weeks reduce the likelihood of a significant increase in consumption.

Early August heat spreading east is the one catalyst that could shift the balance, and the forecast has not delivered it. A hot revision across the East can still produce a sharp short-covering move — the market has demonstrated repeatedly this summer that it will rally on a forecast shift. The current outlook does not support it.

Regional price behaviour through August bidweek illustrated the split precisely: prices popped across the West as lofty temperatures boosted cooling demand from Texas to California, while a sharp pullback in the East countered the move and kept the national average in check.

For the forecast, weather is a two-sided risk with an asymmetric current position. Positioning is short, the surplus is wide, and the trend is down — which means a hot eastern revision produces a violent short-covering rally toward $2.810 and possibly $2.979. Absent that revision, cooler eastern forecasts remove the last demand support and $2.592 comes into play.

Seasonality works against buyers from here regardless. Cooling demand peaks in late July and early August, then fades through September while injection capacity remains. Every week without eastern heat is a week the surplus widens into the shoulder season.

Cash Prices Confirm What The Forecast Shows

The physical market is corroborating the futures decline, which removes the argument that this is a positioning-driven selloff detached from fundamentals.

Cash prices weakened across Texas, the Gulf Coast and parts of the Midwest earlier this week. The physical market is confirming what the forecast is showing: daily demand is not tight enough to challenge the futures selloff.

That alignment matters because cash and futures diverging is how a bottom typically forms. When physical buyers pay premiums to prompt futures, it signals genuine tightness that the paper market has not priced. When cash weakens alongside futures, there is no hidden demand waiting to be discovered — the market is loose in both time frames simultaneously.

The regional dispersion has been extreme this year, which is a separate story from the direction. Prices have ranged from a decrease of $1.49 per MMBtu at one northeastern hub to an increase of $2.08 at the Houston Ship Channel across a single report week, with a majority of U.S. hubs printing above $4.00 per MMBtu during the January cold event. That volatility reflects pipeline constraints and local weather rather than the national balance.

Infrastructure additions have compounded the pressure at the benchmark itself. Increased deliveries toward Louisiana following the completion of a Texas-Louisiana expansion project added supply pressure directly to Henry Hub spot prices. Physical gas arriving at the delivery point is the most direct mechanism by which pipeline capacity translates into a lower futures price.

The read-through for the forecast is that the weakness is real rather than technical. A market where the front month has broken multi-month lows, cash has weakened in three separate regions, storage is 6.4% above average and production is at 111 Bcf per day has no hidden bullish variable in the physical layer.

What would change that is nominations data. The export story turns bullish when liquefaction plants come back from maintenance and start pulling harder, and the market will not price that until it shows up in daily nominations. Watching feedgas nominations rather than headlines is the operative discipline through August.

LNG Feedgas At 18 Bcf/d Below The Spring Peak

Liquefaction demand is the strongest thing buyers have and it is not strong enough. Feedgas is running near 18 Bcf per day — a large volume leaving the domestic market, and below the summer peak pace near 19 Bcf per day and well below records north of 20 Bcf per day.

The trajectory through the summer has been choppy rather than upward. Flows to major export terminals averaged 17.2 Bcf per day at one point, down from 17.4 Bcf in June, partly because of scheduled maintenance at a major Texas liquefaction facility. A seven-day estimate put feedgas near 17.5 Bcf per day in mid-July. Exports were projected to average 18.5 Bcf per day around the Juneteenth holiday, more than 3 Bcf per day above year-earlier levels.

So feedgas has improved from 17.2 to roughly 18 Bcf per day, and it remains below the spring peak. That gap of 1 to 2 Bcf per day, held across the injection season, is 30 to 60 Bcf of monthly storage — enough to explain why the surplus has widened rather than narrowed.

Maintenance is the identified cause and it is temporary. Summer maintenance at the Texas facility and others limited how much gas could leave the country. Export terminals' consumption of feedgas climbed in June as seasonal maintenance events culminated. The capacity exists; the utilisation does not.

One major exporter raised its full-year outlook Thursday on strong export demand, and the domestic market did not care, because the gas is not leaving fast enough. That is the cleanest statement of the disconnect available: forward export demand is confirmed and rising, and the physical pull on Henry Hub inventories is running below what the price needs.

The bull case rests entirely on that gap closing. Strong exports could quickly erase the surplus by fall — a genuine possibility given that returning maintenance capacity plus new trains could push feedgas from 18 Bcf toward 20 Bcf per day, which at 2 Bcf per day incremental is 60 Bcf a month of additional draw.

Sixty Bcf a month from mid-August through October is 150 Bcf, against a projected end-October surplus of 5%. That arithmetic is how $2.642 becomes $3.50 by November. It requires nominations to confirm first.

July Exports Slipped To 10.48 Million Tonnes

The export data delivered the most damaging single datapoint of the week, and it directly undercuts the liquefaction bull case.

July exports actually slipped to 10.48 million metric tons from 10.6 million in June — a decline of 120,000 tonnes, or 1.1% — while overseas prices were screaming for more supply.

That is the sentence that matters. Global buyers wanted U.S. cargoes. Asian and European price signals supported maximum export volumes. And American shipments fell month over month, because summer maintenance at multiple facilities capped physical throughput regardless of demand.

Read against the price action, that explains why the domestic market has ignored every bullish export headline. Forward demand is real; delivered volume declined. A market that cannot lift exports when overseas prices are at premiums has a supply-side constraint at the liquefaction stage, and that constraint is the reason gas keeps accumulating in domestic storage.

The regional demand picture supports the case that the pull is there. Hot weather across South Korea and northeastern China has helped preserve Asia's price advantage for flexible U.S. cargoes, though a cooler European forecast and typhoon activity have complicated the arbitrage. Demand exists in both basins; the routing decision depends on netbacks that shift weekly.

The forward math is genuinely constructive once maintenance clears. Ten point four eight million tonnes a month annualises to roughly 126 million tonnes. Feedgas at 18 Bcf per day sustained annualises considerably higher, and the record pace north of 20 Bcf per day would represent another meaningful step. Every incremental Bcf per day of feedgas is roughly 30 Bcf a month removed from the domestic balance.

The problem is timing against the calendar. Cooling demand peaks now and fades through September. If feedgas recovery arrives in late August and September, it offsets fading power burn rather than adding to it — which stabilises the balance without tightening it.

For the surplus to close before winter, feedgas has to exceed 20 Bcf per day and hold. That has not happened, and the market will not price it until daily nominations show it.

Hugh Brinson At 1.5 Bcf/d Arrives September 1

The bearish structural development with a fixed date is the one nobody can hedge around. The Hugh Brinson pipeline reaches full capacity at 1.5 Bcf per day by September 1 — more Permian gas heading straight to Henry Hub just as summer demand fades.

The timing could not be worse for buyers. September 1 falls immediately after peak cooling demand, at the start of the shoulder season, and delivers additional supply directly to the pricing point rather than into a regional market. That is 45 Bcf a month of incremental deliverability arriving as power burn declines.

Against a projected end-October inventory of 3,966 Bcf at 5% above the five-year average, two months of 1.5 Bcf per day represents roughly 90 Bcf — enough on its own to push the October surplus back toward the current 6.4% level rather than the modelled 5%.

The infrastructure story has been consistent all year. A Texas-Louisiana expansion project brought to full service stepped up deliveries toward Louisiana and added supply pressure to Henry Hub spot prices. Each new line that connects Permian associated gas to the benchmark reduces the basis discount in West Texas and raises the effective supply at the delivery point.

That is the mechanism by which record production translates into a weak Henry Hub price rather than merely a weak Waha price. Permian gas that used to be stranded, flared or sold at negative basis now reaches the national benchmark, and it competes directly with Appalachian and Haynesville supply for the same storage and export demand.

The forecast implication is that liquefaction growth has to outrun pipeline growth for the price to recover. Feedgas moving from 18 Bcf to 20 Bcf per day adds 2 Bcf per day of demand. Hugh Brinson at full capacity adds 1.5 Bcf per day of supply. Net tightening of 0.5 Bcf per day is 15 Bcf a month — a rounding error against a surplus running above 200 Bcf.

Liquefaction demand has to fight existing production, rising pipeline capacity and storage that is already above average. Three headwinds, one tailwind, and the tailwind is on maintenance.

European Storage At 58% Is The Bull's Only Card

There is one genuinely bullish datapoint in this market and it sits 4,000 miles from Henry Hub. European storage is 58% full heading into winter, well below normal for early August.

That figure matters because European inventories are the demand of last resort for U.S. cargoes. A continent entering the withdrawal season at 58% capacity has to buy aggressively through autumn or accept price risk in January, and the marginal molecule comes from American liquefaction.

Run the arithmetic on what that implies. European storage capacity runs above 1,100 Bcf equivalent, so a shortfall against a normal early-August level of roughly 70% to 75% represents well over 100 Bcf equivalent of incremental purchases needed before winter — on top of normal replenishment.

Global buyers want U.S. cargoes and the raised full-year outlook from a major exporter confirms the demand is real. Hot weather across northeastern Asia is preserving that basin's price advantage for flexible cargoes, which means American exporters face two competing bids rather than one.

That is the setup that produces a fourth-quarter squeeze if maintenance clears. Feedgas at 20 Bcf per day, European restocking urgency, Asian summer demand and a U.S. surplus that has already stopped widening would compress storage into a winter with no cushion. Strong exports could quickly erase the surplus by fall.

The reason the market is not paying for it is sequencing. Between now and the moment that scenario becomes visible in nominations sit four weeks of shoulder-season injections, the September 1 pipeline addition, and cooler eastern forecasts. The bull case is a fourth-quarter trade being asked to support an August price.

There is also a counter-argument on the European side. A cooler European forecast has complicated the arbitrage for flexible cargoes, which is why July exports fell rather than rose despite the storage deficit. Buyers with weak prompt demand defer purchases even when they know winter risk is elevated.

For anyone forecasting past September, European storage at 58% is the single most important variable on the board. For anyone trading August, it is a footnote against a 33 Bcf injection.

Technical Map: $2.592, $2.676, $2.810 And $3.028

The chart is unambiguous and the levels are tight. The session began with the market in a downtrend on both the swing chart and the 50-day moving average, and the downtrend was reaffirmed when sellers took out Tuesday's low at $2.659.

Support below is thin. If selling pressure persists, long-term support at $2.592 comes into play — 1.9% beneath the $2.642 print. That level is the last identifiable structure before the market enters territory unvisited since the spring, and losing it on a closing basis opens the low $2.50s where model projections cluster.

The first meaningful sign of stabilisation would be recovering the former long-term bottom at $2.676. That reclaim would suggest the market may be nearing a value area — and the buying would likely be short-covering and modest bottom-picking rather than genuine accumulation. That distinction matters for anyone tempted to treat a bounce off $2.592 as a trend change.

The trend changes only if the swing top at $2.810 is taken out, 6.4% above spot. Beyond that, a second swing top at $2.979 has to fall before the market can even reach the 50-day moving average at $3.028 — 14.6% above the current price. Strong buying is required to take out that second top, and the sequencing means a genuine trend reversal requires three separate levels to break.

That structure is why every rally this summer has failed. Short-covering carried futures to $2.991 ahead of one report and died at the top of the range. The market has repeatedly approached the $2.979 swing top and never cleared it.

Model-driven projections point lower rather than higher: a five-day estimate at $2.51, a one-month figure at $2.46 and a three-month reading at $2.50. All three sit beneath the current price, and the one-month number implies a further 7.0% decline.

The practical framework into next week: sell rallies into $2.676 to $2.810 with stops above $2.979, or wait for $2.592 to break and trade the continuation. Buying $2.642 requires a weather revision that the current forecast does not support, and the 50-day average at $3.028 is a long way overhead.

The January Spike To $4.875 Proved What Can Happen

The reason nobody should size a short position casually here is what this market did seven months ago.

The February contract increased $1.76 in a single report week, moving from $3.120 per MMBtu to $4.875 — a gain of 56.4% in five sessions. The Henry Hub spot price rose $1.86 from $3.12 to $4.98 across the same window. A majority of U.S. hubs reported prices over $4.00 per MMBtu. The 12-month strip averaging February 2026 through January 2027 climbed 65 cents to $3.970.

Note the divergence in that last figure. Prompt futures rose 56.4% while the 12-month strip rose 19.6%. The increase was mostly a reaction to anticipated changes in 2026 storage balances and had no material influence on the longer-dated part of the curve. The market repriced the front and left the back alone, which is exactly how a weather-driven squeeze in a tight-storage environment behaves.

Houston-area temperatures averaged 53°F that week, down 6°F, producing 86 heating degree days — 29 more than the prior week and two above normal. Total Southeast consumption rose 26%, or 3.5 Bcf per day, led by a 54% increase in residential and commercial demand. Electric power consumption in the region rose 17%.

A two-degree deviation from normal in one region produced a 56% move in the front month. That is the volatility profile of this commodity, and it has not changed because the price is $2.642 instead of $3.120.

The difference between then and now is the storage cushion. January's spike occurred against inventories that markets feared would end the withdrawal season tight. Today's market carries a 6.4% surplus with 3,966 Bcf projected for end-October at 5% above normal. That buffer is what stands between a cold November and a repeat of $4.875.

The contract also tested a two-year high at $4.55 in May before losing 41.9% to today's level. Round trips of that magnitude inside nine months are normal here.

For position sizing, the lesson is directional conviction with defined risk. The trend is down and the fundamentals support it. The tail is fat and it is on the upside, because storage surpluses are consumed by weather far faster than they are built by production.

Official Forecasts At $3.70 Against A $2.642 Screen

The gap between institutional projections and the traded price is the widest it has been this year, and it frames the medium-term opportunity.

Official forecasting has the Henry Hub spot price averaging close to $3.70 per MMBtu across 2026 before declining below $3.50 next year. Record production helps meet rising demand and puts moderate downward pressure on prices, while inventories remaining above the five-year average limit upward pressure.

At $2.642, the September contract sits 28.6% below that $3.70 full-year average. Part of that gap is arithmetic — the January spike to $4.98 spot and the May high near $4.55 pull the annual average upward, so a weak August does not invalidate the projection. But it also means the second half has to average materially above $2.642 for the full-year number to hold, and the forward curve does not currently price that.

An earlier projection had the 2026 average just under $3.50 with 2027 rising sharply to just under $4.60, driven by demand growth outpacing supply growth on liquefaction feedgas and reducing storage. The current framework reverses that shape: $3.70 in 2026 falling below $3.50 in 2027.

That revision is significant and bearish. Moving 2027 from $4.60 to below $3.50 is a 24% reduction in the out-year forecast, and it reflects a supply picture — record Permian-led production and new pipeline capacity — that has proven more durable than the demand growth from exports.

Wholesale electricity prices are also forecast lower this summer than last, primarily on lower gas costs delivered to power plants, though heatwaves could still cause spikes. That confirms the demand-side read: cheap gas is being consumed for generation, and it is still not enough to draw storage.

Model-based projections sit well below the official view: $2.51 on a five-day horizon, $2.46 at one month and $2.50 at three months. Those numbers extrapolate the current trend and imply the price grinds into the low $2.50s.

The next official update lands August 11. It will carry the first full read on July feedgas, the 33 Bcf build and the September pipeline addition, and any downward revision to the $3.70 figure would remove the last institutional support for a higher price.

The Trade Into Next Week: $2.592 Or $2.810

The forecast resolves into a downtrend with defined levels and a fat upside tail. September gas at $2.642 sits 1.9% above long-term support at $2.592 and 6.4% below the swing top at $2.810.

The bear path is the path of least resistance and it is already in motion. Tuesday's low at $2.659 has been taken out. Losing $2.592 on a closing basis opens the low $2.50s, where five-day and three-month model projections cluster at $2.51 and $2.50. Below that, the one-month projection at $2.46 becomes the objective — a further 7.0% decline. The triggers are already in place: production near 111 Bcf per day, storage 6.4% above the five-year average, a 33 Bcf build that came in heavy, cooler eastern forecasts, and 1.5 Bcf per day of new Permian capacity arriving September 1.

The bull path requires three sequential breaks. Reclaim the former long-term bottom at $2.676 to signal a value area. Take out the swing top at $2.810 to change the trend. Clear the second swing top at $2.979 to reach the 50-day moving average at $3.028, 14.6% above spot. That path needs a hot revision across the Midwest and Northeast, feedgas nominations pushing above 20 Bcf per day as maintenance clears, and a storage build materially below the five-year average.

The base case is continuation lower into the shoulder season. Range $2.50 to $2.81 through August, with the surplus widening as cooling demand fades and pipeline capacity arrives. Next Thursday's storage report and the August 11 forecast update are the scheduled catalysts.

Position sizing must respect the January precedent. A two-degree temperature deviation moved the front month 56.4% in five sessions from $3.120 to $4.875 while the 12-month strip moved only 19.6%. This commodity does not correct — it gaps. Short positions here carry defined fundamental support and undefined weather risk.

The fourth-quarter setup is the opposite trade and it is worth tracking now. European storage at 58% heading into winter, well below normal. Feedgas recovering from maintenance with a raised full-year export outlook. Asian demand preserving a price premium for flexible cargoes. A projected end-October surplus of 5% that could compress fast if exports exceed 20 Bcf per day.

Base case into month-end: range $2.50 to $2.81, targeting $2.592 then $2.51, with invalidation on a daily close above $2.810. The heat is in the wrong place and the gas keeps coming.

That's TradingNEWS