NG ($2.84) Fights Record Supply as Cove Point Outage Sinks Appalachian Hubs — $2.75 Next

NG ($2.84) Fights Record Supply as Cove Point Outage Sinks Appalachian Hubs — $2.75 Next

Power burn fell below 40 Bcf/d and LNG feedgas softened with Cove Point offline for maintenance | That's TradingNEWS

Itai Smidt 9/22/2026 4:00:00 PM
Commodities NG1! NATGAS XANGUSD

Key Points

  • October NG traded at $2.84 after Monday erased the 3% gain from the week ended September 18.
  • Lower 48 output hit a record 113.2 Bcfd in September while demand rose only 2.7% year over year.
  • A storage build under 30 Bcf targets $2.912 and $3.00; above 35 Bcf opens $2.75 and $2.70.

October Henry Hub natural gas futures traded at $2.84 per MMBtu on Tuesday, September 22, up 0.26% from Monday's close. The contract opened at $2.830 and steadied through the morning as a drop in power burn below 40 Bcf/d and softer LNG exports were offset by a decline in production readings. That followed Monday's pullback, when the prompt month erased the entire gain it had built the prior week.

The recent path shows the tug-of-war. On Friday, September 18, the October contract settled at $2.912, up 1.1 cents or 0.4%, leaving it up about 3% for the week after a roughly 5% decline the week before. Monday wiped out that weekly gain. Over the past month, prices are up 0.30%. Over 12 months, they are down 9.44%.

The thesis for this forecast is direct. U.S. production is running at record levels and it is overwhelming every bullish input the market can find. Average Lower 48 output reached 113.2 Bcf/d so far in September, above August's record monthly average of 112.2 Bcf/d. Friday's dry gas production hit 113.1 Bcf/d, up 4.4% year over year, while Lower 48 demand was 80.9 Bcf/d, up 2.7%. Supply is growing at nearly double the rate of demand. That gap is why rallies keep failing near $2.90 and why the calendar 2027 strip has fallen to an average of $3.31 per MMBtu, its lowest since February 2022.

The bullish counterweight is real but narrower. The storage surplus is shrinking. In the week ended September 11, energy firms injected 44 Bcf, well below the 87 Bcf added a year earlier and the five-year average of 74 Bcf. That narrowed the surplus over the five-year average to 118 Bcf from 148 Bcf a week earlier. Late-season heat in the South and Gulf Coast is doing the work.

The seasonal setup adds tension. The October contract expires on September 28, which means positioning is rolling to November and the market is beginning to price winter risk. The reminder of what winter can do is recent: Henry Hub spot prices averaged $7.72 per MMBtu in January 2026 before collapsing to $3.62 in February.

The forecast bias is bearish near term with a defined range. Support sits at $2.75–$2.80 and resistance at $2.90–$2.912. A daily close above $3.00 would flip the bias.

Tuesday's Session: Power Burn Below 40 Bcf/d and Steady Prices

The session's structure shows a market balanced between two weakening forces.

Futures opened at $2.830 and steadied through the early hours. The bearish inputs were a dip in power burn below 40 Bcf/d and softer LNG export flows. Power burn, the amount of gas consumed by electric generators, is the single largest source of summer demand. When it drops below 40 Bcf/d, the market has lost its seasonal support and is transitioning to the shoulder period between cooling and heating seasons.

The offsetting input was a decline in production readings. Daily output estimates came in lower than the recent run rate, which supported prices enough to keep the contract positive at $2.84, up 0.26%.

That balance produced a narrow move. On Monday, futures pulled back amid healthy supply readings and forecasts for benign weather through the final stretch of September and into early October. That session erased the gains from the week ending September 18, when the contract had climbed about 3%.

The weather picture is the most immediate driver. Above-average temperatures are now expected to cover a smaller area of the South and Southeast from September 23 to October 2, which curbs gas demand from power generators for air conditioning. A sharp cooldown across the eastern United States is cutting late-season demand, with storms threatening Appalachia, while heat holds across Texas and the Gulf Coast.

The regional divergence is visible in cash markets. Six Appalachian hubs traded at their lowest prices since November 2024 after Cove Point LNG went offline on Saturday for annual maintenance, pulling roughly 0.85 Bcf/d of feedgas demand out of the region. When a major export outlet closes, the gas that would have moved to it competes for pipeline space, and regional prices collapse.

The technical read matches the fundamentals. Daily indicator-based signals are pointing to sell, downgraded from neutral on Monday evening.

For traders, Tuesday's session established the near-term range. The contract has failed at $2.912 and found support in the $2.80 area. The next directional catalyst is Thursday's storage report, which covers the week ended September 18.

Production at 113.2 Bcfd: The Structural Weight on Every Rally

Supply is the dominant variable in this market, and the numbers show why the bearish structure persists.

Average gas output in the U.S. Lower 48 states rose to 113.2 Bcf/d so far in September, up from a monthly record high of 112.2 Bcf/d in August. Friday's daily dry gas production was 113.1 Bcf/d, a 4.4% increase from a year earlier. For perspective, output averaged 107.4 Bcf/d in September 2025 and set what was then a record at 108.0 Bcf/d in August 2025. Production has added roughly 5 Bcf/d of supply in 12 months, an increase larger than the output of most producing countries.

The growth is concentrated in two basins. Rising production in the Permian and Haynesville regions has supported inventory builds through the summer, according to the EIA's September Short-Term Energy Outlook. Permian gas is largely associated production, meaning it comes out of the ground alongside crude oil. That makes it price-insensitive: producers drill for oil economics, and the gas arrives regardless of what Henry Hub pays. With oil at $89.42 for November WTI, well above pre-war levels, Permian drilling economics remain strong, which means associated gas keeps flowing.

Haynesville is different. It is a dry gas play in Louisiana, close to Gulf Coast LNG terminals, and its producers respond directly to gas prices. Haynesville output rising at $2.84 means breakeven costs have fallen enough for producers to expand even at these levels.

The demand side cannot keep pace. Lower 48 demand on Friday was 80.9 Bcf/d, up 2.7% year over year. Production growing 4.4% against demand growing 2.7% produces surplus gas, which goes into storage.

That surplus explains the price structure. Record output and mild spring weather allowed inventories to stay above the five-year average since March, reaching a peak of 7.7% above normal in April.

The medium-term demand story is stronger. The EIA forecasts U.S. electricity sales of 4,135 billion kilowatt-hours in 2026 and 4,211 BkWh in 2027, driven by data center development and increased manufacturing. Gas-fired generation supplies much of that load growth. But that demand arrives over years, while the supply is arriving now.

For the forecast, production is the reason the bias is bearish. Until the Lower 48 run rate falls below 112 Bcf/d, rallies will keep failing.

Storage: A 44 Bcf Build and a Surplus Narrowing to 118 Bcf

The storage data is the bullish half of the argument, and it is the metric that has kept prices from breaking down.

In the week ended September 11, energy firms injected 44 Bcf into storage. That was well below the 87 Bcf added in the same week a year earlier and below the five-year average build of 74 Bcf. Total inventories reached 3.298 trillion cubic feet, about 3.6% below last year's level but 3.7% above the five-year average. The below-average build narrowed the surplus over the five-year average to 118 Bcf from 148 Bcf a week earlier.

The driver was late-season heat. High temperatures and strong air-conditioning demand limited the amount of gas added to storage. Power generators burned gas that would otherwise have gone into the ground.

The trend matters more than the single print. The surplus peaked at 7.7% above normal in April and has narrowed to 3.7%. Over one week, it shrank by 30 Bcf. If that pace continued for four more weeks, the surplus would be close to erased before the injection season ends on October 31.

The EIA's own forecast disagrees. It projects natural gas inventories will total 3,969 Bcf on October 31, 2026, the end of the injection season, 5% above the five-year average. That forecast was completed on September 3, before the recent run of below-average builds. From the September 11 level of 3,298 Bcf, reaching 3,969 Bcf would require 671 Bcf of injections over seven weeks, an average of 96 Bcf per week. Recent builds have been running at 44 Bcf. That gap suggests the EIA's October 31 estimate may be too high, which would be bullish.

Thursday's report, covering the week ended September 18, is the next test. A build under 30 Bcf would confirm that demand is cutting into the surplus and would strengthen the bullish case. A build above 35 Bcf would show the tightening is fading and would support the bearish case.

The storage picture ahead of winter is the medium-term driver. Inventories above the five-year average at the start of the heating season give the market a buffer against cold snaps. Inventories at or below average leave it exposed. At 3.7% above average and narrowing, the cushion is thinner than it looked in April.

LNG Exports: 18.2 Bcfd Friday, Cove Point Offline and Cameron Recovering

Export demand is the swing factor between record production and a bearish price, and it is currently unstable.

LNG feedgas, the gas piped to export terminals for liquefaction, was on track to rise to 18.2 Bcf/d on Friday, up from a three-week low of 17.1 Bcf/d on Thursday. The increase came primarily from higher flows to Sempra's 2.0 Bcf/d Cameron LNG plant in Louisiana as it recovered from maintenance. Earlier in the week, average flows to the nine major U.S. LNG export facilities were expected to fall to a three-week low of 17.5 Bcf/d, primarily because of that Cameron maintenance.

The recovery did not last. Berkshire Hathaway Energy took its 0.8 Bcf/d Cove Point LNG export plant in Maryland offline on Saturday for a few weeks of annual maintenance. That pulled roughly 0.85 Bcf/d of feedgas demand out of Appalachia. Six regional gas hubs in the area traded at their lowest prices since November 2024 as a result.

Softer LNG exports were among the bearish inputs on Tuesday. That is the pattern to watch: at 18.2 Bcf/d, LNG absorbs roughly 16% of Lower 48 production. Every 1 Bcf/d of feedgas lost to maintenance is 1 Bcf/d of extra supply that has to find a home in storage or be priced into the market.

The structural trend is upward. LNG feedgas averaged 15.7 Bcf/d in September 2025 and 16.1 Bcf/d in early October 2025. At 18.2 Bcf/d, current flows are more than 2 Bcf/d higher year over year. New liquefaction capacity coming online continues to add baseload demand, which is the main reason Henry Hub has held near $2.84 rather than falling further against record production.

The global picture adds a variable. Global natural gas prices fell on Monday ahead of a series of meetings between world leaders this week, on renewed hopes for Middle East diplomacy, even though the conflict has widened. Lower international prices narrow the arbitrage that makes U.S. LNG exports profitable. If European and Asian prices fall far enough, cargo economics weaken and feedgas demand could soften.

For the forecast, LNG is the key bullish lever. A sustained move in feedgas above 19 Bcf/d would tighten the balance meaningfully. A drop below 17 Bcf/d, with Cove Point offline and other plants in maintenance, would add pressure toward $2.75.

The 2027 Strip at $3.31: The Market's Verdict on Long-Term Supply

The forward curve carries a message that the front month does not, and it is the most bearish data point in this market.

Futures for calendar 2027 have fallen to an average of $3.31 per MMBtu, their lowest level since February 2022. That is the price at which producers can lock in revenue for every month of next year. It sits 16.5% above the current October contract at $2.84, which reflects normal seasonality, since winter months carry higher prices than shoulder months.

The significance is in the decline. A calendar strip at a four-year low means the market does not believe that rising LNG exports and data center power demand will tighten the balance enough to lift prices next year. Traders are pricing continued production growth from the Permian and Haynesville that offsets demand growth.

That has consequences for producers. At a $3.31 average for 2027, drilling economics in dry gas basins are thin. Gas producers hedge forward production using the strip, so a falling strip reduces the revenue they can lock in and eventually slows drilling. The mechanism is self-correcting but slow: low forward prices reduce future supply, which eventually lifts prices.

The equity market is pricing that pressure. Coterra Energy (CTRA), one of the largest U.S. gas and oil producers, fell 8.62% to $32.56 on Tuesday, among the largest declines in the S&P 500. Energy was the weakest S&P sector, down 1.3% after the open, though much of that reflected the 3.19% drop in crude oil to $89.42 on Iran's offer to reopen the Strait of Hormuz within seven days.

The oil-to-gas ratio highlights how differently the two markets are trading. At $89.42 for WTI and $2.84 for gas, the ratio stands at 31.5 to 1. Crude carries a large war premium from a seven-month Middle East conflict. U.S. natural gas carries none, because it is a landlocked market with its own supply picture and only limited exposure to global prices through LNG.

The EIA's forecast for the region is relevant to both. It expects Middle East oil production to rise as flows through Hormuz gradually increase, while some export constraints persist through the end of the year.

For the forecast, the 2027 strip at $3.31 confirms that the structural bearish case remains intact. The front month's direction is about weather and maintenance. The strip is about supply, and supply is winning.

Seasonality: Shoulder Season Now, and the Memory of $7.72 in January

The calendar is the most important context for anyone trading natural gas in late September.

The market is in the shoulder season, the period between summer cooling demand and winter heating demand. Injection season ends October 31. Power burn has dropped below 40 Bcf/d, and forecasts call for benign weather through the final stretch of September and into early October. This is the annual window when demand is at its weakest and the market has the least support.

The October contract expires on September 28, six days from now. That expiry forces positioning to roll to November, which is the first month of the heating season and carries winter risk premium. Roll periods often produce volatility, as traders close October positions and establish November ones.

The winter reference is dramatic. Henry Hub spot prices averaged $7.72 per MMBtu in January 2026, then collapsed to $3.62 in February. Before that, the monthly averages were $3.19 in October 2025, $3.79 in November 2025 and $4.26 in December 2025. That January spike shows what a cold winter does to a market with inventories near average. From a $2.84 October contract, a repeat of January 2026 would mean a 172% move.

That asymmetry defines the trade. At $2.84, downside is limited by production economics and by the fact that the front month is already near multi-month lows. Upside, if winter delivers, is unlimited in practical terms because the market has no fast way to add supply in January.

Storage is the buffer that determines how violent a winter move can be. At 3,298 Bcf as of September 11, 3.7% above the five-year average and 3.6% below last year, inventories are adequate but not abundant. The EIA projects 3,969 Bcf on October 31, 5% above the five-year average. If actual builds continue at the recent 44 Bcf pace rather than the 96 Bcf weekly rate required to hit that forecast, the market would enter winter with a thinner cushion than expected.

The current cooldown is the near-term factor. A sharp cooldown across the eastern United States is cutting late-season demand, while heat holds across Texas and the Gulf Coast.

For the forecast, seasonality argues for a bearish front month and a constructive view of November and December contracts. The October expiry on September 28 is the pivot.

Key Support: $2.80, $2.772 and the $2.70 Floor

The downside map for October natural gas is defined by recent trading and by production economics.

The first support is $2.80. It is a round number and the zone where Tuesday's session found buyers after opening at $2.830. Holding above $2.80 keeps the contract in its recent range.

The second is $2.772, a level the front month traded at during the recent pullback. A break below it would take the contract to fresh lows for the month and would signal that the bearish weather and production picture is overwhelming the storage tightening.

The third is $2.75, followed by the $2.70 round number. Reaching $2.70 would be a 4.9% decline from $2.84. That scenario would require Thursday's storage build to come in above 35 Bcf, continued benign weather into October and LNG feedgas staying below 18 Bcf/d with Cove Point offline.

Below $2.70, the market would be testing levels that pressure Haynesville drilling economics. Dry gas producers respond to sustained sub-$2.75 prices by curtailing completions, which eventually slows production growth. That is the self-correcting mechanism that usually limits how far the front month falls during shoulder season.

The technical backdrop supports caution. Daily indicator-based signals moved to sell on Monday evening after being neutral earlier. The prompt month erased the entire gain from the week ending September 18 in a single Monday session, which shows that sellers are more aggressive than buyers at current levels.

The regional cash markets are already weaker than the futures. Six Appalachian hubs traded at their lowest prices since November 2024 after Cove Point went offline. When physical markets price below the futures benchmark, it signals surplus supply looking for outlets.

The bullish counterweight at lower levels is the storage tightening. Injections running at 44 Bcf against a 74 Bcf five-year average mean the surplus is shrinking every week. Traders who are short the front month into a falling surplus and an approaching heating season face increasing risk as October progresses.

The support rule: above $2.80, the contract is range-bound. Between $2.70 and $2.80, the bearish case is playing out. Below $2.70, production economics begin to matter and the risk-reward for new shorts deteriorates sharply.

Key Resistance: $2.90, the $2.912 Settle and $3.00

The upside map is tight, and every level has been tested in the past week.

The first resistance is $2.87–$2.90. The contract traded at $2.87 and $2.90 during the past week as storage data and weather forecasts shifted. It has repeatedly failed to hold above $2.90 on a closing basis.

The second is $2.912, Friday's settlement. That is the high-water mark of the recent rally and the level Monday's pullback erased. Reclaiming $2.912 on a daily close would mean the market has recovered the entire Monday decline and would signal that the storage tightening is winning over the production weight. From $2.84, that is a 2.5% gain.

The third is $3.00, the psychological round number. A daily close above $3.00 would break the bearish structure and would require a genuine catalyst: a storage build under 30 Bcf on Thursday, heat holding across the South and LNG feedgas staying above 19 Bcf/d. That combination would show demand is cutting into the surplus faster than the market expects.

Above $3.00, the next references come from the forward curve. The calendar 2027 strip averages $3.31, and winter months in the curve carry premiums above the October contract. The November contract, which becomes the prompt month after September 28, will trade at a premium to October because it carries heating season risk.

The historical winter levels sit far above. Henry Hub spot averaged $3.79 in November 2025, $4.26 in December 2025 and $7.72 in January 2026. Those are references for what a cold winter can produce, not near-term targets.

The conditions required for an upside break are specific and currently absent. Weather forecasts call for benign conditions through early October, with above-average temperatures covering a smaller area of the South and Southeast from September 23 to October 2. Production is at a record 113.2 Bcf/d. LNG feedgas is under pressure with Cove Point offline for a few weeks.

The one variable that could change quickly is weather. October cold snaps in the Midwest and Northeast can arrive with two weeks' notice and add several Bcf/d of heating demand. That is the main path to a move above $3.00 before the November contract takes over.

The resistance rule: below $2.912, the bias stays bearish. Above $3.00 on a daily close, the structure flips and the bearish view is invalidated.

Thursday's Storage Report: The Week's Decisive Number

The weekly EIA storage report is the single most important scheduled event for natural gas traders, and Thursday's release carries unusual weight.

The report will cover the week ended September 18. The prior week's figure was a 44 Bcf injection, against 87 Bcf a year earlier and a five-year average of 74 Bcf. That print narrowed the surplus over the five-year average to 118 Bcf from 148 Bcf.

The thresholds are clear. A build under 30 Bcf would strengthen the bullish case substantially. It would mean the surplus is narrowing faster than expected and that demand, including power burn and LNG, is absorbing more of the record production than the market assumed. That would support a push toward $2.912 and potentially $3.00.

A build above 35 Bcf would strengthen the bearish case. It would suggest the tightening seen in the September 11 week was a function of late-season heat that is now fading, and that the underlying balance remains loose. That would open the path toward $2.75 and $2.70.

A build between 30 and 35 Bcf would leave the market range-bound between $2.80 and $2.912.

The week covered included specific conditions worth noting. A sharp cooldown hit the eastern United States, cutting late-season demand, while heat held across Texas and the Gulf Coast. Cove Point LNG was still operating for most of that week, going offline on Saturday, September 19, which falls after the reporting period. That means the Cove Point outage will show up in the following week's report, not this one.

Production during the reporting week was running near 113 Bcf/d. LNG feedgas moved from a three-week low of 17.1 Bcf/d on Thursday, September 17, to 18.2 Bcf/d on Friday, September 18.

Beyond Thursday, the next major data point is the EIA's October Short-Term Energy Outlook, due October 6. It will update the October 31 storage forecast of 3,969 Bcf. A downward revision would confirm that the tightening is real and would be bullish for winter contracts.

The Henry Hub spot price series from the EIA updates on September 23, providing another reference point for physical market conditions.

For the forecast, Thursday is the pivot. The 30 Bcf and 35 Bcf thresholds define the immediate direction.

Risk Scenarios: What Sends October Gas to $3.00 or $2.70

Several specific risks could break the current range in either direction.

The upside risks come from weather and supply disruptions. An early cold snap across the Midwest and Northeast in the first half of October would add heating demand weeks before the market expects it, and traders positioned short into the November roll would have to cover. A hurricane or tropical system in the Gulf of Mexico would shut in production and potentially disrupt LNG loadings, which can move prices several percent in a session. An unplanned outage at a major production basin, or freeze-offs later in the season, would tighten supply quickly.

Demand upside could also come from LNG. If feedgas recovers above 19 Bcf/d once Cameron is fully back and other plants complete maintenance, the balance tightens by more than 1 Bcf/d against current levels. Rising global prices would improve cargo economics and pull more gas to the coast.

The downside risks are supply and weather. Production continuing above 113 Bcf/d through October would keep the balance loose. The forecast for benign weather through early October, with above-average temperatures covering a shrinking area of the South and Southeast, would cut power burn further. Extended LNG maintenance, with Cove Point offline for a few weeks and any additional unplanned outages, would strand supply.

Global factors add pressure. Global natural gas prices fell on Monday on renewed hopes for Middle East diplomacy, even as the conflict widened. Iran offered to reopen the Strait of Hormuz within seven days, which sent Brent to $97.64 and WTI to $89.42. Lower global energy prices reduce the pull on U.S. LNG exports.

The macro backdrop is mixed. Equities are strong, with the Nasdaq near records, but the Federal Reserve raised rates to 3.75%–4.00% on September 16 and markets price a 90% chance of another hike in December. A broadly softening U.S. economy and slower global activity could take some heat out of industrial gas demand.

The contract-specific risk is expiry. The October contract expires September 28. Expiry weeks can produce sharp moves as positions unwind, independent of fundamentals.

The scenario weighting favors the bearish case near term. Production is at a record, weather is benign and LNG is impaired. The bullish case depends on Thursday's storage print and on weather that has not yet appeared in forecasts.

Price Targets: $2.70 Near Term, $2.912 on a Tight Print, $3.00 as the Line

The forecast breaks into three scenarios with specific triggers.

The base case over the next one to two weeks is a range between $2.75 and $2.912, with a downward drift toward $2.75–$2.80 as shoulder season deepens. This assumes production holds above 113 Bcf/d, weather stays benign through October 2, LNG feedgas stays near 18 Bcf/d with Cove Point offline, and Thursday's build lands between 30 and 35 Bcf. In this scenario, October expires on September 28 near the low end of the range and positioning rolls to November.

The bear case targets $2.70 and then $2.65. It requires Thursday's build above 35 Bcf, continued mild weather into mid-October and LNG feedgas falling below 17.5 Bcf/d. From $2.84, a move to $2.70 is a 4.9% decline. Below $2.70, dry gas producer economics come under pressure, which limits further downside.

The bull case targets $2.912 first, then $3.00. It requires a build under 30 Bcf on Thursday, heat holding across the South and LNG feedgas rising above 19 Bcf/d. Reaching $3.00 from $2.84 is a 5.6% gain. A daily close above $3.00 would break the bearish structure and open the path toward the winter premium embedded in the November and December contracts.

The medium-term view is more constructive. The storage surplus has narrowed from 7.7% above normal in April to 3.7% in September, and weekly builds at 44 Bcf are far below the 96 Bcf weekly pace required to reach the EIA's 3,969 Bcf October 31 forecast. If injections keep running light, the market enters winter with a thinner cushion than expected. The January 2026 spot average of $7.72 shows what a cold winter can do.

The risk-reward at $2.84 favors patience. The downside to $2.70 is $0.14, and the upside to $3.00 is $0.16. At a contract value of $10,000 per point, that is $1,400 of risk against $1,600 of reward. For short positions, selling rallies toward $2.90–$2.912 with a stop above $3.00 offers $0.20 of reward to $2.70 against $0.09 of risk, a ratio above 2 to 1.

Traders looking for winter exposure should focus on November and later contracts rather than the expiring October contract.

Verdict: Bearish Front Month, Constructive Into Winter, $3.00 Is the Line

The verdict on October natural gas at $2.84 is bearish in the near term, with a constructive view of the winter contracts.

The bearish case is structural. Lower 48 production is at a record 113.2 Bcf/d in September, above August's record 112.2 Bcf/d, and Friday's dry gas output of 113.1 Bcf/d was up 4.4% year over year against demand growth of 2.7%. Power burn has dropped below 40 Bcf/d as the shoulder season arrives. Weather forecasts call for benign conditions through early October, with above-average temperatures covering a shrinking area of the South and Southeast. Cove Point LNG is offline for annual maintenance, pulling 0.85 Bcf/d of demand out of Appalachia, where six regional hubs traded at their lowest prices since November 2024. The calendar 2027 strip has fallen to $3.31, its lowest since February 2022. Daily technical signals turned to sell on Monday evening, and the prompt month erased a week of gains in a single session.

The bullish case is tightening storage. The September 11 build of 44 Bcf was well below the 87 Bcf added a year earlier and the 74 Bcf five-year average. The surplus narrowed to 118 Bcf from 148 Bcf, and inventories at 3,298 Bcf are 3.7% above the five-year average, down from a 7.7% peak in April. Reaching the EIA's 3,969 Bcf October 31 forecast would require 96 Bcf weekly builds, more than double the recent pace. LNG feedgas recovered to 18.2 Bcf/d on Friday, and structural export capacity keeps growing.

The trading plan follows. Sell rallies toward $2.90–$2.912 with a stop above $3.00, targeting $2.75 and then $2.70. Treat a Thursday storage build above 35 Bcf as confirmation. Treat a build under 30 Bcf as a reason to stand aside, with $2.912 and $3.00 in play. A daily close above $3.00 invalidates the bearish view entirely.

For position traders, the better expression of a bullish winter view is the November contract or later, not the October contract that expires September 28. January 2026's $7.72 spot average is the reminder of what a cold winter delivers when inventories are only modestly above average.

Verdict: bearish front month. Near-term target $2.75, extended target $2.70. Invalidation on a daily close above $3.00.

That's TradingNEWS