How Natural Gas Storage Reports Move Prices (2026)

Natural gas storage reports move prices because the market trades the gap between what was reported and what was expected. A build larger than the consensus forecast means the market is holding more gas than buyers anticipated, and that extra supply pushes Henry Hub futures down. A build smaller than expected, or a bigger-than-expected withdrawal, tightens the balance and sends prices higher. The size of the surprise sets the size of the move, not the raw number on its own.

That is the whole mechanism in three sentences. Everything else that happens on a Thursday morning is noise, positioning, or weather sitting on top of it.

The EIA publishes the Weekly Natural Gas Storage Report every Thursday at 10:30 a.m. ET, drawn from a survey of roughly 90 storage operators. Traders across the curve treat it as the single most consequential scheduled release of the natural gas week, and for a few minutes the front-month Henry Hub contract can move several percent in either direction on the print alone.

Below is a walkthrough of what the report measures, how the actual-versus-consensus comparison works, why the five-year average is the yardstick the market actually uses, and how to read the number yourself without getting caught by the most common trap in this data.

What Does a Natural Gas Storage Report Measure?

The report measures working gas held in underground storage across the Lower 48 at the end of each week. Working gas is the volume actually available to be withdrawn and delivered, as opposed to the total pore space a facility holds. It is reported in billion cubic feet, abbreviated Bcf.

Underground storage comes in two forms. Salt-dome caverns, mostly in the Gulf Coast, can be cycled quickly and tend to hold natural gas pushed out by strong production. Depleted reservoirs and aquifers, concentrated in the Northeast, Midwest, and Rockies, hold larger volumes of working gas but inject and withdraw more slowly.

The data comes from the EIA’s Working Gas in Underground Storage survey, a weekly sample of storage companies across the country. Respondents report their own net volumes, and the EIA aggregates them into national and five regional totals. Because it is a survey rather than an audited census, individual regional figures carry some noise that the national total smooths out.

Three distinctions matter when you read the report. First, total US working gas is not the same as the gas sitting in any particular region, and regional balances drive local basis prices far more than the national number does. Second, working gas is not a reserve. Storage holds gas already produced and moved to the facility; it adds nothing to the resource base. Third, Henry Hub futures reference the broader Lower 48 market, not any single storage facility, so a facility-level issue rarely moves the benchmark on its own.

How Do Storage Reports Move Natural Gas Prices?

How natural gas storage reports move prices, step by step: the data drops, the market measures the deviation from consensus, traders update their forward balance, and the price adjusts until the marginal buyer and seller agree again.

The consensus number is a survey of analysts’ expectations compiled by the wire services just before the release. It is not a forecast of demand or supply. It is a forecast of what other people will print.

Consider the mechanics. Storage level feeds directly into the forward balance. If inventories sit far above the five-year band, the market expects ample supply, which pushes the curve into contango, where later months trade above the front month. That contango is itself a signal: carrying gas into a plentiful future is not worth much, so the prompt month looks cheap relative to the back of the curve. A large surprise against that backdrop can pull the prompt month down hard and steepen the spread.

The reverse holds in winter. When inventories sit near the five-year low or below it, the market prices the risk of a cold snap outrunning available supply, and the curve flattens or flips into backwardation. A small build against that backdrop removes the tightness premium quickly, and prices can fall even though the headline read is mildly positive.

The reaction is fast because positioning is crowded. Managed money and other speculative money hold large outright and spread positions ahead of the release, so a surprise arrives into a book that is already leaning one way. A bullish surprise landing on heavy speculative length can produce a violent squeeze that is not fundamentally driven, which is exactly the whipsaw pattern retail traders complain about in trading forums after the number.

What Data Does the Weekly EIA Storage Report Include?

The report carries more than the headline change, and the details often move futures as much as the headline does.

• Report date and week ending. The data covers the week ending Friday, published the following Thursday. This delay matters. The number you are reacting to describes demand and supply conditions that already ended.
• Weekly net change. The build or withdrawal in Bcf, expressed as both a current-week and prior-week figure so you can see the trend, not just the single print.
• Ending total working gas. The national stock level in Bcf, the figure that carries into the surplus or deficit calculation.
• Year-over-year comparison. Current total against the same week a year earlier, expressed in Bcf and as a percentage.
• Five-year average comparison. Current total against the five-year average for that week. This is the benchmark the market actually uses.
• Regional changes. Five regions: East, Midwest, South, Mountain, and Pacific, plus a Producing-region breakdown.
• Underground gas storage capacity. Total facility capacity and utilization, which frames how much room is left to inject.

Units matter for anyone comparing across data sources. Storage is reported in Bcf, a volume measure. Prices are quoted in dollars per MMBtu, an energy measure. Because the heating value of pipeline-quality gas varies slightly by location and by month, analysts convert volumes to energy using a standard heat content before drawing conclusions about days of supply.

How Are Bullish and Bearish Storage Surprises Calculated?

A bullish or bearish storage surprise is calculated by comparing the actual print to the consensus forecast, never by comparing the print to zero or to some absolute threshold. A 90 Bcf build is bearish if the market expected 70 Bcf and bullish if the market expected 110 Bcf.

Print	What it implies	Typical price reaction
Build larger than consensus	More supply or weaker demand than expected	Bearish, prompt month falls
Build smaller than consensus	Less supply or stronger demand than expected	Bullish, prompt month rises
Withdrawal larger than consensus	Heating or power demand running hot	Bullish, backwardation steepens
Withdrawal smaller than consensus	Mild weather or strong supply offset demand	Bearish, tightness premium deflates

Percentage surprises can look alarming early in a season and mean very little. Take a late-April print. If inventories are still near their seasonal low because winter demand pulled hard on storage, a small Bcf difference can translate into a double-digit percentage surprise. By August, when stocks are near capacity, the same absolute difference is a rounding error. Always check the percentage and the Bcf figure together before reacting.

A worked example makes the mechanic concrete. Imagine a late-spring print: the EIA comes in with a 92 Bcf build against a 96 Bcf consensus, with the prior week at 101 Bcf. The build was smaller than expected, which reads as supportive, because demand absorbed slightly more gas than the market had modelled. The magnitude was only 4 Bcf, small enough that record production caps and weather forecasts kept any rally from running far. That is a fair illustration of most weekly prints: directionally informative, rarely decisive on its own.

Why Do the Five-Year Average and Prior-Year Data Matter?

The five-year average matters because raw storage levels follow a strong seasonal curve. Working gas climbs from spring into early November and falls through the winter. Comparing April’s total to November’s total without normalizing produces nonsense, so analysts measure the current level against the average for the same week across the last five years.

The resulting surplus or deficit figure answers a specific question: is the market tighter or looser than a typical recent year at this point on the calendar? A 5 Bcf deficit against the five-year average in early April, when stocks are still low, is a much tighter market than the same 5 Bcf deficit in late September, when most surplus gas has already been injected. Identical number, opposite meaning.

Here is the honest caveat. The five-year window now includes years with meaningfully different demand structure, primarily because US liquefied natural gas export capacity took a large share of domestic supply off the pipeline. A benchmark built partly from pre-export years will read as tighter than the market really is. That is why some analysts prefer a ten-year linear trend as the normal, and why nobody argues the five-year average is perfect. It is a consistent yardstick, and consistency is what makes it tradable.

Year-over-year comparison is the second lens, and it serves a different purpose. The five-year average smooths weather and supply noise across half a decade. The prior year gives a single concrete reference point, useful when a specific season develops in a specific direction, such as a record-hot summer followed by a freeze-off in the producing region that cuts production sharply.

How Does Weather Change the Storage Signal?

Weather changes the storage signal because weather is the largest single input to the weekly balance. Heating degree days in winter and cooling degree days in summer drive residential and commercial demand, and the surplus or deficit in the print is largely a weather report in disguise.

Weather enters twice: once as realized conditions over the week being measured, and once as the forecast for the weeks ahead. The second pass is often more powerful. A hot forecast for the coming week gets priced into the front month immediately, because the market discounts expected withdrawals rather than waiting to see them.

Freeze-offs are the sharpest example. When temperatures collapse across the Rockies and the Dakotas, wells can be shut in within hours and production drops by several billion cubic feet a day. A single freeze-off can produce a 20 or 30 Bcf weekly drawdown, and traders read it simultaneously as a demand signal and a supply signal pulling in opposite directions. Communities trading the release describe this pattern repeatedly: futures rally on the cold, then stall because record production elsewhere absorbs the strength.

Power burn adds a second demand channel. Gas-fired electricity generation rises during hot weather and during periods when coal or nuclear output is disrupted. The grid increasingly pulls on storage in shoulder months, which has compressed the traditional shoulder season into a shorter, more active stretch.

How Do LNG Exports, Power Burn, and Supply Affect the Reaction?

Several variables decide whether an inventory surprise translates into a lasting price move or gets absorbed within the hour. LNG feedgas is now the largest of them. Export terminals withdraw roughly a quarter of US production at peak run rates, and their feedgas is taken near-flat around the clock regardless of weather. A new train starting up, an extended outage, or a shipping delay can add or remove several Bcf a day from the balance without any change in the headline print.

Dry gas production is the second. Production responds slowly and shows up in storage weeks later, which is why traders watch the monthly production estimates and rig counts rather than expecting the Thursday number to reveal a supply shift.

Pipeline constraints cause the regional effects that the national average hides. Take-away capacity limits mean gas can be stranded behind a bottleneck while inventories elsewhere sit comfortable. When that happens in the Northeast during a cold snap, citygate basis in New England and the Great Lakes can spike to multiples of the Henry Hub benchmark while the national storage surplus looks comfortable. This is the deliverability distinction that gets missed in most retail coverage: Henry Hub measures the benchmark balance, while basis measures whether the gas can physically reach the customer who wants it.

Planned maintenance rounds it out. Scheduled compressor or pipeline work reduces deliverability without changing supply, producing the same regional squeeze. And a single facility outage, such as freeze damage at a salt cavern in the Gulf Coast, can disrupt hundreds of millions of cubic feet per day at one site while barely registering in the national total.

How Traders Read the Report Before, During, and After Release

Before the release, professional desks size the surprise against positioning rather than against zero. The question is not how big the build is. It is how big the deviation needs to be to hurt the positions already on the book.

In the first minutes, the reaction is nearly mechanical. The prompt month gaps, spreads widen or narrow, and implied volatility jumps. Many traders step aside during this window because the spread between bid and offer is wide and the price discovery is unreliable.

Then they read the rest of the table. The order of operations matters: net change against consensus first, ending total against the five-year average second, year-over-year third, and regional splits last. The regional lines deserve real attention because they explain the basis action and frequently contradict the prompt-month move.

After the spike, confirmation comes from three places. The next-day weather forecast tells you whether the tightness will persist. LNG feedgas estimates tell you whether demand outside the domestic balance is growing. The deferred curve tells you whether the market believes the move: if the prompt month rallies but December and the following winter are unchanged, the market read the print as noise.

Volumes fade fast. Whatever the market’s real conviction is usually established within the trading day, and by the following session the print is background information. Chasing the initial spike is the most common retail mistake, and trading forums are full of accounts of positions opened in the first five minutes and closed at a loss by the close.

Why Can Natural Gas Prices Fall After a Bullish Storage Report?

Natural gas prices fall after an apparently bullish storage report when the result beat a consensus that had already priced in the bullish outcome. If the market expected a 70 Bcf build and the print was 90 Bcf, the report reads as bullish in level terms and bearish as a surprise.

Three other explanations account for most of the rest. A bullish print can land into a deteriorating technical setup, where prices break a key support level on the release and stop-loss orders accelerate the move. It can also be read as confirming weak demand, which traders sometimes take as a signal that prices have further to fall rather than bounce.

Finally, the forward curve can dominate. If the market is focused on rising LNG feedgas, a producer freeze-off, or an approaching cold season, a single weekly surplus is noise. Underhedged utilities and producers buying on dips will absorb the prompt-month weakness within the session.

Why Storage Matters More Than a Weekly Inventory Print

Storage level is the running scoreboard of the whole gas balance, which is why it carries information that any single weekly change cannot. A total number tells you how much cushion sits between current supply and a cold snap, and that cushion is the real constraint on winter prices.

The metric analysts use for this is forward demand cover: current working gas divided by the average daily consumption expected over the coming period. Expressed in days of supply, it strips out the seasonal shape of consumption and turns a volume into a buffer. A market with the same absolute inventory can have very different forward demand cover depending on how cold the winter forecast is and how much LNG is pulling feedgas.

Days of supply is also the reason a comfortable-looking national total can still produce regional price spikes. Cover is a national average, but constraints are local. Where pipeline capacity binds, the marginal molecule is not the national average molecule.

How Positioning Data Complements the Storage Print

Storage tells you what the balance did. Positioning data tells you who is exposed to it. Traders who want an edge on the Thursday reaction read both, because the same print produces very different outcomes depending on how crowded the trade already is.

The Commitments of Traders report, published each Friday, shows how non-commercial participants classified as managed money hold their futures and options positions, broken out by contract and by spread versus outright. A book that is already heavily long going into a bearish print has little room to add, which amplifies the move. A book carrying heavy short length into a bullish print sets up the squeezes retail traders describe.

Position limits set by the exchange add a second layer. When a contract approaches its speculative limit, traders start rolling into deferred months, and that mechanical flow shows up in the spread structure rather than the outright price.

Neither dataset forecasts the weather or the balance. Read together, they explain why the market reaction was the size it was, which is the question most people are actually asking after the number prints.

Frequently Asked Questions

What to Watch First

Start with the deviation from consensus, not the headline number. Then look at where the ending total sits against the five-year average for that week, since that tells you whether the market is looser or tighter than a typical recent season.

Third, check the regional splits and see whether they confirm or contradict the national picture. Fourth, confirm with the weather forecast for the coming week and the latest LNG feedgas estimate. Finally, check whether the reaction is consistent with the shape of the deferred curve.

A storage report is one input among several, published weekly with a six-day reporting lag. It describes what already happened rather than what will happen next, and a single print should never be treated as a forecast. Natural gas futures trading carries substantial risk, and rules, position limits, and reporting conventions change over time.

This article is general educational information about how a public data release interacts with commodity markets. It is not trading advice, and it is not individualized investment guidance.