Natural gas is the most weather-sensitive contract on NYMEX, and that sensitivity shows up as one of the most repeatable calendar patterns in commodities. A natural gas seasonality chart plots the average behaviour of the front-month Henry Hub contract (ticker NG) across the calendar year, letting you see when demand, storage flows and volatility historically cluster. It is not a trade signal on its own — but it is a map of where the risk usually lives.
This guide walks through the gas year month by month, explains the storage mechanics that create the pattern, and shows how to use a seasonality overlay without falling into the classic traps.
A seasonality chart aggregates many years of price data into a single average path — usually indexed to 100 at the start of the year, or expressed as average monthly percentage returns. Some versions use the continuous front-month series; others use a specific delivery month (for example, every January contract) to avoid distortion from roll yield.
That distinction matters more in natural gas than in almost any other market. The NG forward curve is not flat: winter contracts trade at a premium to shoulder-month contracts almost structurally, because storage costs money and winter demand is inelastic. A continuous front-month chart therefore blends real price moves with the mechanical effect of rolling from a cheap October contract into an expensive January one. Before you interpret any chart, check which construction method is used.
For a broader framework on how these patterns are built and where they break down, see our explainer on commodity seasonality.
US natural gas storage operates on a simple annual cycle:
Everything on a natural gas seasonality chart is a second-order effect of this cycle. The weekly EIA Natural Gas Storage Report — released Thursdays — is the scoreboard, and it is the single most reliable volatility event in the gas calendar. Traders compare the reported build or draw against consensus and against the five-year average for that week; the gap between them drives the immediate price reaction.
Two demand layers sit on top of this: residential and commercial heating in winter, and power-burn for air conditioning in summer. Gas is the marginal fuel in much of the US power stack, so a hot July can produce genuine summer strength — the reason the seasonal curve is not a simple one-hump shape.
January–February. Peak withdrawal. Demand is at its structural maximum and storage is draining fast, so any cold anomaly hits a market with limited slack. This is where the largest single-day percentage moves in NG historically occur. Directionally, though, the average result is ambiguous: winter risk premium is usually already priced in by late December, so a normal winter can leave prices drifting lower even as inventories fall.
March. The transition month. Withdrawal season formally ends March 31, and by mid-month the market is trading spring, not winter. Late cold snaps still matter, but their impact on end-of-season storage shrinks by the day. Volatility stays elevated while the curve reprices.
April–May. Injection begins. Heating demand collapses, cooling demand has not started, and power burn is at a seasonal low. These are the shoulder months — the period when the market is most driven by production and least by weather. Historically a soft stretch, though a low end-of-winter storage level can produce a spring rally as the market bids for refill.
June–August. Cooling season. Power burn becomes the swing demand factor and heat waves in Texas and the Southeast can flip weekly builds into draws. Summer strength is real but conditional — it depends on temperatures, and increasingly on renewables output and LNG export volumes.
September–October. The second shoulder. Injection season winds down and the market starts pricing end-of-October storage. If inventories are comfortable, gas tends to be heavy; if the refill has fallen short, October can see aggressive positioning into winter.
November–December. Winter risk premium is rebuilt. The first genuine cold event of the season usually produces an outsized reaction because positioning is light and the market has not yet tested demand. December often sets the tone for the whole withdrawal season.
The most famous seasonal trade in natural gas is the March/April calendar spread — long March, short April, or the reverse. March is the last withdrawal-season contract; April is the first injection-season contract. They sit on either side of the storage year boundary, so they can decouple violently.
In a tight winter, March can spike while April barely moves, blowing the spread out. In a mild winter with a comfortable storage overhang, the spread collapses toward zero or inverts. The nickname "widow-maker" comes from exactly this asymmetry: the spread has a floor near zero when winter is warm, but almost no ceiling when it is cold and inventories are thin.
Seasonality charts of the spread itself are more instructive than charts of flat price, because the spread isolates the storage-year effect from the general level of gas prices. If you trade it, size for the tail — not for the average.
A few discipline points:
To work with the current picture rather than a static average, pull up the natural gas price page on commodity-hub.eu and toggle the chart to a multi-year range. Overlay the same calendar window across several years to see whether the current path is tracking or diverging from the typical seasonal shape. Because gas rarely trades in isolation, it is also worth watching crude oil and refined products alongside it — energy complex correlations tighten during extreme weather events, when heating oil and gas compete at the margin.
For definitions of the terms used above — contango, backwardation, roll yield, front month — the glossary section is the place to start, alongside the commodity seasonality entry.
What months are historically strongest for natural gas? The late-autumn build into winter (roughly November–December) and periods of extreme summer heat tend to be the constructive windows, because that is when the market rebuilds winter risk premium or when power burn spikes. The shoulder months of April–May and September are historically the softest, since neither heating nor cooling demand is meaningful. These are tendencies, not rules — dispersion around the average is very wide in natural gas.
Why is the March/April natural gas spread called the widow-maker? March is the final contract of the withdrawal season and April is the first of the injection season, so they sit on opposite sides of the storage year. In a cold winter with low inventories, March can spike while April stays anchored, and the spread widens with no natural ceiling. In a mild winter the spread compresses toward zero. The payoff is highly asymmetric, which has repeatedly caused outsized losses for traders who sized the position off average outcomes.
Is a natural gas seasonality chart still reliable? The underlying storage cycle — inject in summer, withdraw in winter — has not changed, so the broad shape remains valid. What has changed is the amplitude and the demand mix, driven by shale supply, LNG exports and gas's role in power generation. Use seasonality as a base rate and context tool, then confirm with current storage data versus the five-year average and near-term weather forecasts.
What data should I watch alongside the seasonal pattern? The weekly EIA storage report, degree-day forecasts, dry gas production estimates, LNG feedgas flows, and the shape of the forward curve. Together these tell you whether the market is confirming the seasonal script or trading against it.