top of page

The Cold Shot Is Coming November 24–26 — and Whether It Stays Is a New York Story

Writer: Tony Zelinski
Tony Zelinski
20 minutes ago
6 min read
The Cold Shot Is Coming November 24–26 — and Whether It Stays Is a New York Story
The Cold Shot Is Coming November 24–26 — and Whether It Stays Is a New York Story

For once, the weather models agree. Both of the major global models — the American GFS and the European ECMWF — call for a national cold surge on November 24–26, the most reliable near-term demand signal on the board. Demand-weighted heating degree days peak on November 26—Thanksgiving Day—at roughly 23.7 on the GFS and 19.9 on the ECMWF, up from about 16 today. A four-model consensus from GasAlpha tells the same story: population-weighted heating degree days climb to 20–26 on November 25–26, the highest of the 10-day window. In Baltimore, the temperature drop is plain to see: highs near 71°F on Friday, November 20, falling to 50°F on Tuesday the 24th and 46°F on Wednesday the 25th.


That much is settled. What happens next is not — and the disagreement is almost entirely an East Coast story. For buyers in New Jersey and Massachusetts, that means the swing factor for the next two weeks of gas demand is sitting in their own backyard.


What the Models Agree On


The November 24–26 cold shot is the shared signal, and it is the one worth planning around. Both models produce it. The four-model GasAlpha consensus produces it. Private forecaster BAMWX is flagging cooler and stormier weather for the eastern half of the country in the last week of November, with heating degree days rising again into late November.

The cold arrives as millions of Americans travel for Thanksgiving — and as the December natural gas contract heads into its final week before expiry on November 25.


The market has already shown how it reacts to this kind of signal. On Tuesday, December futures fell to about $3.06 in morning trading, then reversed hard after a midday model update added cold and settled 2.8% higher at $3.208/MMBtu. That is what a market with a record speculative short — 227,227 contracts net short, per the latest CFTC data — does when a cold model run lands. Every forecast revision becomes a squeeze.


Where the Models Split


After November 26, the two models go their separate ways, and the gap is striking. Over the 15-day window, the ECMWF runs about 17.4 demand-weighted heating degree days colder than the GFS — 251.4 versus 234.0, or roughly 7.4% more heating load. But essentially all of that gap is one city: New York. The ECMWF is 57.7 heating degree days

colder than the GFS for New York City over the window, contributing +17.3 of the +17.4 national gap. Chicago adds +6.2 and Boston +3.2; Houston (−6.5) and Atlanta (−2.8) pull the other way.


The New York disagreement is a tale of two patterns. The GFS embeds two warm intrusions — heating degree days of just 8.8–10.0 on November 20–21 and 8.3–12.3 on November 28–29 — with a cold spike of 28.4 in between, on November 25. The ECMWF shows no warm relief in week two at all: New York heating degree days build steadily to 27.9 and 30.2 on

December 1–2, against 20.9 and 21.3 on the GFS — a disagreement of about 9 heating degree days per day at the end of the window. Boston mirrors New York: the GFS has cold front-loaded (30.7 on November 25) then moderating to 10.6 by November 29, while the ECMWF's cold intensifies late, from 24.7 to 31.0 by December 2.


Houston flips sign entirely. The GFS is cool there November 27–29, with temperatures in the high 40s to mid-50s; the ECMWF has Houston at 69–74°F with cooling degree days instead. Atlanta is colder on the GFS by 28.2 cumulative heating degree days. Chicago, by contrast, is the best-agreed hub in the country — the two models differ by just 2.55 heating degree days per day there.


In the middle of the window, the models nearly cancel out: days 7–12 are essentially tied nationally, about 118.0 versus 117.4 weighted heating degree days, because the ECMWF's Eastern cold is offset by the GFS's Houston and Atlanta cold. The divergence reappears in days 13–15, which tilt about 6.7 heating degree days colder on the ECMWF as it holds East Coast cold into early December.


What Not to Over-Read


Before anyone trades on a 17.4 heating-degree-day gap, the fine print matters. The models' convergence score is just 0.19 — low — with a 4.24 heating-degree-day average daily difference. Arc Research, which publishes the comparison, says plainly to treat either single-model cumulative number as low confidence.


More importantly, the early-window gap is not even news. The days 1–6 spread of 10.4 weighted heating degree days sits below the seasonal norm of 13.4 and is labeled "typical" — the models' usual bias, not a fresh demand signal. The ECMWF runs systematically colder than observed station weather from October through February. And the newest ECMWF cycle already warmed materially — down 23.1 weighted heating degree days from the prior run — even as the disagreement persists. These are single deterministic runs with no ensemble spread, and skill at days 8–15 is limited, precisely where the models diverge most.


So the honest read is two-part: the November 24–26 cold is bankable, and the December cold is a coin flip that the market will re-price model run by model run. That is the pattern Tuesday already showed — a 15-cent round trip in a single session on one forecast update.


What It Means for Buyers in New Jersey and Massachusetts


For commercial and industrial buyers, the takeaway is where to watch. The national average temperature is not the number that moves your price. The East Coast pattern is — and the East Coast is where you live.


In Massachusetts, the exposure is winter deliverability. January 2027 forwards at Algonquin Citygate still top $20/MMBtu, with January basis near $17.64 over Henry Hub. Last January, Winter Storm Fern drove Algonquin spot gas to $122.58/MMBtu. If the ECMWF's version of

December verifies — no warm relief, heating degree days near 30 in New York and Boston into December 2 — the Northeast will get its first real deliverability test of the season, and it will come with storage at 3,862 Bcf, just 2.0% above the five-year average, and record LNG

feedgas of 21.8 Bcf/d pulling gas out of the Gulf Coast at the same time.


In New Jersey, the same cold lands on a market already carrying capacity costs. PJM capacity prices went from $28.92 to $269.92 per megawatt-day, and the July 2026 auction cleared at the $325 cap for 2028/29. A cold Thanksgiving week lifts both gas and power demand at once — and the December contract expires November 25, the day before

the cold peaks, with a record short still in place. Expiry-week volatility is the norm in this setup, not the exception.


The practical point is timing. The EIA still expects Henry Hub to average just $2.87/MMBtu this winter, and the national storage picture is adequate. But the price your facility pays is not the national average — it is basis, and basis is priced on the East Coast pattern.

The buyers who do best in this market are the ones who hedge winter exposure before the model runs decide which version of December shows up, not after the first cold front arrives and the same forecast that was a coin flip on Monday is a squeeze on Tuesday.


PEM Perspective


Weather is the market's oldest story, and this week it is a particularly honest one: the models agree on the cold that is coming and disagree on the cold that follows, and they tell you exactly where the disagreement lives. It lives in New York. It lives in Boston. It

lives in the same hubs that set the price for every facility in New Jersey and Massachusetts.


That is not a reason to bet on a forecast. It is a reason to plan around the range. A cold shot on November 24–26 is now the base case, and a cold early December is a live possibility with a record short sitting underneath it. The gas is in storage, the export terminals are

running at record levels, and the market will decide the rest, one model run at a time. Secure your winter supply before it does.


Winter is almost here, and the market is already sending its first signals.




Winter is almost here...

Would you like a review of your facility's Winter Energy strategy? We are here to help!



 
 
 

Comments


Commenting on this post isn't available anymore. Contact the site owner for more info.
bottom of page