Drudge Retort: The Other Side of the News
Wednesday, October 07, 2026

Tropical Depression Nine has formed in the southwestern Gulf and is forecast to become a hurricane, potentially bringing significant flooding rain and strong winds to parts of the northern Gulf Coast and South late this week.

More

Alternate links: Google News | Twitter

Forecasters were monitoring a tropical depression in the Gulf of Mexico on Tuesday. If the system develops as they predict, it would become Tropical Storm Isaias and eventually the first hurricane of the 2026 Atlantic season.

[image or embed]

-- The New York Times (@nytimes.com) 8:40 PM · Oct 6, 2026

Comments

Admin's note: Participants in this discussion must follow the site's moderation policy. Profanity will be filtered. Abusive conduct is not allowed.

... That will make Isaias the first hurricane of an Atlantic season squashed by El Niño. ...

Looks like a Cat 2 when it makes landfall.



#1 | Posted by LampLighter at 2026-10-07 02:56 PM | Reply

Now officially a tropical storm.

www.nhc.noaa.gov

#2 | Posted by REDIAL at 2026-10-07 05:23 PM | Reply

Looks like a Cat 2 when it makes landfall.
#1 | Posted by LampLighter

This could be a chance for Google to score a win for their AI model. The official forecast is cat 2 but Google's model predicts Cat 3 which would put it into the "major storm" category.

#3 | Posted by johnny_hotsauce at 2026-10-07 05:33 PM | Reply

@#3 ... This could be a chance for Google to score a win for their AI model. ...

From what I've seen, the NWS is also looking at and testing AI weather models.

Fortunately, though, the NWS is a bit more conservative in their adoption of new models than Google might be.

If the Google AI model shows itself to be correct, that would be good. But it would be just one more data point in the evaluation of AI weather forecasting models.

#4 | Posted by LampLighter at 2026-10-07 06:50 PM | Reply

I thought King Dotard eliminated all hurricanes coming his gulf. Time to bust out the Sharpie.

#5 | Posted by TFDNihilist at 2026-10-07 07:24 PM | Reply

But it would be just one more data point in the evaluation of AI weather forecasting models.
#4 | Posted by LampLighter at 2026-10-07 06:50 PM | Reply | Flag:

They've been using AI models in weather forecasting for 40 years. In the 2010s it moved on to CNNs & RNNs. 15 years later it's on to GNNs and Vision Transformer models. The new models beat them across the board. Have you used python before? If so, what models have you trained?

#6 | Posted by sitzkrieg at 2026-10-07 08:28 PM | Reply

Guess that's a no. How train - test - validation splits work.

#7 | Posted by sitzkrieg at 2026-10-08 07:52 AM | Reply

And instead of "shows itself to be correct"

here's how accuracy, precision, and recall in machine learning is really measured. It's not binary.

#8 | Posted by sitzkrieg at 2026-10-08 07:59 AM | Reply

Cat 2 now.

#9 | Posted by REDIAL at 2026-10-08 09:37 PM | Reply

This could be a chance for Google to score a win for their AI model. The official forecast is cat 2 but Google's model predicts Cat 3 which would put it into the "major storm" category.

#3 | Posted by johnny_hotsauce at 2026-10-07 05:33 PM | Reply | Flag:

The Category 3 storm was packing 120 mph (193 kph) winds, and was expected to continue strengthening. Category 3 storms, considered major hurricanes, are capable of causing devastating damage, the center said.

#10 | Posted by sitzkrieg at 2026-10-09 09:07 AM | Reply

Running into predicted shear.

#11 | Posted by REDIAL at 2026-10-09 07:30 PM | Reply

@#11 ... Running into predicted shear. ...

Yup.

But still looks rather formidable.


#12 | Posted by LampLighter at 2026-10-09 07:34 PM | Reply

@#6 ... They've been using AI models in weather forecasting for 40 years. ...

Oh please, AI has only been around for a decade or so.

Yes, the forecasters have been using computers to try to model the weather, but those models have only had the intelligence that humans specifically programmed into them.

One of the things I noted about this current hurricane is that the ~cone of probability~ seems to be a lot narrower than it had been shown to be in the past. Is that due to the new prediction powers of NWS's recent adoption of AI in their forecast models?


#13 | Posted by LampLighter at 2026-10-09 07:41 PM | Reply

As of 4PM CT, the wind speed is 115mph, so a Cat 3.

Cat 3 is 111-129 mph

Saffir-Simpson Hurricane Wind Scale
www.nhc.noaa.gov

... 3 (major)
111-129 mph
96-112 kt
178-208 km/h

Devastating damage will occur: Well-built framed homes may incur major damage or removal of roof decking and gable ends. Many trees will be snapped or uprooted, blocking numerous roads. Electricity and water will be unavailable for several days to weeks after the storm passes. ...



#14 | Posted by LampLighter at 2026-10-09 07:46 PM | Reply

The Waffle House Index helps Southerners " and FEMA " gauge a storm's severity
apnews.com

... Golden hash browns, gravy-smothered biscuits and crispy waffles with a hearty helping of syrup are among the classic Southern comfort foods. But when hurricanes tear through Southeastern towns, the hot meals and bold yellow signs of the local Waffle House provide another kind of comfort.

If a Waffle House stays open in town, even in a limited capacity, neighbors are reassured that the coming storm is unlikely to cause devastation. A closed location of the dependable diner chain has come to indicate impending disaster. The metric is known as the Waffle House Index.

What might sound like silly logic has become one of the most reliable ways for Southerners " and even federal officials " to gauge a storm's severity and identify communities most in need of immediate aid.

What is the Waffle House Index?

The South's favorite disaster authority provides an informal measure of how significantly a storm will affect or has affected a community.

A map of the chain's over 1,900 locations, concentrated in the Southeast and mid-Atlantic, helps residents of storm-prone states assess whether they're likely to lose power, experience severe flooding or endure other extreme conditions that might cause a resilient restaurant to close its doors. For some, it's a telltale sign of whether they need to evacuate.

Waffle House is known not just for serving breakfast 24/7, 365 days a year, but also for its disaster preparedness. For decades, people across the South have noticed that the local Waffle House seemed to be the only business still open during a storm or the first to reopen after it passed.

The restaurant chain's reputation for remaining open when people desperately needed a place to warm up, charge devices and grab a hot meal became a fairly reliable " albeit amusing " resource to help track recovery efforts. The company even operates its own storm center.

How does the index work?

Waffle House's social media shares color-coded maps of its restaurant locations in certain regions that will soon be hit or are recovering from storm damage. The Federal Emergency Management Agency also offers some live tracking.

Green means the location is serving a full menu, indicating minimal damage in the surrounding area. The lights are on and the syrup is flowing.

Yellow means the restaurant is serving a limited menu, a signal that it's pulling power from a generator and might have a low food supply. The area might not have running water or electricity, but there's enough gas to fry up bacon for hungry customers.

Red means the location is closed, a sign of unsafe operating conditions and severe destruction to the restaurant or nearby communities. ...


#15 | Posted by LampLighter at 2026-10-09 07:51 PM | Reply

As a statistician once explained it to me, weather forecasting is considered virtually impossible because of the number of variables contributing to the future. Chaos in a more scientific sense.

So how much reliable input does AI have to go on here? And in climate, the future is not going to be acting like in the past, because there's no recorded precedent.

If that's a dumb question, I am out of my element. Apologies.

#16 | Posted by Dbt2 at 2026-10-09 07:52 PM | Reply

@#16 ... Chaos in a more scientific sense. ...

imo, yes, there is still chaos theory acting in weather forecasting.

But the accuracy has been improving over the years.

For example, the 5-day forecast of late is about as accurate as the 3-day forecast was 20 years ago.

Ditto for the current 7-day forecast vs the olden 5-day forecast.

Still, the accuracy is not 100%. So there is still a miss here and there.

I do look forward to any improved accuracy AI can bring to the weather forecasting table. But I also note that the current forecasting models have decades of experience behind them, so there may be a reluctance to jump to a new AI forecasting model without a similar track-record.



#17 | Posted by LampLighter at 2026-10-09 08:04 PM | Reply

Landfall.

#18 | Posted by REDIAL at 2026-10-10 12:25 AM | Reply

The following HTML tags are allowed in comments: a href, b, i, p, br, ul, ol, li and blockquote. Others will be stripped out. Participants in this discussion must follow the site's moderation policy. Profanity will be filtered. Abusive conduct is not allowed.

Anyone can join this site and make comments. To post this comment, you must sign it with your Drudge Retort username. If you can't remember your username or password, use the lost password form to request it.
Username:
Password:

Home | Breaking News | Comments | User Blogs | Stats | Back Page | RSS Feed | RSS Spec | DMCA Compliance | Privacy

Drudge Retort