If you can read this, either the style sheet didn't load or you have an older browser that doesn't support style sheets. Try clearing your browser cache and refreshing the page.

(NBC Chicago)   Coming soon - all weather forecasts in the Netherlands to be changed to 50% probability   (nbcchicago.com) divider line 26
    More: Asinine, Coming Soon, Netherlands, Dutch, weather predictions, raw data, forecasts  
•       •       •

2225 clicks; posted to Geek » on 18 Jul 2012 at 12:12 PM (2 years ago)   |  Favorite    |   share:  Share on Twitter share via Email Share on Facebook   more»



26 Comments   (+0 »)
   
View Voting Results: Smartest and Funniest

Archived thread
 
2012-07-18 08:46:07 AM
*rtfa*

I'm kind of OK with this. I mean, being off by as much as 5 or so degrees, or the weather is sprinkly instead of sunny, is not a big deal. But when they forecast 50 F temperatures and sun, but you drive into A farkING SNOWSTORM, they should be driven out on a rail.

I'M LOOKING AT YOU, ALL THE WEATHER FORECASTERS IN NEW ENGLAND!!1!
 
2012-07-18 11:03:47 AM
They wouldn't be having these problems is people would just follow my suggestion and kill all the damn butterflies.
 
2012-07-18 11:48:52 AM

xanadian: *rtfa*

I'm kind of OK with this. I mean, being off by as much as 5 or so degrees, or the weather is sprinkly instead of sunny, is not a big deal. But when they forecast 50 F temperatures and sun, but you drive into A farkING SNOWSTORM, they should be driven out on a rail.

I'M LOOKING AT YOU, ALL THE WEATHER FORECASTERS IN NEW ENGLAND!!1!


justamiser.files.wordpress.com

I make the weather! All of this moisture coming up out of the Gulf is gonna push off to the east and hit Altoona.
 
2012-07-18 12:14:39 PM
I love how politicians think. They can just make a law and solve a problem.

Make fines for predicting bad weather that turns out being nice? Always predict nice weather
 
2012-07-18 12:17:24 PM
"Last week it was really good weather over most of the country but the weather forecasts were full of heavy rain and thunderstorms, so people stayed home," Joep Thonissen, head of the Dutch tourist attraction association told The Telegraph.

So... What you want to do is make it illegal to forecast bad weather. Why don't you include dire economic predictions while you're at it?
 
2012-07-18 12:19:50 PM
 
2012-07-18 12:23:12 PM
In today's weather forecast, we have chances of sun, rain, sleet, snow, thunder, hail, fog, and mist. It may be windy or calm, and the temperature could have a possible low of -40 degrees and a possible high of 40 degrees. Much the same the next few days.

Now, back to sports...
 
2012-07-18 12:30:54 PM
Maybe purchase a storm tracker 3000, or a climate machine 10000, or maybe an ultra super twin Doppler storm slayer 850.
 
2012-07-18 12:48:38 PM

Ponzholio: How Valid Are T.V. Weather Forecasts?

Answer- Not very.


Wow, that was a pretty crappy study. For starters, the precipitation data is worthless since they only looked at whether or not it rained at the airport. In a metropolitan area of roughly 8,000 sq mi, they focused on exactly 16 sq mi to determine whether or not it rained in the city. I very well could have rained in Independence, Overland Park, or Lenexa without raining at the airport, especially considering that the airport is in the absolute farthest northwest corner of the city.

With regard to the temperature data, I think a deviation of 3-4 degrees is pretty good. If I tell you it's getting to hit 106 today and it only hits 102, I've still given you a pretty good idea of what to expect.

I think the biggest problem is that people don't understand that even the best meteorologists lose confidence more than about three days out.
 
2012-07-18 12:59:39 PM
People should get penalized for inaccurate reports.

Predictions, on the other hand....
 
2012-07-18 01:00:44 PM
See proper scoring rules and Jaynes' honest weatherman problem (p. 5 of the PDF). The idea is to design a forecaster's reward system so that they receive the best expected reward for the best-calibrated forecasts (in the sense that their predicted probabilities match the actual frequencies). Unfortunately Jaynes' relative entropy rule involves unbounded rewards and so isn't completely practical.
 
2012-07-18 01:05:20 PM

xanadian: *rtfa*

I'm kind of OK with this. I mean, being off by as much as 5 or so degrees, or the weather is sprinkly instead of sunny, is not a big deal. But when they forecast 50 F temperatures and sun, but you drive into A farkING SNOWSTORM, they should be driven out on a rail.

I'M LOOKING AT YOU, ALL THE WEATHER FORECASTERS IN NEW ENGLAND!!1!


This is why we can't have nice things. Apparently there is nothing people think can't be improved through legislation. Every annoyance, inconvenience and perceived slight must be stamped out and the offenders must be made to suffer. Let me guess... next on your list is misdemeanor charges against those who leave up those "back in X minutes" signs and aren't back by then, followed by those who say they'll meet you at 7:00 and don't show up until 7:40.
 
2012-07-18 01:17:05 PM
Stephanie Abrams can tell me any farking thing she wants, about any subject.
I'll just nod and smile.
 
2012-07-18 01:39:32 PM
If a weather forecaster says there is a 95% chance of rain, and it doesn't rain, THIS WAS AN ACCURATE REPORT. Rolling a natural 20 in a D&D skill check isn't exactly a rare occurance.
 
2012-07-18 01:41:21 PM

rugman11: I think the biggest problem is that people don't understand that even the best meteorologists lose confidence more than about three days out.


Every meteorologist will tell you they wish they could only present three days out, because that's all the science can predict with enough accuracy to air/publish. But people insist on 7 day forecasts, and TV stations and other media have caved in.
 
2012-07-18 01:50:05 PM

jack21221: If a weather forecaster says there is a 95% chance of rain, and it doesn't rain, THIS WAS AN ACCURATE REPORT. Rolling a natural 20 in a D&D skill check isn't exactly a rare occurance.


Not quite. Its a 30% (for example) chance of rain because they are saying rain will cover 30% of the area. So if you're going to be having a BBQ in your backyard, there's a 30% chance it will hit you.

But the 30% is an actual prediction of an actual event: 30% of the area will be seeing rain. That can be verified as true or untrue if it turns out 60% of the area was covered by rain.
 
2012-07-18 01:53:00 PM

jack21221: If a weather forecaster says there is a 95% chance of rain, and it doesn't rain, THIS WAS AN ACCURATE REPORT. Rolling a natural 20 in a D&D skill check isn't exactly a rare occurance.


Or think of it this way: when they say 30% chance of rain... they're saying two absolute things:

At any point during the day (or hour if its an hourly forecast):

A. There is an absolute 100% chance that 30% of the area will see rain
B. There is an absolute 0% chance that 70% of the are will not be seeing rain
 
2012-07-18 02:15:31 PM
And now, the weather. How's it look Dr. Schrödinger?
 
2012-07-18 02:51:32 PM

downstairs: jack21221: If a weather forecaster says there is a 95% chance of rain, and it doesn't rain, THIS WAS AN ACCURATE REPORT. Rolling a natural 20 in a D&D skill check isn't exactly a rare occurance.

Not quite. Its a 30% (for example) chance of rain because they are saying rain will cover 30% of the area. So if you're going to be having a BBQ in your backyard, there's a 30% chance it will hit you.

But the 30% is an actual prediction of an actual event: 30% of the area will be seeing rain. That can be verified as true or untrue if it turns out 60% of the area was covered by rain.


Uh, no. That is absolutely not true. At all.
 
2012-07-18 03:00:37 PM

downstairs: jack21221: If a weather forecaster says there is a 95% chance of rain, and it doesn't rain, THIS WAS AN ACCURATE REPORT. Rolling a natural 20 in a D&D skill check isn't exactly a rare occurance.

Not quite. Its a 30% (for example) chance of rain because they are saying rain will cover 30% of the area. So if you're going to be having a BBQ in your backyard, there's a 30% chance it will hit you.

But the 30% is an actual prediction of an actual event: 30% of the area will be seeing rain. That can be verified as true or untrue if it turns out 60% of the area was covered by rain.


No.

"The "Probability of Precipitation" (PoP) describes the chance of precipitation occurring at any point you select in the [forecast] area".

Light reading: http://www.srh.noaa.gov/ffc/?n=pop
 
2012-07-18 03:07:04 PM
Having lived there for over half my life, I can safely say the following:

i114.photobucket.com


/ lekker weer vandaag, he?
 
2012-07-18 03:50:12 PM

WalkingSnake: downstairs: jack21221: If a weather forecaster says there is a 95% chance of rain, and it doesn't rain, THIS WAS AN ACCURATE REPORT. Rolling a natural 20 in a D&D skill check isn't exactly a rare occurance.

Not quite. Its a 30% (for example) chance of rain because they are saying rain will cover 30% of the area. So if you're going to be having a BBQ in your backyard, there's a 30% chance it will hit you.

But the 30% is an actual prediction of an actual event: 30% of the area will be seeing rain. That can be verified as true or untrue if it turns out 60% of the area was covered by rain.

No.

"The "Probability of Precipitation" (PoP) describes the chance of precipitation occurring at any point you select in the [forecast] area".

Light reading: http://www.srh.noaa.gov/ffc/?n=pop


Huh. That's not how it has always been explained to me. My bad.
 
2012-07-18 03:52:34 PM

WalkingSnake: No.

"The "Probability of Precipitation" (PoP) describes the chance of precipitation occurring at any point you select in the [forecast] area".

Light reading: http://www.srh.noaa.gov/ffc/?n=pop


Ok, now that this is explained... shouldn't the probability generally go down the further you go out... as the "confidence" part *has* to go down over time.

Yet I generally don't see that.
 
2012-07-18 04:04:22 PM

downstairs: Ok, now that this is explained... shouldn't the probability generally go down the further you go out... as the "confidence" part *has* to go down over time.


Probability of precipitation isn't the same thing as confidence in the prediction. You can predict a 20% chance of rain and be very confident about that, or an 80% chance of rain but not be very confident. In practice, what they do is average together a bunch of forecast simulations, some of which predict rain and some of which don't. The average gives the forecast probability of rain; the spread around that average gives the confidence. Basically it's mean vs. variance.
 
2012-07-18 04:26:15 PM
In Chicago, most of the forecasts keep punting any chance of rain to days 6 and 7 of their 7 day forecasts during this drought. Though we might possibly have a small chance of rain today and tomorrow, maybe, which means it will rain like hell over a 10x10 sq mile area and the rest of us schlepps get nothing. Maybe a sprinkle if you're lucky.
 
2012-07-19 09:50:28 AM

VonEvilstein: Having lived there for over half my life, I can safely say the following:

[i114.photobucket.com image 480x329]


/ lekker weer vandaag, he?


Ek wens ons kan sulke weer kry. Dis net koud en droog hier.
 
Displayed 26 of 26 comments

View Voting Results: Smartest and Funniest


This thread is archived, and closed to new comments.

Continue Farking
Submit a Link »






Report