Accuweather launches "revolutionary" 45-day forecasts.....

[jtmal0723]

You have got to be kidding me, right? Evidently earlier this week, Accuweather launched their "Revolutionary" 45-day forecasts. I have one thing to say: lolwut?

http://www.washingtonpost.com/blogs/capital-weather-gang/wp/2013/08/06/accuweather-you-cannot-be-serious-new-45-day-forecasts/

[twcclassics]

A joke indeed. I use the AccuWeather Firefox add-on (b/c the weather.com one is STILL incompatible), and they rarely even get my 3 day forecast right!

[Trevor]

As I posted on Facebook earlier this week:

[Metarvo]

A joke indeed.

Seconded.  It's hard enough for most meteorologists to get 7-day forecasts right, except for the really good ones of course.  I believe it was said a while back that AccuWeather is actually AccuWrong.

[jtmal0723]

Weather forecasting is generally only accurate up to 48 hours in to the future. Anything after that is usually left up to chance and how things flow.

[TWCCraig]

Accuwrong. They can't even get the 5-day forecast right, not even close as NWS. I remember when they started doing 25-day forecasts, now 45? The only models I know of that go out that far are the Euro weaklies and CFSv2. Even at that range, those model's accuracies are poor, and only useful for mid and upper level patterns in the 750mb-100mb layers, they often show poor surface reflections and outputs. I saw a mini study someone did on how accurate Accuweather's 25-day forecasts are, you're better off using climatology standards (1981-2010), which were shown to be more accurate than Accuweather.

[Jonathan]

A joke indeed. I use the AccuWeather Firefox add-on (b/c the weather.com one is STILL incompatible), and they rarely even get my 3 day forecast right!

Heck, I've seen them say there were storms here on the current conditions...when the skies were virtually clear for hundreds of miles. How does that even happen?

[gt1racerlHDl]

Accuweather must be really bored to come up with that 

[phw115wvwx]

Just use my scoring system from the forecasting contest here to verify these day 10-45 forecasts.  I'm almost certain AccuWeather will bust quite a bit on each one.

[twcclassics]

I believe it was said a while back that AccuWeather is actually AccuWrong.

Maybe someone else did, but I call them Not-So-Accu-Weather.

Heck, I've seen them say there were storms here on the current conditions...when the skies were virtually clear for hundreds of miles. How does that even happen?

That happens to me all the time! I don't understand it. I think the weather is at least in the general area, but it's not in Omaha.

[WeatherWitness]

Is AccuWeather just trying to embarrass themselves?   As Craig and the article said, I remember when AccuWeather launched the 25-day forecasts and everyone making a big fuss over that.  Now we're at 45, even though 25 was still far from accurate.  Heck, even the NWS (which I consider to be a better source than AccuWeather) had a difficult time with a forecast only 2 days out when I was in New York.  A forecast of 81 degrees for a high and sunny skies was in reality a high of 68 degrees and rain.  If we struggle with forecasts only a few days out, how can we even begin to think (detailed) forecasts over a month out will be of any use?

And I know AccuWeather claims it's for planning purposes, etc. but I bet even the average person who knows nothing about weather models or whatever would think that relying on a 45-day forecast was crazy.

[Metarvo]

I just got my eyes on one of those forecasts, and it disgusted me for two reasons.  First of all, there were weather conditions, highs, and lows listed for over a month out.  As if that weren't bad enough, it was nearly the same forecast every day.  Partly cloudy on most days (a hedge if you ask me), with mid to upper 80's for highs.  The lone exception was Halloween, with a high of 59 °F.  I dearly hope that doesn't pan out, because it would mean a very warm start to fall.  That would be the other reason I'm disgusted.  I'm actually hoping it's wrong.