MGorse Posted April 10, 2012 Share Posted April 10, 2012 I noticed that Philly's original forecast had a high of 58 with 0.71" of rain for April 23rd...the forecast for the same date as of this morning has April 23rd with sunshine and a high of 79. What is wrong with that? Link to comment Share on other sites More sharing options...
Superstorm93 Posted April 11, 2012 Share Posted April 11, 2012 Matching the forecast to its value? Well according to the graph above, they do seem rather poor. I'm also just frustrated with the lack of Pro support over the six years I've been subbed to them. Link to comment Share on other sites More sharing options...
Billabong Posted April 11, 2012 Share Posted April 11, 2012 The old "the public doesn't understand the forecast anyway" argument. The bottom line here folks- this is a money making decision pure and simple. You do not really believe for a minute that any Met worth his or her salt, including Jesse who is bravely defending this product here, really deep down thinks this is worth a hill of beans. However, in the corporate world, if you can sell something worthless why not do it - integrity be damned. The venerable P.T. Barnum principle in effect. Couldn't agree more... Accuweather is a business (obviously, a very successful one), which just like any other business, has a main goal of making money. I've taken a couple weather communication courses with Ray Ban (VP of the Weather Channel) and we've discussed this several times. Time after time, TWCs research shows that the public wants specific numbers (opposing ranges/probability) for all forecasts, even when they know it's far from accurate. Surprised me too..... even though the people knew the forecast that far out was inaccurate, they still wanted it. It's entertainment. If the people want it, the business strategy is to give it to them. Link to comment Share on other sites More sharing options...
ohleary Posted April 11, 2012 Share Posted April 11, 2012 The Dude said it best IMO. They are telling the public what they want to hear instead of telling them that what they want to hear isn't possible to tell them...at least not now or in the near future. Why? $$. Science and the reputation of the field be damned. I'm just grateful I don't work for Accuwx, as I'd be terribly torn, either sacrifice my scientific integrity and defend this travesty, or stick to my principles and quit I guess. Link to comment Share on other sites More sharing options...
BillB Posted April 11, 2012 Share Posted April 11, 2012 The Dude said it best IMO. They are telling the public what they want to hear instead of telling them that what they want to hear isn't possible to tell them...at least not now or in the near future. Why? $$. Science and the reputation of the field be damned. I'm just grateful I don't work for Accuwx, as I'd be terribly torn, either sacrifice my scientific integrity and defend this travesty, or stick to my principles and quit I guess. Although, to be fair to Accuwx, most of us are in this dilemma...just not out to an extreme such as 25 days. Most of the time, I have no confidence in the Day 6/Day 7 forecasts. Regardless of what any verification may say. I think we spend valuable time on these periods...time that could be more effectively used on either the earlier parts of the forecast or on the Forecast Discussion. But this is what the Powers-that-Be have demanded, so "Hail Caesar!" Link to comment Share on other sites More sharing options...
winterymix Posted April 11, 2012 Share Posted April 11, 2012 Although, to be fair to Accuwx, most of us are in this dilemma...just not out to an extreme such as 25 days. Most of the time, I have no confidence in the Day 6/Day 7 forecasts. Regardless of what any verification may say. I think we spend valuable time on these periods...time that could be more effectively used on either the earlier parts of the forecast or on the Forecast Discussion. But this is what the Powers-that-Be have demanded, so "Hail Caesar!" I think it involves rethinking the verbiage. It involves not over-reaching on the shoulders of limited verifiable skill. Simply put, forecasts out to 5 days are worded appropriately. During setups where confidence is good out to days 6 and 7, the current verbiage is fine as well. However, when confidence is low, the verbiage of the forecast should reflect the diminished confidence. For days 6 and 7 in situations of reduced confidence, why not state; "Trending cooler and dryer or trending warmer and more showery". In other words, the reduced confidence is justifiable and it serves the public needs if the forecast isa best guess in a situation of reduced confidence. For days 16 to 25, it would be appropriate to state, "This is the climatology, i.e. average high and low, average frequency of precip." After stating the climatology, it would be appropriate, within the ocnfines of current skill to state that the period from days 16 to 25 appear to be normal, below or above average wrt temperature and the same for precip. What is so wrong about putting out a forecast to day 25 as long as the skill and confidence are not mis-represented? Link to comment Share on other sites More sharing options...
New England Storm Posted April 12, 2012 Share Posted April 12, 2012 AccuWeather should not have called this a "25 day Forecast." How about a "long range outlook." Accuweather should not say on day 20 at point a will be sunny and 60 with a 12 mph wind from the sw. How about day 15-20, " Turning warmer with long spells of rain and thunder possible, some storms might be strong. Day 20-25 Cooler weather is possible, dry and calm days with chilly nights, some frost in rural areas. AccuWeather would still have something new to "sell" to general market Link to comment Share on other sites More sharing options...
phlwx Posted April 22, 2012 Share Posted April 22, 2012 I noticed that Philly's original forecast had a high of 58 with 0.71" of rain for April 23rd...the forecast for the same date as of this morning has April 23rd with sunshine and a high of 79. Consistency fail... They should've stuck to the original forecast...would have been only a day off with the storm. Link to comment Share on other sites More sharing options...
LithiaWx Posted April 23, 2012 Share Posted April 23, 2012 lol Link to comment Share on other sites More sharing options...
Cheeznado Posted April 25, 2012 Share Posted April 25, 2012 Today was the day Norman, OK was supposed to get 1.7" of snow on the 25 day forecast. High today: at least 90. Also, there may have been an all-time record high today for the state of OK. :lol: Link to comment Share on other sites More sharing options...
ohleary Posted April 26, 2012 Share Posted April 26, 2012 Today was the day Norman, OK was supposed to get 1.7" of snow on the 25 day forecast. High today: at least 90. Also, there may have been an all-time record high today for the state of OK. :lol: Seems there are a few bugs to work out. Link to comment Share on other sites More sharing options...
OKpowdah Posted April 29, 2012 Share Posted April 29, 2012 Technically, with enough of a negative correlation, Accuweather could actually become useful... just forecast the opposite Link to comment Share on other sites More sharing options...
metalicwx366 Posted April 29, 2012 Share Posted April 29, 2012 Technically, with enough of a negative correlation, Accuweather could actually become useful... just forecast the opposite Link to comment Share on other sites More sharing options...
GaWx Posted April 29, 2012 Share Posted April 29, 2012 Technically, with enough of a negative correlation, Accuweather could actually become useful... just forecast the opposite Climo was closer on 11 of the 18 days and MUCH closer on 9 of the 18 days lol! Link to comment Share on other sites More sharing options...
cbmclean Posted April 29, 2012 Share Posted April 29, 2012 No way, I've seen it and this is all science, not voodoo. The data is out there (it is NOT free however, which is why take this seriously as a business venture), we're just better use of it. In talking to some other mets who have posted here today, I think the disagreement is an approach for the masses versus one for researchers. The public doesn't want "a 20% chance of temperatures between 49 and 51" -- they want to hear 50. It may be a sad fact for us scientists, but it's true. I'll retell a story that I posted on a blog earlier: We had a large client here a few years ago, that we had presented with an hour-long demo of attractive graphs of the probability of their forecast values being reached (a range of temperatures that dictated whether they could do their work or not). We were all very proud of it. They said, "Well this is all very impressive, but all we need is a Yes/No answer for each time period. Green or red, that's it." So that's what we gave them. I know this is a late reply but today is the first time that I went through this thread. Mr. Ferrell, in my honest opinion, when your client told you this, the morally correct response would have been something like "We are sorry, but we cannot provide the information that you seek, and we cannot in good conscience accept your money to provide a product that we know does not meet your requirements." Link to comment Share on other sites More sharing options...
donsutherland1 Posted June 18, 2012 Share Posted June 18, 2012 With regard to the 20-day and 25-day AccuWeather forecasts, I took a closer look at the temperature forecasts for New York City. I took the 20-day and 25-day forecasts from the April 4 through May 3 period. The 20-day forecasts corresponded to the April 23 through May 22 period. The 25-day forecasts corresponded to the April 28 through May 27 period. 20-day Forecasts: Highs: Days forecast to be above normal: 10 Days forecast to be normal: 1 Days forecast to be cooler than normal: 19 Lows: Days forecast to be above normal: 18 Days forecast to be normal: 2 Days forecast to be cooler than normal: 10 During the 30-day period in question, AccuWeather expected high temperatures to be generally cooler than normal and low temperatures generally warmer than normal. During the 30-days in question, daytime maximum readings were predominantly cooler than normal, while the split between low temperatures was nearly equal. Actual Highs: Above normal: 9 Normal: 1 Below normal: 20 Actual Lows: Above normal: 14 Normal: 4 Below normal: 12 Although the information from the AccuWeather 20-day forecasts seemed to reasonably describe the 30-day period as a whole, the daily details had less agreement with the actual outcomes, particularly when it came to the maximum temperatures. Highs: Days forecast to be above normal: 10; Actual: 1; Correct anomaly: 10% of forecast days Days forecast to be below normal: 19; Actual: 11; Correct anomaly: 58% of forecast days Lows: Days forecast to be above normal: 18; Actual: 15; Correct anomaly: 83% of forecast days Days forecast to be below normal: 10; Actual: 7; Correct anomaly: 70% of forecast days When it came to forecasts of deviations of more than 5° from normal, the following outcomes were seen: Highs: Days forecast to be more than 5° above normal: 2; Actual: 1; Correct anomaly: 50% of forecast days Days forecast to be more than 5° below normal: 7; Actual: 5; Correct anomaly: 71% of forecast days Lows: Days forecast to be more than 5° above normal: 7; Actual: 4; Correct anomaly: 57% of forecast days Days forecast to be more than 5° below normal: 0; Actual: 0; Correct anomaly: Not Applicable AccuWeather vs. Climatology: Average Error in Forecast Highs: AccuWeather: 6.4° Climo: 5.2° Average Error in Forecast Lows: AccuWeather: 4.4° Climo: 4.0° Overall, Climo did better on forecast highs by 0.328 sigma and better on forecast lows by 0.164 sigma. 25-day Forecasts: Highs: Days forecast to be above normal: 6 Days forecast to be normal: 4 Days forecast to be cooler than normal: 20 Lows: Days forecast to be above normal: 12 Days forecast to be normal: 4 Days forecast to be cooler than normal: 14 During the 30-day period in question, AccuWeather expected high temperatures to be generally cooler than normal and low temperatures nearly equally distributed between cooler and warmer than normal. During the 30-days in question, daytime maximum readings were nearly equally distributed between above and below normal, while daily minimum temperatures were predominantly warmer than normal. Actual Highs: Above normal: 13 Normal: 1 Below normal: 16 Actual Lows: Above normal: 19 Normal: 4 Below normal: 7 In terms of the daily details, the following outcomes were seen: Highs: Days forecast to be above normal: 6; Actual: 3; Correct anomaly: 50% of forecast days Days forecast to be below normal: 20; Actual: 11; Correct anomaly: 55% of forecast days Lows: Days forecast to be above normal: 12; Actual: 9; Correct anomaly: 75% of forecast days Days forecast to be below normal: 14; Actual: 5; Correct anomaly: 36% of forecast days When it came to forecasts of deviations of more than 5° from normal, the following outcomes were seen: Highs: Days forecast to be more than 5° above normal: 1; Actual: 0; Correct anomaly: 0% of forecast days Days forecast to be more than 5° below normal: 13; Actual: 8; Correct anomaly: 62% of forecast days Lows: Days forecast to be more than 5° above normal: 5; Actual: 2; Correct anomaly: 40% of forecast days Days forecast to be more than 5° below normal: 2; Actual: 0; Correct anomaly: 0% of forecast days AccuWeather vs. Climatology: Average Error in Forecast Highs: AccuWeather: 5.1° Climo: 5.1° Average Error in Forecast Lows: AccuWeather: 5.1° Climo: 4.2° Overall, AccuWeather and Climo were essentially tied when it came to forecast highs. Climo fared better by a narrow 0.021 sigma margin on forecast highs and better on forecast lows by 0.306 sigma. Link to comment Share on other sites More sharing options...
hm8 Posted June 18, 2012 Share Posted June 18, 2012 Nice work Don. Obviously being worse then climo in most cases does not bode well for accuweather Link to comment Share on other sites More sharing options...
donsutherland1 Posted June 18, 2012 Share Posted June 18, 2012 Nice work Don. Obviously being worse then climo in most cases does not bode well for accuweather It should be noted that my analysis is for a single city over a single 30-day timeframe. I don't have data on how the forecasts fared for other cities. I might run the numbers again at some point during the summer or fall to see how things turn out. Link to comment Share on other sites More sharing options...
RodneyS Posted May 15, 2013 Author Share Posted May 15, 2013 And now -- 30 days: http://articles.chicagotribune.com/2013-05-13/travel/sc-trav-0514-travel-forecast-20130513_1_accuweather-forecast-travel-plans Link to comment Share on other sites More sharing options...
MGorse Posted May 15, 2013 Share Posted May 15, 2013 Here is their press release from earlier this month. Sounds like they have a ton of skill. http://www.accuweather.com/en/press/11918357 30 days of forecasting information helps users plan travel and events State College, PA, May 8, 2013 - With the launch of the new AccuWeather 30-Day Forecast, the leading weather provider has become the only source for free, detailed, 30-Day forecasting information. The addition of this unique forecast allows users of AccuWeather.com and the AccuWeather.com mobile website greater insight in planning trips, sporting events, and outdoor events ahead of time with the Superior Accuracy™ of weather information that is the hallmark of AccuWeather. "AccuWeather is excited about the 30-Day Forecast because it gives our users an unmatched level of insight in planning around weather conditions," Steven Smith, AccuWeather's Chief Digital Officer said. "Our audiences expect deeper and more accurate forecasts from AccuWeather and this is just another example of our company leading the way in meteorological innovation to better serve our customers." The 30-Day Forecast provides the comprehensive level of detail one would expect in an AccuWeather forecast, including: high and low temperatures; the patented AccuWeather RealFeel® temperature; sky conditions and chance of precipitation; wind speed and direction; rain, snow, and ice amounts; thunderstorm probability; UV index; and much more. The AccuWeather 30-Day Forecast is the result of many years of research, backed by great intellectual property and patents, a talented IT staff, and a team of more than 110 meteorologists. In its 50-year history, AccuWeather has consistently pushed the envelope in long-range forecasting. In the 1960s, AccuWeather introduced the five-day forecast when the National Weather Service was offering only a two-day forecast. In later years, AccuWeather has been the first to introduce the 7-day, 10-day, 15-day, 25-day, and now the 30-day detailed forecasts. These forecasts have become industry standards with audiences in all media relying on longer-range forecasts. Link to comment Share on other sites More sharing options...
hurricaneman Posted May 16, 2013 Share Posted May 16, 2013 Here is their press release from earlier this month. Sounds like they have a ton of skill. http://www.accuweather.com/en/press/11918357 30 days of forecasting information helps users plan travel and events State College, PA, May 8, 2013 - With the launch of the new AccuWeather 30-Day Forecast, the leading weather provider has become the only source for free, detailed, 30-Day forecasting information. The addition of this unique forecast allows users of AccuWeather.com and the AccuWeather.com mobile website greater insight in planning trips, sporting events, and outdoor events ahead of time with the Superior Accuracy™ of weather information that is the hallmark of AccuWeather. "AccuWeather is excited about the 30-Day Forecast because it gives our users an unmatched level of insight in planning around weather conditions," Steven Smith, AccuWeather's Chief Digital Officer said. "Our audiences expect deeper and more accurate forecasts from AccuWeather and this is just another example of our company leading the way in meteorological innovation to better serve our customers." The 30-Day Forecast provides the comprehensive level of detail one would expect in an AccuWeather forecast, including: high and low temperatures; the patented AccuWeather RealFeel® temperature; sky conditions and chance of precipitation; wind speed and direction; rain, snow, and ice amounts; thunderstorm probability; UV index; and much more. The AccuWeather 30-Day Forecast is the result of many years of research, backed by great intellectual property and patents, a talented IT staff, and a team of more than 110 meteorologists. In its 50-year history, AccuWeather has consistently pushed the envelope in long-range forecasting. In the 1960s, AccuWeather introduced the five-day forecast when the National Weather Service was offering only a two-day forecast. In later years, AccuWeather has been the first to introduce the 7-day, 10-day, 15-day, 25-day, and now the 30-day detailed forecasts. These forecasts have become industry standards with audiences in all media relying on longer-range forecasts. They should get their product better than climatology before they say its a success, it seems to me that this 25 now 30 day venture is a complete failure Link to comment Share on other sites More sharing options...
Ellinwood Posted June 24, 2013 Share Posted June 24, 2013 http://www.josh-rosenberg.com/2013/06/accuweather-long-rage-forecast-accuracy/ Link to comment Share on other sites More sharing options...
The_Global_Warmer Posted June 26, 2013 Share Posted June 26, 2013 It's morally questionable. It's ethically wrong period. Jesse Farrell and everyone at accuweather knows its Bull****. But they gotta make $$$$$$$. Link to comment Share on other sites More sharing options...
andyhb Posted June 29, 2013 Share Posted June 29, 2013 AccuWeather, laughing stock of the weather community™ since...well...ever. Link to comment Share on other sites More sharing options...
donsutherland1 Posted June 29, 2013 Share Posted June 29, 2013 With regard to the 30-day forecasts, this might be applicable: Dallas Federal Reserve President Richard Fisher recently told an audience: During World War II, future Nobel laureate Ken Arrow served as a weather officer in the Army Air Corps. He and his team were charged with producing month-ahead weather forecasts. Being a disciplined analyst, Arrow reviewed the record of his predictions and, sure enough, confirmed statistically that the corps’ forecasts were no more accurate than random rolls of dice. He asked to be relieved of this futile duty. Arrow’s recollection of the response from on high was priceless: “The commanding general is well aware that the forecasts are no good. However, he needs them for planning purposes.” Perhaps the prospective users of the 30-day forecasts will feel the same way as the general in question? Link to comment Share on other sites More sharing options...
tsteel Posted August 2, 2013 Share Posted August 2, 2013 So this has now been extended to 45 days? I rarely ever go to Accuweather for anything other than a quick forecast while on the go, but it still amazes me that forecasts for such specific items such as wind speed & direction, probability of thunderstorms. etc are included that far out. Link to comment Share on other sites More sharing options...
metalicwx366 Posted August 2, 2013 Share Posted August 2, 2013 So this has now been extended to 45 days?No. Just 25 days. A 25 day forecast. Link to comment Share on other sites More sharing options...
RodneyS Posted August 2, 2013 Author Share Posted August 2, 2013 No. Just 25 days. A 25 day forecast. AccuWeather announced a 30-day forecast on May 8, 2013. I have not seen an announcement of a 45-day forecast; however, if you go to their webpage today (August 2, 2013) and select a city, a 45-day forecast is given. See, for example, Atlanta for September 1st-15th, http://www.accuweather.com/en/us/atlanta-ga/30303/september-weather/348181?monyr=9/1/2013 Link to comment Share on other sites More sharing options...
metalicwx366 Posted August 2, 2013 Share Posted August 2, 2013 AccuWeather announced a 30-day forecast on May 8, 2013. I have not seen an announcement of a 45-day forecast; however, if you go to their webpage today (August 2, 2013) and select a city, a 45-day forecast is given. See, for example, Atlanta for September 1st-15th, http://www.accuweather.com/en/us/atlanta-ga/30303/september-weather/348181?monyr=9/1/2013Oh wow. Apparently they have extended it to 45 days. He was correct. Link to comment Share on other sites More sharing options...
Ed Lizard Posted August 2, 2013 Share Posted August 2, 2013 Does anybody know what they are using for 6 week forecasts? US climate model? European? A weighted average? The Korean? Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.