About

I am a writer and media weather observer. In the last decade, it seems to me that weather has become a bigger topic in the media, often being used as the lead "tease" to kick off the news. As a result, I've often wondered if forecasters feel the ratings pressure and, wittingly or unwittingly, tend toward the sensational in presenting their forecasts. Beyond that, I've always been interested in determining which weather forecasters are more accurate than others and, ultimately, in grading their performance. This is a vehicle through which I hope to do that.

If you a have a question or comment about this blog, you may send it mnforecaster@gmail.com.

35 comments:

  1. Do you have a way to be contacted? I don't see an email address anywhere on your site. Thanks.

    ReplyDelete
  2. Yes, email address mnforecaster@gmail.com

    ReplyDelete
  3. Love the site! I often compare between all the media outlets, and too have wondered if a bit of sensationalism was occurring. I have also noticed in the past that some channels consistently forecast high while others were low, although this year has been more variable. I'll definitely be coming back again. Thanks!

    ReplyDelete
  4. Thanks for the feedback, Anonymous!

    ReplyDelete
  5. @NovakWeather on TwitterDecember 21, 2010 at 4:08 PM

    How do you handle weather forecasts that use percentages or such words as chance, maybe, possibility, could, etc.?

    ReplyDelete
  6. @NovakWeather on TwitterDecember 21, 2010 at 4:13 PM

    Also, I happened to catch Belinda Jensen's weather forecast on Saturday AM. She predicted an inch or less of snow for yesterday and never mentioned much about a potential storm while all the other stations were hitting it hard and hyping it up. However, I see that you graded KARE11 with a B- on this recent storm. How is this possible?

    ReplyDelete
  7. @NovakWeather,
    Words such as the ones you mentioned are, of course, difficult to work into the analysis. I guess I try to get an overall sense of what they are projecting, considering the way they emphasize certain words more than others, etc. KSTP will sometimes include percent chances of precipitation in their forecasts, which are helpful.

    I didn't catch Belinda Jensen on Saturday a.m. But if I can take your words to the bank, I should change their grade. I run into difficulties when some stations don't have the news when others do. So I can't be perfectly uniform in how I gather the forecasts. Sometimes, I'll go to the website to try to get information, but that's often outdated (even though they promote their websites as having "up to the minute information." Also, some webmasters are faster than others at updating the weather video broadcasts online. It's not the perfect systematic process I'd like, but I try to keep to the goal.

    ReplyDelete
  8. This comment has been removed by a blog administrator.

    ReplyDelete
  9. 3 years ago we had an unforecasted snow "event" about a week before Thanksgiving, the first snow of the season. My son was following a woman who went out of control on an overpass. My son avoided her and then was struck from behind by another woman. His car was pushed into the side of the 1st vehicle and then onto the top of the guardrail. His injuries only involved some minor stiffness. I will never forget, though, that a 17 year old Burnsville boy died during that same unforecasted event when his car slid out of control. While some of the forecasters may be a little over the top, if they save one life every "event", I say go for it.

    ReplyDelete
  10. @NovakWeather on TwitterJanuary 3, 2011 at 9:32 AM

    Bill,
    Do you have an average overall grade posted somewhere or simply just individual storm grades?

    ReplyDelete
  11. Yes, I do. If you click on the "Report Cards" link at the top, it will take to what you're looking for. Alternatively, here's the link that will take you directly there: https://spreadsheets.google.com/ccc?key=0ApMK5lrd3TqkdEl3d3JMdDZJeDYxaFZScVFRQnpncnc&hl=en&authkey=CN773YAE#gid=0

    If you have ideas or suggestions on how to improve it, I'm all ears.

    ReplyDelete
  12. @NovakWeather on TwitterJanuary 3, 2011 at 9:47 AM

    Bill, I take it that your grading only goes back to early December?

    Also, below you will see my predictions for MSP and RST metro areas from last week. What kind of a grade would you give me? You can verify these comments by going to @NovakWeather on Twitter.

    Impressive 'dry slot' is cutting off precipitation in eastern MN. Don't expect much, if anything, from this storm.
    31 Dec
    Freezing rain/sleet will start in RST around noon and in MSP by 1pm; will end by 5pm. Roads will become icy.
    31 Dec
    No appreciable snow accum. expected in MSP or RST. An inch at best.
    31 Dec
    Western MN will get hammered again with heavy snow and blizzard conditions later today.
    31 Dec
    Expect a brief period freezing rain/sleet from 1pm - 7pm. That is it. Simply not a big storm for southeast MN. It will be icy.
    31 Dec
    Road conditions will rapidly deteoriorate as you head NW on I-94 from MSP metro today and tomorrow.
    30 Dec
    Fargo, Aberdeen SD, Int'l Falls, Bemidji, Alex will all get heavy snow. St.Cloud, MSP, RST will get mostly ICE or rain.
    30 Dec
    This will prove to be a major Winter Storm for western MN NOT eastern MN today and tomorrow.
    30 Dec
    Everyone will experience some ICE accumulation tomorrow afternoon. RST will warm to near 32 Fri. evening. This will help.
    30 Dec
    Expect all liquid precip. today and tomorrow. The main concern is for ICE accum. by noon Friday, esp. from MSP north and west.
    30 Dec Favorite
    A 36 hour thaw has officially started. Temps crash on Friday.
    29 Dec
    Some areas of western MN will receive well over a foot of snow from Thur. thru Sat. AM.

    ReplyDelete
  13. Yep, I just started the site prior to the big storm.

    It's kind of hard to go back and grade retroactively given the way I check forecasts at specific times so everyone can be compared on equal footing. Maybe I can gather your info for future storms provided you can give forecasts for MSP. Do you have a blog or website or just the Twitter? Feel free to email me at mnforecaster@gmail.com.

    ReplyDelete
  14. @NovakWeather on TwitterJanuary 3, 2011 at 12:32 PM

    Bill,
    You are right on. Following me on Twitter is the best way to go. I see that you are already doing this.

    ReplyDelete
  15. I like the site. Fox rated number 1 surprised me as I think they tease and hype up the weather the most during the whole broadcast. When you take their predictions are you looking online or each channel at a certain time. The last storm ian said 1 to 3 inches that was way off while sevn was a dusting to an inch on tv. Fox 9 also seems the most vague in calling numbers to certain areas.

    ReplyDelete
  16. As for Fox9 and the others, time will tell. I try to gather my information from the telecasts. If I miss them, I can usually find them online (if the stations are timely with the uploading). Otherwise, I have to go with text that appears on each station's website.

    Thanks for the feedback!

    ReplyDelete
  17. Bill,
    After listening to your radio interview and viewing your website, I have a few suggestions for your grading format. Currently, it is way to subjective and you need to make this more objective.

    First off, you must require (via an offer) that each media outlet submit a forecast at 10am and 10pm (or whatever time). This is an appropriate time because the latest NAM computer model run comes out by then.

    Secondly, you must have them make the prediction for a certain location within the MSP metro area (MSP airport or Chanhassen NWS would seem appropriate).

    Finally, you can't let the forecasters use the words chance, maybe, possibly, could, etc. Using these words will make your job more difficult since it will become subjective again.

    Just my .02.

    ReplyDelete
  18. 1" to 3" of snow is underestimating this early week storm totals. Especially on the southwest side of the metro.

    ReplyDelete
  19. Bill, I've always said that it is foolish to attempt an accurate prediction much more than 5 days in advance. Long range computer models can provide a skeletal "blue print" of what may happen, but overall, they are not a good at predicting the nuances that greatly affect our weather from day to day.

    Meteorologists that attempt to accurately predict more than 5 days in advance are foolish. Meanwhile, these same weather professionals are often the ones who tend to glorify or hype a situation. In turn, the public becomes "terrorized" by the weather industry yet it was the few that spoiled it for the whole.

    ReplyDelete
  20. The 'Farmer's Almanac' and Dr. Bill Gray (Hurricane predictor) can be considered 2 of the greatest weather "terrorists" in the country. They attempt to predict months in advance. Absolutely foolish!

    ReplyDelete
  21. I agree @NovakWeather. Credibility is the only thing meteorologists have. Why do anything to jeopardize that?

    ReplyDelete
  22. Bill:

    what a great site this is. I've been watching this cold snap. Keep in mind that Chikage's forecast either last Saturday night or Sunday morning said low's of -15. Most of the other on air local mets have been calling for low's between -5 and -10.

    ReplyDelete
  23. Bill,
    In your last posting about 7 day forecasts, you say "The data suggests that when it's within four days of the day that interests you, the variance among forecasters is rather small; it wouldn't much matter whose forecast you followed."

    What people need to understand is that most weather broadcasters base their 7 day forecasts off the same computer model guidance (GFSx or GFS). That is why their temps are so close to one another.

    I would argue that Paul Douglass & KSTP actually deviate from the the others by sticking their neck's out on a line more often than others and going with a more educated guess rather than straight model statistics (MOS guidance). In other words, utilizing MOS guidance is the "easy" way out.

    I applaud the STRIB and KSTP for putting more of their own spin on a weather forecast. However, as I've said before, it is foolish to attempt to accurately predict the weather more than 5 days in advance.

    ReplyDelete
  24. Agreed. Anyone who shows original thinking rather than total faith in the models is to be admired!

    MPR has quite the attention-getting headline today on their blog: http://minnesota.publicradio.org/collections/special/columns/updraft/

    ReplyDelete
  25. I often wonder why weather reporters keep using the term "unseasonably cold" when referring to this recent cold spell. I have lived in Minnesota for almost 50 years and this seems pretty darn seasonable to me!!

    ReplyDelete
  26. Why is it necessary to get a Goggle account in order to view parts of the site?

    ReplyDelete
  27. Could you tell me more about what parts of the site you can't view? I've not heard about this problem and would like to know more. You may also email mnforecaster@gmail.com. Thank you!

    ReplyDelete
  28. Bill,
    I'm curious why you are not following the snow system for early next week. Please elaborate.

    ReplyDelete
  29. When I got my first computer in 1978 I wrote a small program to track and evaluate the accuracy of local television forecasters. It was a simple program and I only tracked high temperature data. I noted the forecasted high temperature for the next day after an evening telecast. I did this for six months and found that KARE-11 had the best record... but only 35% of the time. Interesting to see that they're still more accurate than the others.

    ReplyDelete
  30. @NovakWeather on TwitterFebruary 1, 2011 at 3:58 PM

    Bill,
    Still no lovin' for @NovakWeather? I believe that your grades on this last storm are too generous. Forecasters should have been on this storm by Friday.

    ReplyDelete
  31. @NovakWeather, please send an email to mnforecaster@gmail.com so we can discuss offline!

    ReplyDelete
  32. Great blog. I am a Minnesota native meteorologist and a general meteorology nerd. Follow me on twitter at pv_anomaly. Great reads and I enjoy your assessments. Check out my weather blog too.

    ReplyDelete
  33. This comment has been removed by a blog administrator.

    ReplyDelete
  34. Thanks for the site, and facebook updates... I like it so much I'm sending you an invite for +, hope to see you on there too!

    ReplyDelete
  35. I use a Private weather forecaster located in metro and he is by far more accurate then the tv forecasters. Craig Berthiaume at Weather Watch.

    ReplyDelete