Jump to content
  • Member Statistics

    17,502
    Total Members
    7,904
    Most Online
    Weathernoob335
    Newest Member
    Weathernoob335
    Joined

Vendor, Blog and TV Channel Forecasts Thread


earthlight

Recommended Posts

Through May 22, the average forecasting error is as follows:

 

AccuWeather 90-day Forecast: 6.0°

Climatology: 5.3°

 

In terms of % of better forecasts (with each high or low temperature constituting a forecast), the tally for better forecasts is as follows:

 

AccuWeather 90-day Forecast: 39%

Climatology: 52%

 

Tied: 9%

Link to comment
Share on other sites

  • Replies 2.8k
  • Created
  • Last Reply

Today saw Central Park reach 90° for the first time this year. Ironically, the 90-day AccuWeather forecast from today still shows no 90° readings (which is through August 23).

 

In terms of the first 90° reading, the crudely-constructed analogs did better than AccuWeather's forecast. Those analogs were also off the mark with a June 27 date (#2751).

 

Just as climo will very likely outperform the 90-day forecasts for the May-June period for which I'm comparing the two, climatology related to the first 90° day would have done very well:

 

1950-2015:

Median first date: June 1

Earliest 10% of dates: April 27 or earlier

Latest 10% of dates: June 24 or later

 

Average daily error (5/1-26*):

AccuWeather: 6.1°

Climatology: 5.5°

 

*-Based on today's 90°-69° range through 2 pm.

Link to comment
Share on other sites

Climatology outperformed the AccuWeather 90-day forecasts (http://www.americanwx.com/bb/index.php/topic/39531-vendor-blog-and-tv-channel-forecasts-thread/?p=4082148) listed earlier in this thread. Climatology did better in terms of average error, % of cases with errors < 5°, and monthly mean temperature. The full data will be posted later today once NYC's high temperature has been registered.

Link to comment
Share on other sites

May 2016 finished with a mean temperature of 62.8°. That was 0.4° above normal.

 

During May, climatology fared better than AccuWeather's 90-day forecast. Climatology had an average error of approximately 13% less than the average error of the AccuWeather forecast. In short, the 90-day forecast showed no skill.

 

May2016_Accu_Wx_Climo.jpg

Link to comment
Share on other sites

  • 3 weeks later...

Through June 22, the average daily forecasting errors from the forecasts in #2751 (April 8) are below:

 

Climo06222016.jpg

 

It is now all but certain that climatology will outperform the AccuWeather 90-day forecast for the May-June timeframe. It is also possible that in the extended part of the forecasting range (June), the simple analogs constructed from ENSO/SSTAs may also outperform the AccuWeather 90-day forecast.

 

If that happens, I will run another test using the AccuWeather 90-day forecast, Climatology, and the simple Analogs for September based on the 7/3 AccuWeather forecast (which would encompass 9/1-30) to see if the outcome is an aberration. I would again expect climatology to do better than the AccuWeather 90-day forecast.

Link to comment
Share on other sites

The September numbers:

 

Monthly Anomaly:

AccuWeather: -0.4°

Analogs: +2.9°

 

September2016.jpg

We're lately finding with our changing climate analogs mean little and besides those averages are based on 1981-2010 normal temps, which don't reflect today's climate..

Link to comment
Share on other sites

We're lately finding with our changing climate analogs mean little and besides those averages are based on 1981-2010 normal temps, which don't reflect today's climate..

I don't expect the analogs to have much relevance. The major point was that the AccuWx 90-day forecasts have no meaningful value. Climatology is superior. Even very simple analogs are competitive (actually they did better in the extended part of that forecast range last time).

Link to comment
Share on other sites

I don't expect the analogs to have much relevance. The major point was that the AccuWx 90-day forecasts have no meaningful value. Climatology is superior. Even very simple analogs are competitive (actually they did better in the extended part of that forecast range last time).

My guess is that July will have the biggest anomaly maybe 2 degrees above average while August an Sept are near normal, a degree either way, nevertheless long range weather predictions have never been reliable  not even today.

Link to comment
Share on other sites

Amy Freeze says 90 and above Wednesday thru at least the 20th with more heat reloading after that

Try quoting another met on that channel because she is horrible. Constantly making mistakes on-air and has the most awkward exchanges with other anchors. A lot of awkward silence when she's on-air. Not saying that she's going to be wrong but she is terrible

Link to comment
Share on other sites

Try quoting another met on that channel because she is horrible. Constantly making mistakes on-air and has the most awkward exchanges with other anchors. A lot of awkward silence when she's on-air. Not saying that she's going to be wrong but she is terrible

I don't think they hired her for her brains.

Link to comment
Share on other sites

Try quoting another met on that channel because she is horrible. Constantly making mistakes on-air and has the most awkward exchanges with other anchors. A lot of awkward silence when she's on-air. Not saying that she's going to be wrong but she is terrible

The mets work as a team there. She isnt making up model info herself anyhow. I like her and actually she is quite popular..nothing awkward about her..maybe you feel awkward watching her and thats the issue

Link to comment
Share on other sites

The mets work as a team there. She isnt making up model info herself anyhow. I like her and actually she is quite popular..nothing awkward about her..maybe you feel awkward watching her and thats the issue

I'm sure Jeff Smith/Lee Goldberg make up her forecasts the night before and she just wings it the next morning. Yes she is pretty but I for one hate the fact that these stations hire mets for their looks. They all do. Put a real, professional met on air at all times. There's a good reason that she's stuck with the weekend morning shift

Link to comment
Share on other sites

I'm sure Jeff Smith/Lee Goldberg make up her forecasts the night before and she just wings it the next morning. Yes she is pretty but I for one hate the fact that these stations hire mets for their looks. They all do. Put a real, professional met on air at all times. There's a good reason that she's stuck with the weekend morning shift

She was the chief met at another station before coming there so I'm pretty sure you're wrong.
Link to comment
Share on other sites

She was the chief met at another station before coming there so I'm pretty sure you're wrong.

the previous female met before her heidi jones was much more knowledgeable and descriptive in her weatehr updates.. than amy.. but she was let go for bizarre reasons..

Link to comment
Share on other sites

He and lee both have Cornell met degrees. I prefer lee. Amy freeze has a masters in environmental science from Penn. So she's not dumb but also not technically a met. I find her ackward too

Lee has turned into a ham. Yeah we get it, its cold out you don't need to stand outside the studio so we can see your breath. I realize they make him do that I just find it annoying

Link to comment
Share on other sites

  • Rjay locked this topic

Archived

This topic is now archived and is closed to further replies.

Guest
This topic is now closed to further replies.
  • Recently Browsing   0 members

    • No registered users viewing this page.

×
×
  • Create New...