Page 35 of 66 FirstFirst ... 25333435363745 ... LastLast
Results 341 to 350 of 651

Thread: Elizabeth Warren's Pregnant Pause

  1. #341
    tWebber
    Join Date
    Jan 2014
    Faith
    Christian
    Gender
    Male
    Posts
    3,367
    Amen (Given)
    148
    Amen (Received)
    633
    Quote Originally Posted by lilpixieofterror View Post
    Asteroids donít take a vote in their chosen path. They are guided by the laws of physics. There a reason I can buy a 2020 calendar with the dates of the phases of the moon printed in it.

    This makes a whole lot less sense when you realize that polls are showing what people say what they think or plan to do.

    --Sam
    "I wonder about the trees. / Why do we wish to bear / Forever the noise of these / More than another noise / So close to our dwelling place?" ó Robert Frost, "The Sound of Trees"


  2. #342
    tWebber Teallaura's Avatar
    Join Date
    Jan 2014
    Location
    In my house.
    Faith
    Christian
    Gender
    Female
    Posts
    15,154
    Amen (Given)
    9337
    Amen (Received)
    5600
    Quote Originally Posted by Cow Poke View Post
    I don't think that's a very good example, because the math could be correct, but based on false assumptions or bad data.
    I hate you. And Lil...


    I'm NEVER gonna hear the end of this!!!

    You know DARN WELL he's NEVER gonna let me forget this!!!


    Sam's correct about how probability works.


    The problem with the polling is that multiple different polls got pretty much the same results. THAT is NOT how probability - or polling - work. Part of why we trust weighted results is their consistency - and if they aren't truly representative, they shouldn't be consistent.

    So yes, sometimes the result simply falls in the lower estimate (poor weathermen know this only too well). But multiple different polls shouldn't have that happen (it's still possible but far less likely - like flipping heads one hundred times in a row).



  3. Amen Sam amen'd this post.
  4. #343
    What's that? lilpixieofterror's Avatar
    Join Date
    Jan 2014
    Faith
    Christian
    Gender
    Female
    Posts
    9,617
    Amen (Given)
    1513
    Amen (Received)
    3228
    Quote Originally Posted by Sam View Post
    In 2018, the RCP average from Oct. 13 - Nov. 3 was D +7.3.

    The final electoral result was D +8.4, a difference of ... 1.1%.

    Aggregated polls have been doing just fine the past few years.

    --Sam
    And yet, they seem to not be getting the probable winner right in numbers important races. How well did things go in the last special election?
    Last edited by lilpixieofterror; 10-14-2019 at 08:56 PM.
    "The man from the yacht thought he was the first to find England; I thought I was the first to find Europe. I did try to found a heresy of my own; and when I had put the last touches to it, I discovered that it was orthodoxy."
    GK Chesterton; Orthodoxy

  5. #344
    tWebber Teallaura's Avatar
    Join Date
    Jan 2014
    Location
    In my house.
    Faith
    Christian
    Gender
    Female
    Posts
    15,154
    Amen (Given)
    9337
    Amen (Received)
    5600
    Quote Originally Posted by Sam View Post
    This makes a whole lot less sense when you realize that polls are showing what people say what they think or plan to do.

    --Sam
    Polls are always cross sectional.

  6. #345
    What's that? lilpixieofterror's Avatar
    Join Date
    Jan 2014
    Faith
    Christian
    Gender
    Female
    Posts
    9,617
    Amen (Given)
    1513
    Amen (Received)
    3228
    Quote Originally Posted by Teallaura View Post
    I hate you. And Lil...


    I'm NEVER gonna hear the end of this!!!

    You know DARN WELL he's NEVER gonna let me forget this!!!


    Sam's correct about how probability works.


    The problem with the polling is that multiple different polls got pretty much the same results. THAT is NOT how probability - or polling - work. Part of why we trust weighted results is their consistency - and if they aren't truly representative, they shouldn't be consistent.

    So yes, sometimes the result simply falls in the lower estimate (poor weathermen know this only too well). But multiple different polls shouldn't have that happen (it's still possible but far less likely - like flipping heads one hundred times in a row).


    The issue is they donít seem to be doing all that great picking the most probable winner lately.
    "The man from the yacht thought he was the first to find England; I thought I was the first to find Europe. I did try to found a heresy of my own; and when I had put the last touches to it, I discovered that it was orthodoxy."
    GK Chesterton; Orthodoxy

  7. #346
    tWebber
    Join Date
    Jan 2014
    Faith
    Christian
    Gender
    Male
    Posts
    3,367
    Amen (Given)
    148
    Amen (Received)
    633
    Quote Originally Posted by Teallaura View Post
    I hate you. And Lil...


    I'm NEVER gonna hear the end of this!!!

    You know DARN WELL he's NEVER gonna let me forget this!!!


    Sam's correct about how probability works.


    The problem with the polling is that multiple different polls got pretty much the same results. THAT is NOT how probability - or polling - work. Part of why we trust weighted results is their consistency - and if they aren't truly representative, they shouldn't be consistent.

    So yes, sometimes the result simply falls in the lower estimate (poor weathermen know this only too well). But multiple different polls shouldn't have that happen (it's still possible but far less likely - like flipping heads one hundred times in a row).


    Poll herding has been pointed out as a reason polls seem to clump together sometimes. I know that Nate Silver has talked about how the 538 algorithms flag suspected herding. But we often should see different pollsters arrive at similar results if they're polling like groups (i.e., Nevada residents, national poll). That could be an example of herding but at least just as likely could show that pollsters are pulling from representative samples.

    --Sam
    "I wonder about the trees. / Why do we wish to bear / Forever the noise of these / More than another noise / So close to our dwelling place?" ó Robert Frost, "The Sound of Trees"


  8. #347
    tWebber
    Join Date
    Jan 2014
    Faith
    Christian
    Gender
    Male
    Posts
    3,367
    Amen (Given)
    148
    Amen (Received)
    633
    Quote Originally Posted by lilpixieofterror View Post
    And yet, they seem to not be getting the probable winner right in numbers important races. How well did things go in the last special election?

    Polls don't "pick" the probable winner.

    An election prediction is necessarily a question of probability, not certainty.

    Special elections are not typically well-polled.

    Trump won 2016 from <100,000 votes in three states. Pretty hard to predict that result specifically but it would fall in the ~30% probability.

    --Sam
    "I wonder about the trees. / Why do we wish to bear / Forever the noise of these / More than another noise / So close to our dwelling place?" ó Robert Frost, "The Sound of Trees"


  9. #348
    tWebber Teallaura's Avatar
    Join Date
    Jan 2014
    Location
    In my house.
    Faith
    Christian
    Gender
    Female
    Posts
    15,154
    Amen (Given)
    9337
    Amen (Received)
    5600
    Quote Originally Posted by lilpixieofterror View Post
    And yet, they seem to not be getting the probable winner right in numbers important races. How well did things go in the last special election?
    Lil, Horsey is skeletal - why are you still beating him?


    The problem lies in the tiny sample sizes and the weighting - yes, there was/is a problem. No, they're not all wrong.

    Why on Earth are you debating this with Sam?!

  10. #349
    tWebber Teallaura's Avatar
    Join Date
    Jan 2014
    Location
    In my house.
    Faith
    Christian
    Gender
    Female
    Posts
    15,154
    Amen (Given)
    9337
    Amen (Received)
    5600
    Quote Originally Posted by Sam View Post
    Poll herding has been pointed out as a reason polls seem to clump together sometimes. I know that Nate Silver has talked about how the 538 algorithms flag suspected herding. But we often should see different pollsters arrive at similar results if they're polling like groups (i.e., Nevada residents, national poll). That could be an example of herding but at least just as likely could show that pollsters are pulling from representative samples.

    --Sam
    Zebra - I'd rather give the horse a good once over before worrying about him.

  11. #350
    tWebber Teallaura's Avatar
    Join Date
    Jan 2014
    Location
    In my house.
    Faith
    Christian
    Gender
    Female
    Posts
    15,154
    Amen (Given)
    9337
    Amen (Received)
    5600
    Quote Originally Posted by lilpixieofterror View Post
    The issue is they donít seem to be doing all that great picking the most probable winner lately.
    Yep - sorta. But y'all are way out in the weeds - and you know danged well he'll never admit a problem with polling now.

    FYI: the way you phrased this makes no sense - they 'pick the most probable winner' just fine - not so hot on the actual winner, however. I get you but this isn't really correct.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •