Princeton Election Model predicts O-303, R-235; Obama with 92-98% odds. Model was only one electoral vote off in 2008.

  • TroyAthlete

    Posts: 4269

    Oct 30, 2012 6:42 PM GMT
    http://election.princeton.edu/


    princeton election


    Sam Wang's model was even more accurate in 2008 than Nate Silver, although Silver did correctly predict EVERY 2008 Senate race using his own model. Silver's model weights national polls, while Wang's Priceton model only uses state polls.


    EV_map.png
  • Posted by a hidden member.
    Log in to view his profile

    Oct 30, 2012 6:43 PM GMT
    Tick tock.
  • CuriousJockAZ

    Posts: 19138

    Oct 30, 2012 6:43 PM GMT
    Yes, by all means, grasp at this straw. It will be proved terribly wrong. See you Election Night icon_wink.gif And Ex_Mil8, since you brought it up...

  • TroyAthlete

    Posts: 4269

    Oct 30, 2012 6:47 PM GMT
    CuriousJockAZ saidYes, by all means, grasp at this straw. It will be proved terribly wrong. See you Election Night icon_wink.gif And Mil, since you brought it up...



    U mad?

    Yes, by all means, offer no data, facts, or figures. Just call a highly accurate model from an Ivy League professor who spends his days studying facts and stats "grasping at straws" because, after all, you're a Republican to whom facts and arithmetic don't matter. Republicans: "We don't need no stinkin' numbers! The local right-wing witch doctor says Romney is up 5000000000 million!!!" Hahaha.

    P.S. Spamming the boards with that stupid countdown isn't going to increase Romney's chances, hahahaha.
  • Posted by a hidden member.
    Log in to view his profile

    Oct 30, 2012 10:19 PM GMT
    CuriousJockAZ saidYes, by all means, grasp at this straw. It will be proved terribly wrong. See you Election Night icon_wink.gif And Ex_Mil8, since you brought it up...



    I've posted this several times---it's correctly called the last few elections with excellent precision, including those where republicans won.

    I look forward to jizzing on your face after the election X
  • coolarmydude

    Posts: 9190

    Oct 31, 2012 1:29 AM GMT
    TroyAthlete said
    U mad?

    Yes, by all means, offer no data, facts, or figures. Just call a highly accurate model from an Ivy League professors who spends his days studying facts "grasping at straws" because, after all, you're a Republican to whom facts and arithmetic don't matter. Republicans: "We don't need no stinkin' numbers! The local right-wing witch doctor says Romney is up 5000000000 million!!!" Hahaha.

    P.S. Spamming the boards with that stupid countdown isn't going to increase Romney's chances, hahahaha.


    Damn! That's the best smack down on CJAZ ever!
  • Webster666

    Posts: 9217

    Oct 31, 2012 1:32 AM GMT
    TroyAthlete said
    CuriousJockAZ saidYes, by all means, grasp at this straw. It will be proved terribly wrong. See you Election Night icon_wink.gif And Mil, since you brought it up...



    U mad?

    Yes, by all means, offer no data, facts, or figures. Just call a highly accurate model from an Ivy League professors who spends his days studying facts "grasping at straws" because, after all, you're a Republican to whom facts and arithmetic don't matter. Republicans: "We don't need no stinkin' numbers! The local right-wing witch doctor says Romney is up 5000000000 million!!!" Hahaha.

    P.S. Spamming the boards with that stupid countdown isn't going to increase Romney's chances, hahahaha.



    I like this response, a lot.
    And, I LOVE the original post !
  • musclmed

    Posts: 3284

    Oct 31, 2012 1:53 AM GMT
    Webster666 said
    TroyAthlete said
    CuriousJockAZ saidYes, by all means, grasp at this straw. It will be proved terribly wrong. See you Election Night icon_wink.gif And Mil, since you brought it up...



    U mad?

    Yes, by all means, offer no data, facts, or figures. Just call a highly accurate model from an Ivy League professors who spends his days studying facts "grasping at straws" because, after all, you're a Republican to whom facts and arithmetic don't matter. Republicans: "We don't need no stinkin' numbers! The local right-wing witch doctor says Romney is up 5000000000 million!!!" Hahaha.

    P.S. Spamming the boards with that stupid countdown isn't going to increase Romney's chances, hahahaha.



    I like this response, a lot.
    And, I LOVE the original post !

    http://spectrum.ieee.org/geek-life/profiles/electoral-college-math

    No sense in arguing over a prediction a week from the election.

    Despite Romney's advance in the polls, it seems to have ensuired Obama's win in the eyes of Wang.

    in September the chance of a Obama win was 87 %, now its 98%
  • Posted by a hidden member.
    Log in to view his profile

    Oct 31, 2012 2:04 AM GMT
    TroyAthlete said
    CuriousJockAZ saidYes, by all means, grasp at this straw. It will be proved terribly wrong. See you Election Night icon_wink.gif And Mil, since you brought it up...



    U mad?

    Yes, by all means, offer no data, facts, or figures. Just call a highly accurate model from an Ivy League professors who spends his days studying facts "grasping at straws" because, after all, you're a Republican to whom facts and arithmetic don't matter. Republicans: "We don't need no stinkin' numbers! The local right-wing witch doctor says Romney is up 5000000000 million!!!" Hahaha.

    P.S. Spamming the boards with that stupid countdown isn't going to increase Romney's chances, hahahaha.


    BEST.RESPONSE.EVER.
  • musclmed

    Posts: 3284

    Oct 31, 2012 2:09 AM GMT
    see above David
  • CuriousJockAZ

    Posts: 19138

    Oct 31, 2012 3:45 AM GMT
    TroyAthlete said

    P.S. Spamming the boards with that stupid countdown isn't going to increase Romney's chances, hahahaha.




    Pssssst...Spamming the boards with one model or another, graphs, whatnot are not going to help Obama save his sinking campaign.
  • TroyAthlete

    Posts: 4269

    Oct 31, 2012 7:02 AM GMT
    coolarmydude said
    TroyAthlete said
    U mad?

    Yes, by all means, offer no data, facts, or figures. Just call a highly accurate model from an Ivy League professors who spends his days studying facts "grasping at straws" because, after all, you're a Republican to whom facts and arithmetic don't matter. Republicans: "We don't need no stinkin' numbers! The local right-wing witch doctor says Romney is up 5000000000 million!!!" Hahaha.

    P.S. Spamming the boards with that stupid countdown isn't going to increase Romney's chances, hahahaha.


    Damn! That's the best smack down on CJAZ ever!


    In his defense, he's fighting with the handicap of having his head buried up his Republican masters' butts.
  • TroyAthlete

    Posts: 4269

    Oct 31, 2012 7:03 AM GMT
    CuriousJockAZ said
    TroyAthlete said

    P.S. Spamming the boards with that stupid countdown isn't going to increase Romney's chances, hahahaha.




    Pssssst...Spamming the boards with one model or another, graphs, whatnot are not going to help Obama save his sinking campaign.


    HahahaHaha. You gotta love the Republican brain. Facts, figures, and numbers are "spam." This explains a lot.
  • CuriousJockAZ

    Posts: 19138

    Oct 31, 2012 2:31 PM GMT
    TroyAthlete said

    Just call a highly accurate model from an Ivy League professor who spends his days studying facts and stats "grasping at straws" because, after all, you're a Republican to whom facts and arithmetic don't matter



    Let's revisit this "highly accurate model" next Tuesday night and maybe you will understand why this is yet another "grasping at straws" example you've served up of which alleged "facts and arithmetic" DON'T matter when predicting a future outcome of an election.
  • Posted by a hidden member.
    Log in to view his profile

    Oct 31, 2012 3:02 PM GMT
    I think next Tuesday Curious will be crying and wondering why he looks like such a silly little girl.
  • CuriousJockAZ

    Posts: 19138

    Oct 31, 2012 3:08 PM GMT
    smartmoney saidI think next Tuesday Curious will be crying and wondering why he looks like such a silly little girl.


    Guess we'll have to wait and see won't we. Please make note that if Obama wins, I will gladly show up here and take all the ribbing you guys will surely hurl my way and eat countless mouthfuls of crow. However, if Romney emerges the victor, I don't expect to see any of the seemingly self-described political geniuses around here being anywhere to be found. We'll see who the "crying little girls" really are, and it won't be me regardless of the outcome.
  • Posted by a hidden member.
    Log in to view his profile

    Oct 31, 2012 3:18 PM GMT
    See you Tuesday girlfriend.
  • Posted by a hidden member.
    Log in to view his profile

    Oct 31, 2012 3:34 PM GMT
    CuriousJockAZ said
    TroyAthlete said

    Just call a highly accurate model from an Ivy League professor who spends his days studying facts and stats "grasping at straws" because, after all, you're a Republican to whom facts and arithmetic don't matter



    Let's revisit this "highly accurate model" next Tuesday night and maybe you will understand why this is yet another "grasping at straws" example you've served up of which alleged "facts and arithmetic" DON'T matter when predicting a future outcome of an election.


    But from 2004, 2008 and 2010---it rather appears that they do!

    2004: http://election.princeton.edu/history-of-the-2004-race/

    2008: http://election.princeton.edu/2008/11/11/post-election-evaluation-part-2/

    2010: http://election.princeton.edu/2010/11/05/howd-we-do-2010-edition/

    But I don't think you actually grasp these things...
  • morleyq

    Posts: 175

    Nov 01, 2012 5:38 AM GMT
    TigerTim saidBut I don't think you actually grasp these things...

    Recently single for effectively the first time since the advent of the internet, I checked out a few dating sites and was floored at the number of people who thought that evolution and creationism deserved equal time. Or that it was worse to burn the flag than books. This not in the deep south or bible belt, but also among college kids.

    I'd go on about models and the scientific method, but I fear I would either be preaching to the choir or casting pearls before Republicans. icon_twisted.gif

    Edit: I will say one thing. CJAZ is probably thinking "past performance is no guarantee of future earnings." But to use this alone to dismiss a functional model is insane. Otherwise he should take my financial advice as seriously as he does some experts - who also might be wrong. Sure, there is a (small) chance come next week that CJAZ will get the last laugh. Yet it's much more likely that Alabama will beat whoever they are playing this weekend. (Ignore that I've chosen a bad week to use this analogy given that they play at LSU on Saturday).
  • CuriousJockAZ

    Posts: 19138

    Nov 01, 2012 5:42 AM GMT
    TigerTim said
    But from 2004, 2008 and 2010---it rather appears that they do!

    2004: http://election.princeton.edu/history-of-the-2004-race/

    2008: http://election.princeton.edu/2008/11/11/post-election-evaluation-part-2/

    2010: http://election.princeton.edu/2010/11/05/howd-we-do-2010-edition/

    But I don't think you actually grasp these things...


    Oh, I grasp them. What I don't think you grasp is that this is 2012, and you might think about tossing those stats from the past decade right out the window as they may not apply to THIS election. We'll know on Tuesday.
  • morleyq

    Posts: 175

    Nov 01, 2012 5:52 AM GMT
    CuriousJockAZ said
    TigerTim saidBut I don't think you actually grasp these things...

    Oh, I grasp them. What I don't think you grasp is that this is 2012, and you might think about tossing those stats from the past decade right out the window as they may not apply to THIS election. We'll know on Tuesday.

    Oh, no you don't.

    This is NOT about dated "stats".
    It IS about a MODEL for analyzing CURRENT data and using it as a basis for a prediction.

    This is how science works. The model wasn't devised to produce desired results. It may be the case that the model will fail, If so there'll be some head-scratching followed by revision of the MODEL.
  • CuriousJockAZ

    Posts: 19138

    Nov 01, 2012 6:39 AM GMT
    morleyq said

    This is how science works. The model wasn't devised to produce desired results. It may be the case that the model will fail, If so there'll be some head-scratching followed by revision of the MODEL.



    Ohhhh, I get it. Well, in THAT case, I hear the tick tock of a countdown to when the head scratching (okay EXPLODING) will begin and MODEL REVISION will ensue icon_wink.gif

  • Posted by a hidden member.
    Log in to view his profile

    Nov 01, 2012 10:43 AM GMT
    CuriousJockAZ said
    TigerTim said
    But from 2004, 2008 and 2010---it rather appears that they do!

    2004: http://election.princeton.edu/history-of-the-2004-race/

    2008: http://election.princeton.edu/2008/11/11/post-election-evaluation-part-2/

    2010: http://election.princeton.edu/2010/11/05/howd-we-do-2010-edition/

    But I don't think you actually grasp these things...


    Oh, I grasp them. What I don't think you grasp is that this is 2012, and you might think about tossing those stats from the past decade right out the window as they may not apply to THIS election. We'll know on Tuesday.


    The stats from past elections are not being used---it's the METHOD that's being used!

    You obviously don't get it!
  • morleyq

    Posts: 175

    Nov 01, 2012 4:28 PM GMT
    CuriousJockAZ said
    morleyq saidThis is how science works. The model wasn't devised to produce desired results. It may be the case that the model will fail, If so there'll be some head-scratching followed by revision of the MODEL.

    Ohhhh, I get it. Well, in THAT case, I hear the tick tock of a countdown to when the head scratching (okay EXPLODING) will begin and MODEL REVISION will ensue icon_wink.gif

    Sorry, but it is clear that you still don't get it. You are postulating wishful thinking - desired results - as science. Which is a complete reversal of the scientific method.

    I'm open to hearing criticism of the model, but absent that you sound like someone 100+ years ago who said that it is impossible to build an airplane and that those who do will crash and "EXPLODE" and have to revise their model (to not flying).

    If you wish to learn more about what you are (currently not) criticizing, see:

    http://election.princeton.edu/romentum-rove-1nov2012.php

    "National polls show a tie, while state polls show a decisive Obama advantage. Here I suggest that the difference may arise from the fact that the same systematic pollster errors can have different effects depending on whether they occur in national vs. state surveys. Based on past elections, national poll aggregates differ from election results by as much as 2.5%. During the same period, state-poll aggregation has been considerably more accurate. Even if state polls have the same accuracy as national polls, races at that level are usually decided by larger margins, leaving room for aggregation to remove the effect of the error. I suggest that the Meta-Analysis of state polls provides a more accurate poll-based prediction of next Tuesday's outcome than national polls."
  • CuriousJockAZ

    Posts: 19138

    Nov 01, 2012 4:48 PM GMT
    morleyq said

    I'm open to hearing criticism of the model, but absent that you sound like someone 100+ years ago who said that it is impossible to build an airplane and that those who do will crash and "EXPLODE" and have to revise their model (to not flying).



    The election results will be all the criticism needed of your "scientific model". icon_wink.gif