1. Welcome to CowboysZone!  Join us!  Come on!  You know you want to!

Most Accurate Poll

Discussion in 'Political Zone' started by irvin88, Oct 16, 2008.

  1. irvin88

    irvin88 Active Member

    1,668 Messages
    0 Likes Received
    IBD/TIPP Tracking Poll
    Posted: Wednesday, October 15, 2008


    Obama maintained a 3-point lead over McCain going into their third and final debate. Among the early trends in our tracking poll is an increase in undecideds among respondents 65 and older. There are now almost as many sitting on the fence in this group as there are among independents, where the race is tight. McCain could also be doing better among conservatives and married women.


    About IBD/TIPP: An analysis of Final Certified Results for the 2004 election showed IBD's polling partner, TIPP, was the most accurate pollster of the campaign season.
  2. cowboys#1

    cowboys#1 Finish!

    2,408 Messages
    15 Likes Received
    i dont think any polls are accurate. we all have to wait until election night
  3. irvin88

    irvin88 Active Member

    1,668 Messages
    0 Likes Received
    I agree. Mr. Bradley is going to have a say as well.
  4. masomenos

    masomenos Less is more

    5,971 Messages
    0 Likes Received
    The only polls that matter are the state polls.
  5. Doomsday101

    Doomsday101 Well-Known Member

    75,324 Messages
    1,817 Likes Received
    Only thing that matter is the election returns on election day. Asking someone who are you going to vote for tends to be very different from who actually went out and cast their vote. They don't count the I was going to vote but something else came up ballot. :laugh2:
  6. NinePointOh

    NinePointOh Active Member

    1,536 Messages
    18 Likes Received
  7. irvin88

    irvin88 Active Member

    1,668 Messages
    0 Likes Received
    Just explain New Hampshire in Dem Primary this year.

    Hussein was up by 12 the Friday before, and lost by 3 to Hillary on that Tuesday.

    15 point swing in 3 days ??:laugh2:
  8. masomenos

    masomenos Less is more

    5,971 Messages
    0 Likes Received
    lol that's very true, but as far as pre-election polling goes, the only way to look at the possible outcome with any accuracy is state by state projections.
  9. masomenos

    masomenos Less is more

    5,971 Messages
    0 Likes Received
    Hasn't this been covered? Obama did better, on average, in the primaries then he did in in polling. One state or a handful of states doesn't express a trend.
  10. Doomsday101

    Doomsday101 Well-Known Member

    75,324 Messages
    1,817 Likes Received
    The last 2 election did not prove that at all, even the exit polling on election day was off. Exit polling had Kerry winning some key states but the election returns showed a different reality. I think Polls give a sense of where things are but polls tend to change and sometimes quickly.
  11. NinePointOh

    NinePointOh Active Member

    1,536 Messages
    18 Likes Received
    I take it you didn't bother to read the linked article. For the sake of discussion, you can keep New Hampshire and its 4 electoral votes. Now can you explain the rest of the country?


    [IMG]
  12. Beast_from_East

    Beast_from_East Well-Known Member

    14,290 Messages
    2,028 Likes Received
    Actually this has been debunked as proof of a Bradley Effect.

    The problem was not with Obama's numbers. In fact, he got about the same percentage of the vote as the polls predicted. The problem was that the pollsters really screwed up on Hillary.

    The pollsters vastly underestimated her turnout and as a result, her poll numbers were way off.

    I dont have a link, but they were talking about this on TV and said the study was done by Princeton if I remember correctly.
  13. masomenos

    masomenos Less is more

    5,971 Messages
    0 Likes Received
    It really depends on how the projections are done. This site had the '04 election going to Kerry http://www.geocities.com/samboni1342/state_polls.htm. When you take the MoE for each projection into account the polling is really pretty accurate when compared to the actual results http://en.wikipedia.org/wiki/U.S._presidential_election,_2004#Results_by_state.
    The problem with sites like this are that they called states for one candidate or the other even if the percentage from polling was within the margin of error. An example of this is Iowa, which polling suggested was for Kerry by 3.2%. The MoE in that poll however was 4.1%. On election day Bush won Iowa by 10,000 votes, giving him just under a 1% victory. States with favoring candidates by percentages that are within the MoE should remain toss up states.

    Only a few states had real results that were outside the margin of error when switching from one candidate to the other. Another thing to take into consideration is that these projections were based off of outdated data.
  14. jimnabby

    jimnabby Well-Known Member

    3,583 Messages
    426 Likes Received
    Obama's numbers didn't change between the polls and the voting.

    The NH primary was only a week after IA, and there was a ton of news between the two. Obama's surprise win in IA (where he outperformed the polls dramatically), Biden (and others, I think) dropping out of the race, Hillary crying. Things were moving really fast and the polls couldn't keep up with the flux.

    And again, the polls measured Obama's support accurately. There's no support for a Bradley effect there.
  15. NinePointOh

    NinePointOh Active Member

    1,536 Messages
    18 Likes Received
    The exit polls were off. The tracking polls were not.

    The methodology is vastly different between the two.
  16. Doomsday101

    Doomsday101 Well-Known Member

    75,324 Messages
    1,817 Likes Received
    All I'm saying is I don't trust polls, I trust results. When McCain was even in the polls I did not put much into that. When Obama was taking a double digit lead I did not put much into that and now that McCain could be narrowing the gap I don't put much into that. People saying one thing and doing another is a common event. Those who are hard core dems are going to vote for Obama that is a given, the hard core republicans will vote for McCain. The only question I have is who can get the votes out on election day. Telling a pollster who you are going to vote for and actually getting out going to the voting booth and casting a vote are 2 different things which is why I don't put much into peoples intentions.

Share This Page