home

The Polls - 10/24

By Big Tent Democrat, Section Elections 2008
Posted on Fri Oct 24, 2008 at 08:22:00 AM EST
Tags: (all tags)
Share This: Digg! StumbleUpon del.icio.us reddit reddit

In polls released in the past 24 hours, DKos/R2000 has Obama leading by 12, 52-40. WaPo tracker has Obama by 11, 54-43. CBS/NYTimes poll has Obama by 13, 52-39.

The best polls for McCain in the past 24 hours were IBD/TIPP, which has Obama by 1, 45-44; Battleground, which has Obama by 4, 49-45; Hotline which has Obama by 5, 48-43; Gallup Expanded, which has Obama by 6, 51-45. There has not been a poll in the past month that I can remember that has had McCain ahead.

By Big Tent Democrat, speaking for me only

< The Narrative | Friday Morning Open Thread >

  • The Online Magazine with Liberal coverage of crime-related political and injustice news

  • Contribute To TalkLeft


    • Display: Sort:
    The Polls - 10/24 | 15 comments (15 topical)

    Readers Must Login or Create an Account to Comment

    IBD/TIPP (5.00 / 2) (#1)
    by TruthMatters on Fri Oct 24, 2008 at 07:55:09 AM EST
    had McCain winning the 18-25 by a 72-22 margin

    so lets just say its safe to say that poll was wrong.

    Really? (none / 0) (#2)
    by Big Tent Democrat on Fri Oct 24, 2008 at 08:12:49 AM EST
    That's hilarious.

    Parent
    yeah (none / 0) (#4)
    by TruthMatters on Fri Oct 24, 2008 at 08:41:38 AM EST
    Nate silver at 538.com, had a post about it.

    so totally throws the poll out the window.

    link to nates post here.

    Parent

    At least you know when they put out (none / 0) (#7)
    by Exeter on Fri Oct 24, 2008 at 08:58:26 AM EST
    a totally ridiculous result, you know that it is a bad batch of polling and that they are being honest in reporting the results.

    Parent
    Bringing up history again... (none / 0) (#3)
    by mike in dc on Fri Oct 24, 2008 at 08:36:36 AM EST
    Clinton 1992, roughly 6 point margin
    Reagan 1980, 9 point margin
    Eisenhower 1952, 11 point margin
    FDR 1932, 18 point margin

    I think a Reaganesque margin is fairly likely for Obama, an Ike-sized margin is a slight possibility, and if there's any kind of strong close combined with a McCain collapse in the final week (and it looks like the market's going to crater today, which can't be a good thing for anybody), then we could watch the results on November 4th, wide-eyed, as we find ourselves in FDR territory.  
    But even a Clinton-sized win would mean a major realignment, at least for the next 8-10 years.  An FDR sized win would mean realignment for a whole generation.

    list (none / 0) (#10)
    by connecticut yankee on Fri Oct 24, 2008 at 10:38:43 AM EST
    I saw a similar list a few days ago and added a few.

    Carter beat Ford by 2.1
    Reagan beat Carter by 9
    Reagan beat Mondale by 18
    Bush I beat Dukakis by 7.7
    Clinton beat Bush I by 5.6  
    Clinton beat Dole by 8.5
    Bush II beat* Gore by  -.5
    Bush II beat Kerry by 2.5

    I'd say we are heading toward Clinton-Bush but might get a Bush-Dukakis.

    An interesting factoid I just discovered is that third party candidates took 7% in 1980.

    Parent

    John Anderson... (none / 0) (#12)
    by mike in dc on Fri Oct 24, 2008 at 12:08:10 PM EST
    ...took about 7%, yes, in 1980.  He was a moderate Republican who, if I remember correctly, finished third in the GOP primary races and decided to bolt the party and run as an independent.  

    The reason why I picked those examples was that they were as close as I could find to match the situation in this election--unpopular incumbent party, bad economy, strong challenging candidate.

    1932, 1952, 1980, and 1992 all fit that criteria (though I don't think the economy was that bad in 1952).  In 1992 there was a very strong third party candidate, as there was in 1980.  In 1952 and 1932 there was not.  The actual closest analogy is 1952, I think, because there were no incumbents running and the incumbent president was very unpopular.  Ike won big, about 55% of the vote.  
    Obviously every Dem dreams of another FDR-sized victory to sweep into power and vanquish the GOP for a generation, though.:)

    Parent

    It's all about the money (none / 0) (#5)
    by Yotin on Fri Oct 24, 2008 at 08:45:18 AM EST
    In an environment toxic for McCain and company, what would one attribute a slim lead by Obama. I have my doubts his lead is due to any of his positive attributes. IMO, it's all about money overcoming his negative attributes.

    I agree w/homonidviews: Obama has 100% (none / 0) (#6)
    by Exeter on Fri Oct 24, 2008 at 08:56:30 AM EST
    chance of winning. Link Only question now is get a super mandate by winning in places like Georgia and Montana

    Pollsters must be wondering (none / 0) (#8)
    by Cream City on Fri Oct 24, 2008 at 09:13:00 AM EST
    about models based on all those new voter registrations, reported by ACORN at 1.3 million.

    An NYT report yesterday puts the total at 450,000 at most -- because ACORN and its affiliated projects were not reporting just the forms they did but all the forms turned in around the country.  And in both cases, many were not new voters but just previous voters changing addresses, names, etc.

    The NYT also reports that an estimated 30 percent of the actual new registrations are in error -- reregistrations unneeded, for example, as well as intentionally wrong (aka fraud).  

    That comes down to about 300,000 actual new voters (or, of course, newly registered voters, as it won't be known until Nov. 4 whether they vote.)  So pollsters who adjusted models for a million or more new voters may be in error, too.

    Then again, if the turnout is higher among all groups, certain groups, etc., as predicted, who knows whether pollsters have guessed correctly on that, too?  One pollster pointed out that the AA vote may have maxed in the primaries, so expecting a much larger turnout in November based on previous primary-to-general election numbers would be another problem.  

    A large grain of salt is required when looking at polls, with all the guessing that the pollsters had to do.  This also probably will mean major adjustments made at the last minute a la those exit polls in 2004 that had to be adapted a lot on election day, as actual turnout was seen.

    Gallup already investigated this phenomenon (none / 0) (#11)
    by BrianJ on Fri Oct 24, 2008 at 11:59:13 AM EST
    And, to be blunt, "all those new voter registrations" really aren't going to add up to much.  They found that the proportion of new voters is no greater than in 2004, and while Obama's got a big lead among such voters, so did Kerry.  Obama will win (if he does) by winning over the geezers-  or more precisely, because McCain and the Republicans alienated them.

    http://www.gallup.com/poll/111331/No-Increase-Proportion-First-Time-Voters.aspx

    Parent

    Thanks for the link (none / 0) (#13)
    by Cream City on Fri Oct 24, 2008 at 01:12:33 PM EST
    as that helps to explain Gallup's interesting decision to do multiple poll designs at once.  So the new, non-traditional ones are based on better (perhaps) projections of increased turnout not tied to greatly increased voter registrations.

    I wish it were more clear, though, how many other pollsters factored the earlier voter-registration reports into their models.  I recall reading that some definitely were doing so, seeing those reports as indicative.  So I still wonder whether that could help to explain the widely differing numbers that we see in the polls now (among other factors).  

    Parent

    well (none / 0) (#14)
    by connecticut yankee on Fri Oct 24, 2008 at 01:18:05 PM EST
    Enthusiasm for the candidate might also figure in turn-out projections.  Democrats have a 2-1 advantage there iirc.  It's certainly altered the early voting trends. Previously republicans won the early vote, this time its heavily Democratic.

    Parent
    Yes, that is a factor in projecting turnout (none / 0) (#15)
    by Cream City on Fri Oct 24, 2008 at 01:32:15 PM EST
    of course.  (Btw, I recall some polls showing that 2-1 advantage tightening, but perhaps it has widened again.)

    Parent
    Strategic Vision (none / 0) (#9)
    by CoralGables on Fri Oct 24, 2008 at 10:06:11 AM EST
    Anyone know a way in the backdoor to the cross tabs on Strategic Vision polling. As each of their state by state polls done this morning appear to be outliers, digging into the core of them could be fun.

    Then again, Strategic Vision is known as a Republican outfit so perhaps they are just trying to grab some free advertising for the GOP in an attempt to continue the "McCain is gaining" theme.

    The Polls - 10/24 | 15 comments (15 topical)

    Readers Must Login or Create an Account to Comment