home

WaPo/ABC Poll: Obama By 11

One thing is for sure, if Obama does not win this election, the WaPo/ABC poll will have a lot of egg on its face. Today they have Obama leading McCain by 54-43 among likely voters. Earlier today, the WaPo pollster Jon Cohen wrote:

[I]f the trustworthy polls continue showing Obama ahead and McCain wins, it would be a monumental failure for political scientists, reporters and pollsters alike -- an indictment worse than New Hampshire, worse even than 1948. I think the quality pollsters have done a good, professional job this year. I don't think we'll get bitten. Even so, I'll be a little worried until it's all over. I'm not sure what kind of night I'll have on Tuesday. But I'm sure I'm going to have a nervous one on Monday.

By Big Tent Democrat, speaking for me only

< ATTN CA Voters: Vote No On Prop. 9 | McCain and Palin Returning to Colorado - Separately >
  • The Online Magazine with Liberal coverage of crime-related political and injustice news

  • Contribute To TalkLeft


  • Display: Sort:
    What still concerns me (5.00 / 1) (#18)
    by cal1942 on Sun Nov 02, 2008 at 05:46:20 PM EST
    is that the national polls reflect the popular vote only.  On Tuesday 51 elections will be held, not one.

    So I look at some states likely to stay blue or red as they did in '00 and '04 and I see this:

    CA Gore won by 11.8% Obama leads by 24.4%
    NY Gore won by 25.0% Obama leads by 29.7%
    IL Gore won by 12.0% Obama leads by 22.2%

    TX Bush won by 21.3% McCain leads by 13.0%
    AZ Bush won by 6.3%  McCain leads by 3.5%

    It doesn't really matter what percentage wins a given state.  A win is a win and with it all of the EVs. So what if McCain wins TX by only 13 points it's still the same 34 EVs as Bush won in '04.  So what if Obama wins CA by 24.4 points it's still the same EVs that Gore (54) and Kerry (55)won.

    My point is that if Obama wins blue states by large margins it matters not at all and if McCain wins red states by smaller margins than Bush it doesn't matter at all.  The EV result is the same.

    The only way that Obama wins this election is to pick off enough red states to make up for the EV short fall of the states won by Gore and Kerry. The Gore/Kerry states only add up to 264 EVs and I'm worried that the national polls only reflect declining margins in red states for the GOP and increasing margins in the blue states for Democrats.

    That may be why the Obama's spent so much time in Colorado this past week.  Nine elctoral votes.  Enough to put him over the top. He may have simply felt that was his best bet among the red states.

    Ahh, the "wasted vote" (none / 0) (#21)
    by andgarden on Sun Nov 02, 2008 at 05:54:52 PM EST
    That's the principle behind gerrymandering, and if you believe the serious pundits, it's why the Republicans were supposed to hold the House in 2006. Indeed, in Ohio, they did keep a majority of the seats, even though they lost the statewide House vote. But nationally, Democrats won seats more-or-less proportionally to the generic ballot. Why? Gerrymanders are naturally self limiting, and self defeating. Want more seats? You've got to have fewer of your voters in each seat.

    And so it generally goes for the Electoral Electoral college. If this is a 7 point race, then there's nothing to worry about.

    If it's a 2 point race, then worry.

    Parent

    Another reason (5.00 / 1) (#23)
    by Lora on Sun Nov 02, 2008 at 06:05:38 PM EST
    [
    I]f the trustworthy polls continue showing Obama ahead and McCain wins, it would be a monumental failure for political scientists, reporters and pollsters alike --

    Or a monumental failure for our election system.  Somehow, people will want to believe that all the polls are wrong, rather than they are right and our election system is wrong.

    Occam's razor.

    Thanks for the link (none / 0) (#1)
    by andgarden on Sun Nov 02, 2008 at 04:27:54 PM EST
    Again, if this poll is right, then we're going to have an early night. But I share Cohen's concern about nonresponse bias. What if the Republicans just aren't answering the polls?

    We'll know Tuesday night.

    Bias? (none / 0) (#6)
    by TheRealFrank on Sun Nov 02, 2008 at 04:59:04 PM EST
    If the LV model uses party affiliation for weighting, then any such bias should not be visible in the final result.


    Parent
    Well, Dems suddenly have a 6-10pt (none / 0) (#9)
    by andgarden on Sun Nov 02, 2008 at 05:01:32 PM EST
    party ID advantage this year. Some people say that's impossible.

    Parent
    Hm. (none / 0) (#12)
    by TheRealFrank on Sun Nov 02, 2008 at 05:07:13 PM EST
    I don't know.. I think most pollsters use quite a lot of data to come to those numbers. I think Rasmussen mention some 40,000 interviews.

    It is definitely true that newly registered voters have favored the Democrats this year.

    I don't think the "shy Republicans" theory has ever been seen in actual polling. It has definitely been seen in exit polling, Rasmussen had a good post about exit polls on his site the other day.


    Parent

    Agreed (none / 0) (#13)
    by andgarden on Sun Nov 02, 2008 at 05:09:46 PM EST
    with the caveats that 1) registration does not not equal party ID, and should never be used interchangeably; indeed, you can poll nationally based on registration, as many states do not register by party, and 2) just because it hasn't happened before doesn't mean that it can't happen now.

    I am trusting the polls--cautiously.

    Parent

    I suspect if they don't answer the phone, (none / 0) (#20)
    by steviez314 on Sun Nov 02, 2008 at 05:49:10 PM EST
    they're probably not voting.

    After seeing all the vile hate that some people don't mind displaying on some videos of McCain/Palin rallies, I can't imagine why they'd feel too bad about telling a pollster that they're voting for McCain not Obama.

    Jeez, if they're not embarassed to call Obama an Arab terrorist on TV, why would they care what a pollster thought?

    Parent

    Answering the phone has nothing to do (none / 0) (#22)
    by nycstray on Sun Nov 02, 2008 at 05:55:51 PM EST
    with polls/voting. Answering the questions honestly might though.

    I screen my calls and don't pick up unknown numbers.

    Parent

    New polling question!!! (none / 0) (#24)
    by steviez314 on Sun Nov 02, 2008 at 06:21:04 PM EST
    Do you have Caller-ID?

    I'm curious to see the D-R-I and age breakdown.

    Parent

    My age breakdown would (none / 0) (#40)
    by nycstray on Mon Nov 03, 2008 at 04:33:39 AM EST
    disappoint you  ;)

    Parent
    Disagree. (none / 0) (#2)
    by LarryInNYC on Sun Nov 02, 2008 at 04:48:40 PM EST
    it would be a monumental failure for political scientists, reporters and pollsters alike -- an indictment worse than New Hampshire, worse even than 1948.

    Unless it's demonstrated that pollsters somehow changed their methodology in a meaningful way this year, the indictment would not be of them but of the American electorate.

    All polls -- without exception -- agree that Obama will win.  If there's something so fundamentally different about this election (that is, about having a black candidate) that makes people lie to pollsters to the extent observed in this polling the pollsters (the messengers) would not be to blame -- the liars would.

    I don't think that will happen.  I think the anti-black vote is already baked into the the fact that this will be a ten point election, and not the twenty point blowout that the incredible disgust that the general population feels towards the current Administration coupled with the incompetence and instability of the Republican ticket should be producing.

    The people who answer the polls (none / 0) (#4)
    by andgarden on Sun Nov 02, 2008 at 04:52:29 PM EST
    may not by lying. Some people just might not be answering the polls. Nonresponse would be the best explanation.

    Parent
    Whatever the reason. . . (none / 0) (#7)
    by LarryInNYC on Sun Nov 02, 2008 at 04:59:04 PM EST
    my point is that if the polling landscape has shifted suddenly because people respond differently when there's a black candidate involved, it isn't reasonable to blame the subsequent inaccuracies -- at least in the first episode of this kind of polling discontinuity -- on the pollsters.

    Parent
    I don't think any sane person (none / 0) (#10)
    by andgarden on Sun Nov 02, 2008 at 05:02:55 PM EST
    is going to blame a loss on a pollster. Especially not the collective wisdom of the pollsters.

    Parent
    It's the failure of the polls. . . (none / 0) (#26)
    by LarryInNYC on Sun Nov 02, 2008 at 06:39:39 PM EST
    (if that happens) about which we're trying to apportion blame.  If the polling game has changed, I wouldn't blame the pollsters for being off simply because they didn't identify, in advance, the new polling conditions.

    Parent
    Indeed (none / 0) (#8)
    by Democratic Cat on Sun Nov 02, 2008 at 05:01:24 PM EST
    Obama may win by less than the polls indicate because of a combination of the effects of voter suppression and non-representativeness of poll samples. It wouldn't surprise me if Republicans are less inclined to respond to polls this year because they are not enthused about the election and they expect to lose.  But there doesn't seem to be any real chance that Obama would actually lose, not with the spreads we are seeing in polls. Even if the sample is a little bit unrepresentative, that shouldn't be enough to overcome the gap.

    Parent
    If the polls are off. . . (none / 0) (#27)
    by LarryInNYC on Sun Nov 02, 2008 at 06:41:39 PM EST
    due to previously unknown conditions, how can we say "okay, they're off, but not by more than seven percent".  If the polls aren't functioning correctly this year, we could see a McCain win (or an Obama blowout) -- if they're not functioning, then whatever they say now is meaningless.

    Parent
    It's judgment (none / 0) (#36)
    by Democratic Cat on Sun Nov 02, 2008 at 09:34:45 PM EST
    Although we don't know everything,that doesn't mean that we know nothing.

    The polling gap is large. The problem of non-respondents and whether they are different from respondents is not a brand new problem. Pollsters and statisticians confront selection bias all the time. It seems unlikely that the problem is severe enough to flip the election. Do I know this is true for certain? No, of course not. Statisticians make judgments all the time. That's mine.

    Parent

    Non-response (none / 0) (#37)
    by andrys on Sun Nov 02, 2008 at 10:48:10 PM EST
    I don't see how Obama can lose, considering he is currently expanding his lead after some tightening for a couple of days.

      However, this non-response you speak of:  I have no idea whether or not this is true, but someone wrote that about 80% of those queried opt out of being part of a polling program.

      That seems extremely high to ME.   Some people are 'tight lipped' and our voting is considered by many a private matter.    This race may be somewhat tighter (though the difference can't be nearly as large as the lead) because, rather than BEING racist, some are worried to say they won't vote for Obama because they'll be accused of being racist, since Obama supporters can see no other reason not to vote for him.

      In the CNN focus groups, when they ask for votes, Obama voters hold their hands high -- they're happy to say they're voting for him.  But when they ask who's switched to McCain, they're difficult to count because their arms are only half-way up -- you can barely even see the hands, which rarely go higher than the shoulder.

      That may also be the enthusiasm gap.  There's no question that McCain doesn't attract enthusiasm or fervor.  

    Parent

    I Agree (none / 0) (#5)
    by zvs888 on Sun Nov 02, 2008 at 04:53:51 PM EST
    I don't see how the pollsters would be blamed.

    It's not like the Dems won't pick up 7 Senate seats and 25-30 House seats even if Obama loses.

    It all just adds up to the fact that lower income whites who generally participate a lot less in these polls went to McCain in greater numbers than expected and showed up in greater numbers than expected.

    That can't really be blamed on anyone, so much as it's just an uncaptureable event.

    Parent

    If Obama loses (none / 0) (#11)
    by Big Tent Democrat on Sun Nov 02, 2008 at 05:03:08 PM EST
    I will be blaming him for not picking Clinton as his VP.

    Parent
    You know that some other people (none / 0) (#15)
    by andgarden on Sun Nov 02, 2008 at 05:10:47 PM EST
    would find a way to blame her. . .

    Good news is that we have a 99% chance of not having to go through that nightmare.

    Parent

    "If..." (none / 0) (#19)
    by Lil on Sun Nov 02, 2008 at 05:48:52 PM EST
    You're scaring me.

    Parent
    That is an unknowable thing. (none / 0) (#28)
    by LarryInNYC on Sun Nov 02, 2008 at 06:45:21 PM EST
    Clinton might have helped in him some states, but I think the media narrative would have been completely different.

    First off, McCain probably wouldn't have sabotaged his own campaign by selecting Palin.  Secondly, even if he had, a great deal of the energy that's gone into reporting the internecine McCain / Palin campaign struggles would have been directed at the Democrats.  Can you imagine the effect of three months of Olbermann and Mathews analyzing every aspect of every single campaign appearance to attempt to determine every (imagined) way in which Clinton was trying to upstage Obama?

    There's no way of knowing how things would have worked out with Clinton on the ticket.

    Parent

    true (1.00 / 1) (#30)
    by progrocks on Sun Nov 02, 2008 at 06:48:02 PM EST
    but Armando loves Palin, so that is hardly sabotage.

    Parent
    I very much doubt. . . (none / 0) (#31)
    by LarryInNYC on Sun Nov 02, 2008 at 06:50:07 PM EST
    BTD has a kind word to say about Palin.  He is simply worried about the effect having such a doofus on a national ticket will have on female candidates for high office in the future.

    Parent
    In my one canvassing excursion (none / 0) (#35)
    by Coral on Sun Nov 02, 2008 at 09:28:08 PM EST
    I ran into people in NH who were undecided because they had been for Hillary. They'll probably vote Obama, but...boy, it would have been so much easier if she were on the ticket.


    Parent
    It would be a sign (none / 0) (#33)
    by sallywally on Sun Nov 02, 2008 at 08:52:52 PM EST
    that the GOP's voter suppression and other dirty tricks and the Diebold voting machines have done their nefarious work.....imho.

    Parent
    That pollster is concern trolling himself. (none / 0) (#3)
    by steviez314 on Sun Nov 02, 2008 at 04:49:04 PM EST
    48 short/long hours to go.

    Any poll on early voting (none / 0) (#14)
    by Saul on Sun Nov 02, 2008 at 05:10:08 PM EST
    I'm curious on how many of the  young voters 18 to age 25 that got all fired up during the primary and went out to register voted in early voting.  That's what Obama was counting on those that usually do not vote would be voting.

    CBS yesterday... (none / 0) (#16)
    by Edger on Sun Nov 02, 2008 at 05:19:31 PM EST
       (CBS) With just three days left until Election Day, a new CBS News poll finds that the Democratic presidential ticket of Barack Obama and Joe Biden leads its Republican counterpart by 13 points among likely voters, 54 percent to 41 percent. That margin reflects an increase of two points in the Obama-Biden ticket's lead from a CBS News/New York Times poll released Thursday.

        About one in five voters say they have already cast their vote, either in person or through the mail, and these early voters prefer the Democratic ticket by an even greater margin. Obama leads among early voters 57 percent to 38 percent, a nineteen point advantage.



    Parent
    I read that in Colorado (none / 0) (#38)
    by andrys on Sun Nov 02, 2008 at 10:55:59 PM EST
     - as of yesterday - about 50% had voted already (!) and that with that vote, exit polling was showing Obama about 20 points ahead of McCain, an almost impossible situation for McCain, so that looks good for Obama despite the tightening of polls there.

      Another factor is that Obama has been pushing early voting and thank god he did since they're already having long lines outside the places where you can do that in person.  So he would tend to be leading, but not by 20 points, so that's a sign.

    Parent

    I would like to see Obama win by (none / 0) (#17)
    by ThatOneVoter on Sun Nov 02, 2008 at 05:21:56 PM EST
    a margin that surprises the pundits, and makes it impossible to deny the Democrats' mandate.

    It is always possible. . . (5.00 / 1) (#29)
    by LarryInNYC on Sun Nov 02, 2008 at 06:47:12 PM EST
    to deny a "mandate" -- a term basically invented for reasons of political posturing.  And since I doubt many people actually agree on what "mandate" the Democrats have actually been seeking during this campaign, I don't know how they'd actually act on it.

    Parent
    And then there's Bushian analysis (none / 0) (#39)
    by andrys on Sun Nov 02, 2008 at 11:00:30 PM EST
    ... When Bush lost the popular vote in yr 2000, he declared he had a clear mandate...

      Re electoral college voting, he'd won Florida by only something like 537 votes (certified, though the Fed system counting had the difference a bit larger), and Bush won the presidency when Florida went for him with the help of the Supreme Court.  It's so alarming what can make so much difference to all of us.

    Parent

    CBS today. (none / 0) (#25)
    by TheRealFrank on Sun Nov 02, 2008 at 06:23:31 PM EST
    54-41

    Unchanged from yesterday.


    Combine That (none / 0) (#32)
    by zvs888 on Sun Nov 02, 2008 at 06:59:15 PM EST
    With the just announced USAToday/Gallup with Obama with his biggest lead: 11 among LVs.

    And now we have all 3 of the big media polls with Obama way up.

    This should be headed towards landslide territory.

    Parent

    It's almost over (none / 0) (#34)
    by Militarytracy on Sun Nov 02, 2008 at 09:16:00 PM EST
    My winter vacation in hell.  Tonight I find myself revisiting my blogging in the leftzone past as an active duty serving soldiers wife. After the initial Crawford protest I became easy pickins for many left extremists and blogs and bloggers.  I was the thing that anchored baby killers and illegal wars.  It was okay to say anything to me except on Talkleft.  I wish that I hadn't been exposed to any of that sort of thing.  It is very strange to soon be the heroish sort of family that those who long to end genocide during an Obama presidency will pat on the back.................and eight years ago neoconish republicans, or isolationist republicans, or end of the world evangelicals were trying to pat me on the back for my efforts.  I ran from such pats and my duty to family was unavailable for such BS.  My spouse serves the military of this "experiment" come hell or high water - forever believing that you guys......the majority........will be steering his course.  Don't forget to vote so he'll know what the hell he's supposed to be doing for you and what's expected of him.  He has more glands than I do, not numerically, but his aura contains more glands and more faith in all of us (you and me and all these other characters that are considered Americans) than mine ever will.  In every day life he is often labeled a pessimist and I'm called an optimist.  Sort of funny though.  I'm not laying my life on the line for any of ya or what you think and believe!