The Battleground State Version of the David Axelrod Turnout Model Take-Down

DISCLAIMER: Blogging may be light this afternoon due to some issues away from the blog.  I’m still apparently 2-3 days away from getting electricity and heading home so adjustments will crimp into blogging.  I’m trying to get out the Clark County early vote post which was a good day for Democrats but not nearly the big day they hoped/needed. I’m not even monitoring my usual source for scoops and posts so I can get the backed-up posts out.  So big news like Romney heading to Harrisburg, Pennsylvania (tonight!) will have to wait. My ability to monitor comments is limited.  Play nice.  To newcomers on both sides: no name calling, hair pulling, nonsensical comments or you’re outta here as soon as I notice. Disagree all you want but offer sound reasons not just your blind belief. End Disclaimer.

Today’s must read:

Click on this link and read this whole piece by Reid Wilson in the National Journal.  It addresses my exact point in the David Axelrod Turnout Model take-down. My post was on the national numbers but the same story applies at the state level. This column talking to Rob Jesmer walks you through the exact same arguments state-by-state.  A must read:

A few days ago, I sat down with Rob Jesmer, the executive director of the National Republican Senatorial Committee. Jesmer is usually tight-fisted about his polling; he doesn’t share it with members of the media when the numbers are good for his candidates, which avoids the inevitably uncomfortable dilemma when the numbers are bad for his candidates. But he wanted to open his books, if only for a peek, to demonstrate a phenomenon happening across the political spectrum these days: His polls look nothing like polls Democrats are conducting.

It’s a constant refrain from both sides these days. The two parties, the outside groups that are playing such a big role this year, and even some candidates themselves are so dubious about their own numbers that they are employing two pollsters for one race, using one to double-check the other. What flummoxes them even more is that their own party’s pollsters are getting similar results, while the other side is offering a completely different take.

Republicans say their party is a victim of media bias — but not in the standard Lamestream Media sort of way. Pollsters on both sides try to persuade public surveyors that their voter turnout models are more accurate reflections of what’s going to happen on Election Day. This year, GOP pollsters and strategists believe those nonpartisan pollsters are adopting Democratic turnout models en masse.

Regardless of the cause, strategists on both sides acknowledge the difference in their internal polling. Republicans believe Democrats are counting far too much on low-propensity voters and a booming minority turnout that isn’t going to materialize on Election Day. Democrats believe Republicans are hopelessly reliant on an electorate that looks far more like their party than the nation as a whole. The day after Election Day, somebody’s pollsters are going to be proven seriously wrong.

Deep down, both parties secretly worry it’s their side that is missing the boat.


  1. petep
    Posted November 1, 2012 at 11:47 am | Permalink | Reply

    I was just explaining this point last night to my apolitical wife. There are two understandings of this election and I think Gallup is right and Silver is wrong. But I could be wrong…. I just don’t think so.

    • damien
      Posted November 1, 2012 at 11:53 am | Permalink | Reply

      all we have to do is look at how each side acts…romney acts like he is winning while obama acts like he is losing….or he just decided 5 days before the election to change tactics for the heck of it

    • No Tribe
      Posted November 1, 2012 at 5:50 pm | Permalink | Reply

      There’s actually the real math world that looks at these numbers in the context of real people voting, and statistical world that envisions political polls, including obviously partisan ones with an agenda, lend more certainty than madness.

      Nate Silver concocted a formula that says he’s 75% sure that the he can trust the state polls he chooses despite their multiple disparity… just because.

      I’m about 95% sure that Silver is wrong. I like the odds.

      • TheTorch
        Posted November 1, 2012 at 5:56 pm | Permalink

        Silver 95% wrong sounds about right to me! Also someone posted some information about virginia a few threads back – very bad for Obama, any democrat supporter reading that should realise what is about to happen. The enthusiasm gap is real and growing.

  2. Dan
    Posted November 1, 2012 at 11:52 am | Permalink | Reply

    It seems then that the range is a narrow Obama win to a significant Romney win.

  3. NS
    Posted November 1, 2012 at 11:56 am | Permalink | Reply

    Ras Iowa : R 49 : O 48

  4. Dabrisha
    Posted November 1, 2012 at 11:56 am | Permalink | Reply

    Keith…can you post this. Good for a laugh!

  5. Ranger375
    Posted November 1, 2012 at 12:05 pm | Permalink | Reply

    Like the articles are saying one side or the other in polling is wrong. So there will be shock and great disappointment on one side. There will certainly be some ‘splainin’ to do.

  6. bsmith
    Posted November 1, 2012 at 12:08 pm | Permalink | Reply

    We already can see who is going to have to do some explaining after election day. The Las Vegas Review Journal shows poll today with Obama up 4% in Nevada with a 10% advantage Democrats over Republicans. This is at a point when almost half the people have voted and Democrats have a 6% advantage. Election day should cut into advantage. At best it would be 6% advantage when all is said and done, but we are still being shown 10% advantage polls as if that is turnout model. This is happening in other states as well.

  7. Glen
    Posted November 1, 2012 at 12:08 pm | Permalink | Reply

    New vote totals are in from Colorado. Republicans have increased their votes casted advantage again today. Republicans have now cast 36,120 more votes than the dems (up from +27,261 two days ago). Obama needs to be well ahead in the early vote in this state to have a chance according to the RNC (via Michael Barone’s article this morning).

    • Keith W
      Posted November 1, 2012 at 12:16 pm | Permalink | Reply

      Have any idea what the dem advantage in early voting was in 2008?

      • Glen
        Posted November 1, 2012 at 12:36 pm | Permalink

        It would appear that the dems won the early voting in 2008 by about 30,677 if this site is correct. That would mean that we are 66,797 votes ahead of the last presidential cycle in CO.

    • Tom
      Posted November 1, 2012 at 12:47 pm | Permalink | Reply

      Looks great for the GOP. One has to remember that Romney is winning the Independents quite handily. Couple that with the GOP advantage and obama is in mathematical trouble.

    • Posted November 1, 2012 at 2:20 pm | Permalink | Reply

      I’m curious if Republicans have ever had the advantage in early voting before like this? We’re all assuming (myself included) that on top of this early voting advantage, there is also a majority of Republicans that will turn out on election day and beat the pants off the Democrats there, too. But I’m a registered Republican and voted early here in Colorado a few days ago, the first time I’ve ever voted early.
      ~ Brittany

  8. Posted November 1, 2012 at 12:17 pm | Permalink | Reply

    But the most recent confirmed models for comparison… the 2010 mid-terms, tend to show that democrat polling… or even lamestream polling… was the overly-optimistic, off the mark variety… did it not?

    Of the polling equation, wasn’t the modeling done by the right closer than that of the the left? Weren’t the dems pretty much saying and thinking the same thing two years ago?

  9. Dave
    Posted November 1, 2012 at 12:18 pm | Permalink | Reply

    Gov Christie seems like an uneven fellow. one minute his allegiance is to the people giving him national exposure at the repub nat convention, the next minute his allegiance is to a democrat (who happens to be a presidential candidate locked in a close reelection campaign) coming in with cash and for photoops. Christie needs to learn that you dance with the one who brought you. If he thinks this helps is reelection prospects he’s mistaken. As soon as the dems can find a reasonably charismatic democrat who isn’t a Corozine wannabe to run against him, he’s toast..toast with NJ and toast with republicans.

    • Posted November 1, 2012 at 2:22 pm | Permalink | Reply

      He’s officially lost my respect forever after this last little stunt slobbering all over the Socialist. Plus he’s pro abortion, pro gun control, pro gay marriage…the man is not a Republican and he should not claim to be one. He’s a right-leaning independent at best.
      ~ Brittany

      • Posted November 1, 2012 at 3:30 pm | Permalink

        and sadly that is where the Republican Party has to change. If Romney wins…then maybe we can go a long way to encompassing those who are a little closer to center. If he loses it will need to be a VERY QUICK, VERY DEEP sould search of the party to see where we are going.

        I am sorry but to say just because someone doesn’t believe exactly the right wing ideology to a tee they aren’t a republican is part of the trail we veer off of. I am a hardcore fiscal conservative, and socially nearly a libertarian in the sense stay out of my business (fed gov)….that means the federal government should have no say in abortion, gay marriage, etc. That should be leave up to the states.

        I personally despise abortion but dont believe the federal government should outlaw.Roe v Wade should be overturned and the decision to the states. Gay marriages, rights, etc…I could care less the less government is involved the better. Guns, everyone should have one 🙂

        Our tent has to be inclusive of anyone from dead center to the far right. We cannot keep blasting guys like Christie FOR DOING THEIR JOB. What you are reading, what you are hearing on talk radio about Christie being a traitor, he flipped sides, he is campaigning for Obama….is falling right dead into the trap Obama wanted you to. To prove that Republicans are not bipartisan, that they cannot put politics about the job at hand. Christie, a republican governor of a deep blue state, did exactly what he should and needed to do. The fed has crucified his state financially, education funding wise, etc. He is up for re-election in a year. He did what he should and needed to do. Anyone calling him a traitor, no respect, etc…is handing Barack Obama the middle of the roads he wanted.

      • Posted November 1, 2012 at 4:03 pm | Permalink

        You’re certainly welcome to your own opinions as I am to mine. 🙂
        ~ Brittany

      • Posted November 1, 2012 at 4:52 pm | Permalink

        EXACTLY…we, as a party should be open to all views even if they differ with the specifics

  10. Posted November 1, 2012 at 12:24 pm | Permalink | Reply

    Keith your work is eye opening. Nothing like it anywhere else. Question: you typically go after the party ID in the polls but isn’t that criticizing the result instead of the cause? Ie these polls like Marist don’t weight party ID they weight demographics (race and gender) right? Errors here result in errors of party ID if I understand the point. Therefore wouldn’t it be more important to report the demographic weighting of polls as compared to prior elections and reasonable expectations rather than (or in addition to) the party ID in polls?

    Seems like the key to the whole thing is the racial makeup of the electorate since Romney is winning the white vote at historic rates while Obama is winning the non-white vote at similar historic rates. Eg your piece on Axelrod’s Folly demonstrates to me that the election largely turns on whether the white vote is 72% (Axelrod’s assumption) or closer to 78% (Gallup’s recent conclusion – The center for immigration studies projected 73.4% white vote.

    Since this spectrum seems to the crux of the matter I would curious to see the race assumptions in each poll.

  11. Posted November 1, 2012 at 12:27 pm | Permalink | Reply

    Is it true that the margin of error doesn’t factor the possibility that the assumptions are off?

    • NMVM
      Posted November 1, 2012 at 3:52 pm | Permalink | Reply

      That’s correct. The MOE has to do with sampling error. It doesn’t account for non-sampling error like non-response bias and likely voter screen error.

      In other words, if you increase the number of people in your survey you’ll do a better job of estimating the attitudes of the population that the sample represents. But if your sample (after weighting for demographics) doesn’t accurately represent the population you’re trying to sample, increasing the number of people won’t bring reduce the error.

  12. Evan3457
    Posted November 1, 2012 at 12:31 pm | Permalink | Reply

    Don’t know if I’ve made multiple posts of the two articles or not. Not seeing any of them in my browser. Sorry if I multiple posted.

    • Evan3457
      Posted November 1, 2012 at 12:31 pm | Permalink | Reply

      I see this one, but not the others. I’ll try again.

      • Evan3457
        Posted November 1, 2012 at 12:45 pm | Permalink

        Must be something wrong with the links, so I’ll just mention the articles without them. Dan McLaughlin at Red State, “On polling models, skewed and unskewed”, and Bob Krumm (from Oct. 24) “Garbage In = Garbage Out”.

  13. Posted November 1, 2012 at 12:47 pm | Permalink | Reply

    Seems like Gallup has a good record of predicting election demographics.

  14. Posted November 1, 2012 at 5:29 pm | Permalink | Reply

    ABC NEWS/WAPO Poll – Obama 48.56 Romney 48.49

    Here is some interesting data: “The closer race shows another way: Among likely voters who supported Obama in 2008, he’s retaining 83 percent this year – but 14 percent are moving to Romney. Romney, by contrast, is retaining more of John McCain’s supporters, 94 percent, and losing just 5 percent to Obama. That said, Obama’s making it back among new voters: Seven percent of likely voters say they did not vote in 2008, and they favor Obama widely, by 62-34 percent.”

    Here’s my math:

    Obama in 2008 got 69,456,897 votes….according to this poll he retains 83% = 57,649,225 votes + he gains 5% of McCains voters = 2,996,741 plus there are 7% of supposed new voters = 8,869,606 of which Obama is gaining 62% = 5,499,156 for a grand total of 66,145,122 votes.

    Romney is retaining 94% of McCain’s votes from 2008 (59,934,814) = 56,338,725 + he gains 14% from Obama’s 2008 voters = 9,723,966 plus he gains 34% of the new likely voters that didn’t vote in 2008 = 3,015,666 for a grant total of 69,078,357

    Romney 50.95% Obama 48.79%

    that was fun

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: