Obama Will Win

Speaking in Dubuque, Iowa, Obama fires up the crowd as the campaign enters its final days

The election seems to breaking President Obama’s way.   A tweet from Nate Silver put it this way:  “Sample average of national polls released Thursday Obama +0.9,  Friday Obama +1.2, Saturday Obama +1.3, Today so far Obama +1.4”  

The veritable Pew Research Group, which had Romney up four after the first debate, found them even after the second, and on Sunday found Obama leading 50-47.   The national numbers finally seem in line with the state numbers.

Still, I’ve been struggling with this prediction.   The national polls remain close.   Team Romney wants to claim that this means the election will entail a razor thin victory which could go either way.    That sounds very plausible.  Yet the state polls have shown a convincing lead for Obama, though greater in breadth than depth.   Moreover Republicans have argued that the polls are wrong because they are making false assumptions about voter turn out or purposely skewed to support Obama.

Ridiculed by the right, Nate Silver’s 538 blog has shown the power of statistical analysis in elections and has made him one of the most important pundits of this election cycle

Whose assumptions are right — those expecting a close race and lower Democratic enthusiasm, or those believing Obama is on track to win?

Here are my assumptions:

1)  Voter turnout will be roughly in line with 2004 and 2008, with Latino vote increasing.

2)  While early voting won’t be dominated by the Democrats this time (after 2008 there was no way the GOP would punt on that the way McCain had), the increase of GOP voters doesn’t mean a large increase in Republican votes.  That’s because Republicans are very likely to vote anyway.  It just shows that both parties recognize the importance of early voting and are making it a priority.

3)  The polling this cycle is not off base or skewed.   Blaming the pollsters is common for the side that’s behind.

Dr. Michael McDonald predicts a turn out of 60-61%, and he specializes in making such calls.  That’s in line with 2004 and 2008.   Moreover, 2008 was not a year where voter turnout increased massively due to Obama’s “hope and change.”   To expect a decline in turnout in a hotly contested election in which billions were spent doesn’t make sense.  Finally, in surveys Latinos and blacks show intense enthusiasm for the election, even though Republicans claim their voter share will go down.

In many states efforts to pass voter ID laws and limit early voting may in fact be spurring on minority turnout.   It’s not just about President Obama, it’s personal.   “Governor Rick Scott would prefer I don’t vote,” one black man said.  “I’m not going to let him win.”  Indeed, the so-called “voter suppression efforts” of this election cycle may backfire.

Long lines showed voter enthusiasm on the last day of early voting in Florida

Going over the polls from each day over the last month, the election has followed a clear if often tumultuous path.

After the two conventions it seemed President Obama was on a roll.    The GOP convention had sounded bitter and pessimistic while the Democrats beat an optimistic drum.   Just when Obama’s convention bounce started to fade the “47%” tape came out.   Romney seemed a caricature and Obama supporters like myself started to think this could be a landslide.

Everything changed on October 3.  It wasn’t that Obama was so flat it was that Romney was so different than his image.   Rather than barking out a desire for massive tax and spending cuts with disdain for government, he came off as a reasonable moderate.  He was nothing like the Mitt Romney of the GOP primary season, nor did he sound like the plutocrat dissing the “47%.”  He was reasonable, clever and made the President look ordinary.

The President’s support had been soft, based on a strong negative view of Governor Romney.    Public perception of changed.   He caught up to Obama within days, and by the time of the second Presidential debate had opened up a 3 point lead.   The campaign was slipping away from the President.   He was still ahead in most swing state polls, but it was clear that the firewall he constructed in September was a Maginot line — a defense Romney could circumvent and overcome.

While Vice President Biden’s defeat of Paul Ryan energized Democrats, it was the second Presidential debate that stopped the bleeding for Obama.   He was perceived as having won the debate by putting Romney on the defense and turning potential problems (like Libya) into strengths.

The third Presidential debate saw an inverse of the first.   This time Governor Romney, believing momentum on his side, sat on the lead, hoping to look Presidential and undercut any image of him as a war monger.   He was docile, agreed with the President, and polls gave Obama a victory by about the same  proportion they gave it to Romney in debate one.

This Huffington Post image shows that the trend towards the finish started after the third debate – Sandy is not a major factor

Since then there has been a slow, steady drift back to the President   Moreover, Governor Romney never sealed the deal after the first debate.   He showed that he could be moderate and pragmatic but never gave people a true reason to vote for him.

Until the last few days I was not confident that Obama was going to pull this out.  It “felt” like Obama was winning, but intuition is a dangerous indicator.   Bias is driven by intuition and hunch.   If it “feels” like the race is going the way you want it to, it’s probably because you want it to!   Bias can blind people to the obvious.

I look over at right-leaning blogs and note that their belief Romney will win tends to be driven by conspiracy theories (the polls are purposely skewed in some vast left wing conspiracy to demoralize Republicans) or scenarios not justified by data (there will be very low turnout from minorities and Democrats or people are sick of Obama and will make up their minds late to make a change).

The evidence suggests Obama is heading towards victory.   Not a sure thing, but I think Nate Silver’s 85% odds are on track.

Monday I’ll post my predictions on how the states will go as well as some thoughts about the House and the three same sex marriage ballot initiatives which may end up making this an historic election.   Two more days!

Advertisements
  1. #1 by James P. Melcher on November 4, 2012 - 19:21

    If you’re going to look at the Maine votes, you should enter Mike Tipping’s contest on his Bangor Daily News blog. First prize is a $50 Best Buy gift card; 2d prize is a copy of the book Amy Fried and I wrote.

  2. #2 by margieinitaly on November 4, 2012 - 19:49

    This is great! I had already read Nate silver’s blog and think there is a lot too it…We’ll all know on Tuesday night!!

  3. #3 by SShiell on November 4, 2012 - 19:59

    Romney will win – and here’s why.

    You like to talk a lot about the polls so let’s start there. Before I let you detract from what I am about to state, I could care less what the pundits are saying (including the left’s statistical hero, Nate Silver – who stated a year ago Romney was a 71% lock on winning based upon the economy at the time and GDP and job creation numbers are even lower today – go figure). This is my own analysis for what it is worth:

    There are three basic things I look at when I break down the polls – the first is the D/R breakdown. In virtually every poll out there, state and national, that show an Obama lead there is an average of a D+7 difference in the reflected D/R Party affiliation of the respondents. That is consistent with the 2008 exit polling which showed D+8 nationally. I can tell you right now that is not going to happen this year. D+3 in these polls I could even believe but if that were to happen, or worse from the Democrat perspective, and all else stays the same – Obama loses. (Note: You may say that has been debunked – maybe to your satisfaction, not mine – keep whistling past that graveyard.)

    Next I look at where the independents are breaking. I find it interesting that less than 6 weeks ago, everybody and their dog were saying where the Independents go, so goes the race. In 2008, according to exit polling, independents broke to Obama by 8 points. In every poll, the absolute best independents show is dead even. Most polls show independents breaking toward Romney on average +9. If Independents break to Romney by even half that number – Obama loses!

    I then look closely at right track/wrong track and similar type responses. Why? Individuals who have not made up their mind as they enter the booth tend to follow that line of thinking when they pull the lever. When the stats indicate a greater % for right track, they tend to want to keep the current party in power. Polls today indicate Wrong Track is +8% or more in virtually every poll. On average, last minute voters will not lean toward Obama with that number hanging around his neck. Once again, he loses.

    Finally, outside of the polls, I look at the excitement level being generated by the respective campaigns. In 2008, Obama was the Rock Star and 80,000 showed up to see him in Cleveland during the closing stages of the campaign. McCain was pulling less that 4,000 per event during the same time. Even though I voted for McCain, I knew he was going to lose. This year Obama pulled less than 3,000 in a Cleveland suburb the other night while 30,000+ showed up for a Romney event elsewhere in the state. Obama . . . well, you know – he loses.

    But then again, I could be wrong.

    PS: First indication – Virginia polls close at 7PM. If it is a dead heat, it could be a long night. Should Romney win by 3% or more – pull out your crying towel.

    Cheers & Out.

    • #4 by Scott Erb on November 5, 2012 - 07:08

      The length of your post suggests you’re trying to rationalize rejecting the work of pollsters for whom this is their livelihood, with their reputation on the line. Perhaps you are right, but usually when people are seeking reason to reject the polls it’s wishful thinking — every cycle. But we’ll see. Also cherry picking data from events looks a bit like a stretch; Obama’s had huge rallies the last few days. It looks like you’re seeking a reason to rationalize rejecting the conventional wisdom. We all do that – and sometimes that leads to the right conclusion. Again, we’ll see.

    • #5 by Scott Erb on November 5, 2012 - 07:48

      Or shorter: I’m relying on the data. You’re relying on the data being flawed. I don’t find that to be likely, even though it is possible. I don’t get into the details arguing about this because there is no way to prove who’s’ right — except for watching the results come in.

      Markos Moulitsas tweeted: “To have any hope for Tue, GOPers have to ignore the data. Don’t worry, they have plenty of practice doing that.” I think that happens on the right alot – climate change being a classic example. They have an ideological belief and then seek to cherry pick facts to prove their point and find ways to try to rationalize rejecting the rest of the data.

      In a small percentage of cases the data is flawed, and this election may be one of those times. We’ll see.

    • #6 by SShiell on November 5, 2012 - 10:21

      “The length of your post suggests you’re trying to rationalize rejecting the work of pollsters for whom this is their livelihood, with their reputation on the line.”

      There are only two prominent polling institutions whose livelihood is on the line – Gallup and Rasmussen, and their results differ from much of the others. Of the rest, any one that is aligned with a University or Media outlet has nothing on the line but the possibility of some egg on their face. (CNN, NBC, ABC, Fox, National Journal, IBD, NYT all fall in this category)

      “I’m relying on the data. You’re relying on the data being flawed.”

      I broke down the data. The data was not flawed, the models are. They all depend upon the “2nd to none” Obama GOTV effort. (And this with all indications that there is far more enthusiasm on the R side.)

      Case in point – the recent CNN poll – Dead even between the two candidates. But independents go to Romney +22 points and there was a D+11 sample. Independent to Romney +22 is a 29 pont swing from the 2008 exit polls. D+11 sample is 4 points greater than the 2008 turnout nationally! I don’t question the data, the results tell me one thing – Obama has to turn out his base at a greater percentage than he did in 2008 to even be competitive. That is not flawed data, that is a flawed basic assumption for the model.

      Let’s keep it simple – Occam’s Razor if you like – answer 2 questions:
      1. Do you really think this year’s Democratic turnout will equal or exceed 2008 compared to the Republicans?
      2. Do you really think that if independents turn decisively to Romney, that Obama will still win?

      If either question’s answer is No then Obama is in deep trouble. If the answer to both questions is No then Obama loses. To say otherwise is “Whistling past the graveyard!”

      As you say, we’ll see. Cheers and out.

      • #7 by Scott Erb on November 5, 2012 - 11:54

        There clearly is a chance the polls are statistically biased through false assumptions and methods. I don’t think they are, but we’ll find out tomorrow. Either way the Republic moves on!

  4. #8 by jared cantin on November 4, 2012 - 20:28

    jaredThe latest trends of Rasmussen polling, who had Obama winning Wisconsin by only 7 points in 2008 (when he won by 14 I think), makes me think Obama gets WI, Mi, Ohio, Nevada, and wins by a decent amount in the college, and a point nationally.

  5. #9 by brucetheeconomist on November 4, 2012 - 21:49

    Most recent election, it’s seemed to that the losers knew it was coming. This time whoever loses may be angered by both believing they should have won and expecting they would only to see those expectations dashed.

    Polls seem less reliable, maybe because its harder and harder to get people to be polled, especially on the phone.

    A large number of people expecting to win and not doing so could mean even more anger – as if we needed that.

    • #10 by Scott Erb on November 5, 2012 - 07:02

      Virginia, Florida, North Carolina, New Hampshire and Ohio could give it to Romney, if he ran the table on those states. The data says its unlikely, but small shifts in either could make it happen. Since Republicans believe polls are over sampling demographics that vote Democratic, they expect it to happen. We’ll see!

      • #11 by Gary DeWaay on November 5, 2012 - 10:13

        They forget that nobody likes Romney though. Even the staunchest Pubs I know start out talking about him with the caveat… “I don’t like Romney much… BUT…”

      • #12 by Scott Erb on November 5, 2012 - 11:55

        I have a Republican activist friend who confided with me, “We don’t want Romney, I don’t think people realize how bad he’ll be for the Republican party.” He said Democrats should be hoping Romney wins for that reason! I guess if Romney does win I’ll have to hope he’s right.

  6. #13 by Norbrook on November 4, 2012 - 22:01

    What I’ve been saying for a while is that that the focus on the “national polls” is not indicative of what the election is going to hinge on: The Electoral College count. It’s there that Romney has had the most difficulty all along, and he’s never been able to “seal the deal” when it comes to the swing states.

    Another thing that’s making a difference is that Romney has never given people a reason to vote for him. His whole strategy has been based on being “not Obama,” which may play well to the Republican base, but is not something to draw votes outside of that.

  7. #14 by Snoring Dog Studio on November 5, 2012 - 08:38

    On Wednesday, my efforts will go toward not gloating about an Obama win, which is going to happen, no matter how much in denial some of your commenters are. I might just send a few notes to the uber wealthy pigs who poured millions into the Romney coffers – to remind them that they picked a loser.

  8. #15 by Gary DeWaay on November 5, 2012 - 10:08

    I can see this election breaking either way except for one major problem… NOBODY LIKES ROMNEY! Hell, even the preening prick the Shrub had tons of people enthusiastic about him, and having Chris Mathews say he’s a guy people can see having a beer with you. Romney would more likely pay someone else to have a beer with you. This is going to be a landslide. Hoping the indies and swing voters are going to swing towards a rich a-hole prick is a fools game… it never happened with Kerry (much to my amazement) and its sure as hell not happening now. Especially with Obama being such a formidable, likable candidate. I know I know.. he drives the far right insane, but what I said is true to the non-33 percenters. Landslide… with 53 Dem Senators on the coattails and a slim Dem House. Bank it!

    • #16 by Scott Erb on November 5, 2012 - 11:57

      If we get a slim Dem house that will really be a sign that things broke Obama’s way. Right now it doesn’t look like it’ll happen, but on election night over 435 races, many tight, it’s possible! You gonna be able to boot out Noem?

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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: