One Day: No Surrender, No Retreat

There are 38 hours and 29 minutes until the election, as I write this.

It’s a bit early in the day to do detailed polling, with more polls due to drop before the evening. But here’s what today looks like as of right now:

ec battleground(Updated below with a new version of this graphic with North Carolina spelled correctly, and more recent polling data.)

Something to note is that these numbers are going to start moving more than they have been until now, because Ipsos is polling almost all of these states daily. Every day I will remove one Ipsos poll from the average just to add another. I looked at the poll data and these are dedicated state polls, not breakouts of national polls, so I’m leaving them in. But consider things to be just a bit more noisy now because of that.

Here’s the new graphic that I think best sums up all the data. Again, the margin of error is very low due to poll aggregation. The lead each candidate has is based on an average of the most five recent polls of that state, and the margin of error is calculated from the aggregate sample size of all five polls and the population of each state. The electoral count for each candidate is the cumulative total based on that candidate winning that state, plus all that came before it.

Since Barack Obama begins with 247 safe electoral votes, winning Nevada will give him 253, and winning Ohio would give him the requisite 271 to win the election with no further calculations.

Mitt Romney begins with 191 safe votes. Winning North Carolina would give him 206, Florida 235, and up on the list until Ohio, which would give him 285.

As you can see, Obama is leading in enough states to hit 270, with four states and 26 electoral votes to spare. Mitt Romney is trailing in at least five states that he must win to clear 270. All but one, in fact.

From yesterday, Colorado has slipped into the margin of error, and Romney’s lead in Florida has risen out of it. Without either state, the electoral college result would be 294-235, Obama high.

I think this is what you’ll see on Monday as well, with Mitt Romney gaining nothing substantive, and Obama gaining or losing states based on nothing more than his lead drifting in and out of the margin of error in Virginia and Colorado. But again, that’s not a problem for Obama, because all states below Ohio are firewall states that simply pad his lead beyond what he needs to win, states that like armor can absorb a great deal of damage before failing.

* * *

Of other things going on today …

Wisdom from The Nate:

Since ’88, POTUS candidates with a lead of 0.1 to 5 points in final likely-voter polling average in a state have won 30 of 33 times (91%).

New early voting stats in Ohio: Cuyahoga County = 35% of 2008 turnout so far. Rest of state = 24%. Similar to 2008. http://www.sos.state.oh.us/SOS/mediaCenter/2012/2012-11-03.aspx

We have Obama as ~80% likely to win Electoral College if popular vote is a tie. 98% if it’s O+1. 30% if it’s R+1.

Interesting that you have some conservatives predicting not just a Romney win, but Romney sweep of the swing states, even places like PA.

Conditional upon Romney winning, it’s MUCH more likely to be narrow, e.g. he wins OH/FL/VA/CO but loses IA/PA/WI/NV.

– And Nate on Romney’s reasons for seriously trying to win Pennsylvania. Spoiler: it’s because his chances of winning Colorado, Virginia, New Hampshire, Iowa, and Ohio combined (he needs them all) are worse than his chances of winning Pennsylvania. In other words, it’s not the Obama-is-crumbling¬≠ storyline, it’s the Romney-is-justifiably-desperate storyline.

Nate Cohn on whether or not Pennsylvania is tied. Spoiler: it’s not. But it’s nice for all the Nov 3 polls.

– At least in one national poll, Romney’s lead amongst independents has been completely wiped out.

I thought about this a little bit last night and have come up with the following bullshit. Romney has been leading Obama consistently enough in many polls, while trailing overall, to present a difficult question to answer: why?

You don’t normally see that and I think we need an explanation. Here’s mine. It was kind of a talking point on the left (mine too) that the rise of the Tea Party didn’t represent a new influence on politics. Polls showed that most Tea Party’ers were Republicans, with just a handful of conservative Democrats, Independents, and Libertarians tossed in for seasoning. That made that movement addition by subtraction. People who previously called themselves Republicans were now calling themselves Tea Party’ers, but they still held conservative values and still voted for nothing but Republican political candidates.

That much was true and accurate. Despite impressive gains by the GOP in 2010, the end result of those gains was no different than any other comparable time in our history when Republicans controlled the House and made great gains in state legislatures. In other words, Republicans controlled the House of Representatives and the Senate during most of the Bush administration, well before there was a Tea Party. If anything, due to Democrats holding the Senate now and probably next year and there being no filibuster in the House, the power of the Republican Party in Congress is less after 2010 than it was before 2006.

Now that the Tea Party has largely faded away (along with Occupy Wall Street), you have to account in some way what happened to those people. Did they go back considering themselves Republicans?

Probably not. The recent surge in Independents correlates with the demise of the Tea Party as many Republicans consider themselves Independents, and Mitt Romney’s lead with Independents while he trails overall is consistent with that view. Even if the Tea Party demographic is one of Romney’s weakest supporters, they are still Republicans and he is still their nominee.

This in turn explains another oddity we’ve been seeing this fall: the supposed over-sampling of Democrats in virtually all national and state polls. One explanation is that all pollsters are biased towards Democrats. Another is that Democrats aren’t being over-sampled so much as Republicans are being under-sampled due to so many of them now calling themselves Independents.

A movement from self-identifying Republicans to self-identifying Independents simultaneously explains what appears to be an over-sampling of Democrats that looks like 2008 turnout models, and Mitt Romney leading with Independents in too many polls to discount noise or bias, while still trailing Barack Obama overall in the same polls.

– Same-sex marriage legalization is on pace to win in Maine and in Michele Bachmann’s Minnesota. Despite GOP redistricting that should have made Bachmann a lock to win reelection pretty much forever, her race is tight from what I hear. A landslide Obama win could carry her out with the wave, but it’s more likely that she’ll win narrowly.

– James Fallows has a story well worth your time about what the GOP can/could/should do if Mitt Romney loses on Tuesday night.

My quick thoughts: I’m getting tired of seeing/hearing that “nobody knows what’s going to happen.” Of course that’s true, nobody knows for a fact. But many people have a pretty damn good idea based on the math that it’s well into the might-as-well-know-for-a-fact territory. At some point it becomes a disservice to repeat this point too often, as it gives the impression that nobody even has a good educated guess, which isn’t true. “Nobody knows” is a disclaimer, it is not the whole story. Nobody knows, but enough people think with a confidence level that would make you insanely rich in Vegas that they might as well know.

And this, I find quite insightful:

The question in my mind is “is this the end of the Karl Rove Party?” He pioneered the strategy of shifting the party right to get an energized “base,” also shifting it toward the new Know-Nothings they’ve become.

This was especially true in 2004 when Bush made same-sex marriage a central campaign plank. Of all the states that have banned same-sex marriage, the overwhelming majority of them did it in 2004. Very few states have done it since. In fact, the trend is clearly in favor of legalizing it as a non-political movement over the course of time.

Then, I suppose, the question in my mind is whether or not Rick Santorum or another such radical social conservative could have juiced the base that extra 2-4% that would put the GOP ticket on top this year. Or if Bush soured America enough on social conservatism that anyone that comes after with that agenda won’t make it past the primary ever again. Perhaps one of the stories of George Bush’s reign is that he won in 2000 on the back of something else, then got himself reelected on a socially conservative crusade.

A lesson for the next Republican President that’s running for reelection, perhaps.

* * *

I won’t bother with the House. Democrats will gain a few net seats, but gerrymandering has pretty much put the kibosh on a Democratic takeover unless it’s a Democratic wave election. Perhaps forever.

As for the Senate, the Real Clear Politics no-tossups map features no net change in control. Democrats retain majority control with 53 seats, and Republicans remain stuck at 47. My view of the polls of some notable seats:

Missouri: Claire McCaskill is leading Todd Akin by 5.8 points. This was a solid pickup opportunity that the GOP cost itself because a socially conservative Republican said something he actually believes, about women, and did so in public. That’s a big no-no.

Ohio: Sherrod Brown should retain his seat against state Treasurer and veteran Josh Mandel, probably about 5 points.

Massachusetts: Elizabeth Warren should take over Scott Brown’s seat by about +3. It’s closer to +4.4 without an outlier from a GOP consulting firm that’s not a pollster.

Nevada: Dean Heller should win over Shelley Berkley by about 2.2 points. It’ll be close, but polling has favored Heller consistently all month.

Michigan: Debbie Stabenow is totally sasquatching Pete Hokestra. She should beat him by 13 points, at least. Another lost opportunity where the Republican was leading for a while, then he said something. In this case, it was a racist TV ad that shit all over asians.

Wisconsin: Tommy Thompson had been narrowly leading Tammy Baldwin all year, even like half of last year. I don’t know why, but Baldwin has become the slight favorite to win by about 1.4 points.

Arizona: The polling between Jeff Flake (R) and Richard Carmona (D) is hilarious:

9/25: Flake +6

9/25 – 9/26: Flake +3

10/1 – 10/3: Carmona +3

10/7 – 10/9: Flake +6

10/7 – 10/9: Carmona +4

10/8 – 10/9: Flake +6

10/4 – 10/10: Carmona +4

10/4 – 10/11: Flake +6

10/17 – 10/21: Carmona +4

10/21: Flake +6


Connecticut: Linda McMahon has reportedly spent over $100 million of her family fortune on this election and the last. She’s still losing. She hasn’t lead a poll since October 2nd. Chris Murphy wins by 6.6 points, replacing a conservative Democrat (Joe Lieberman) with a … I don’t know. I don’t know anything about Chris Murphy, but I’d take Linda McMahon over Joe Lieberman.

Indiana: Lots of polling from the campaigns and little else, just like with Arizona. Tom Mourdock pulled an Akin and said something he actually believed in front of a microphone. But it’s still tight. Joe Donnelly (D) is leading by an average of 3.6 points in the last five polls, but that includes a stupid Howey-Depauw outlier that had Donnelly up by 11 points. I’d yank out the biased polls, but if I did, there wouldn’t be any polls. This is the best I can do:

10 poll average: Donnelly +3.2

10 poll median: Donnelly +2.5

5 poll average: Donnelly +3.6

5 poll median: Donnelly +3.0

Make of it what you will.

Texas: Paul Salder (D) is closing hard and fast on Ted Cruz (R). He’s only 18.4 points behind and at this rate may tie the race sometime before New Years.

Update I – New polls continue to come in. Just between around noon and 4:47p EST, five states have changed due to new polling. Obama increased his average lead in Colorado from 1.0 to 1.4, in Virginia from 1.6 to 3.0, and in Ohio from 2.6 to 3.6. Obama cut Romney’s lead Florida from 1.4 to 0.8, and in North Carolina from 2.4 to 1.6.

If a tossup is a state where a candidate’s lead is within the margin of error, than only Colorado and Florida are tossups. Today has been a quantifiably bad polling day for Mitt Romney:

ec battleground2

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.