Wargaming the Electoral College: Quickfire Thursday Edition

Pajamas Media ^ | 10/11/2012 | Stephen Green

[SeekAndFind]

If Suffolk University was correct in its decision to quit polling FL, NC, and VA voters and paint those states red, then here’s our new battleground map, with MI and PA slipping (very slightly) into battleground territory.

Now I never did consider NC to be a battleground state, but it would appear that Romney has moved the playing field to the right. Two battlegrounds (FL, VA) have turned red. And three blue states (MI, PA, WI) are opening up. The GOP looks to have a great ground game in OH, and a win there puts Mitt in the White House. Romney can also score a win with with just PA or just MI. Either IA + WI or NH + WI gets Romney over the finish line, too.

If I’m wrong about CO, that opens things up for Obama a bit more. My lyin’ eyes though tell me CO has rather quickly become a gimme for the GOP.

In a close race, Obama wins it 281-257, or Romney takes it 285-253. But if Romney breaks this open — which he now has some small chance to do — then it’s 331-207, and a pretty complete repudiation of the last four years.

UPDATE: For what it’s worth, RCP moved MI and PA into the tossups this morning, too. All is proceeding exactly as I have foreseen.

[SeekAndFind]

Here’s the NBC take on their swing-state polling:

Romney saw his largest gain in Virginia, where he now edges the president 48 percent to 47 percent, a 3-point reversal from last week’s poll, released the day of the first presidential debate. The spread is within the poll’s margin of error.

In Florida, before the debate, it was a 1-point race with Obama leading 47 percent to 46 percent. Now, it is still a 1-point race with Obama leading 48 percent to 47 percent.

In Ohio, where there has been a renewed focus by the Romney campaign after the former Massachusetts governor’s strong debate performance, Obama leads 51 percent to 45 percent. That’s a 2-point uptick for Romney.

There is, however, a big caveat on those numbers from Ohio. NBC reports it in the very next paragraph:

But the Ohio poll also included an 11-point advantage for self-described Democrats — 40 percent to 29 percent for Republicans. Last week’s poll had a narrower 5-point advantage for Democrats. . (In 2008, the party identification split was 39 percent Democrat and 31 percent Republican, according to exit polls.)

In the 2010 midterms, Ohio had an R+1 turnout model, with a D/R/I of 36/37/28. This sample is ridiculously tilted toward Democrats. Even so, Obama lost two points since their last iteration, and only leads by six in a poll with an eleven-point advantage for Democrats.

But how interesting is it that NBC felt compelled to report that D+11 sample? Looks like the media has figured out that news consumers have become more savvy to sample issues.

[mkboyce]

Wow!

But I must ask...what the ef is wrong with Nevada? Is it the unions in Las Vegas?

[SeekAndFind]

CBS reports on Romney’s rise in their own polling:

On the heels of last week’s presidential debate, Mitt Romney has emerged with a one-point edge over President Obama in Colorado and has cut the president’s lead in half in Wisconsin, according to a new Quinnipiac University/CBS News/New York Times poll.

In Wisconsin, where Mr. Obama led Romney by six percent last month, the president now holds just a three-point advantage, with 50 percent to Romney’s 47 percent support. Last month, the president led Romney 51 to 45 percent.

In Colorado, the two remain locked in a dead heat, with Romney leading Mr. Obama 48 percent to 47 percent, within the poll’s margin of error. Last month, also within that margin, Mr. Obama had the one-point edge, with 48 percent to Romney’s 47 percent.

They also find that the more people pay attention to the presidential race, the more they like Mitt Romney:

Romney made gains among voters who are paying a lot of attention to the presidential campaign. Among these voters, Romney leads in Colorado by 53 percent to 44 percent, and in Wisconsin he leads by a smaller margin (51 percent to 47 percent). When measuring voters who have been paying close attention, the race becomes even in Virginia, 49 percent to 49 percent.

[Perdogg]

I am not so sure about NV for 0bama. Latest Suffolk RV poll for Nevada had 0bama up only 4 points. But get this; 0bama was leading only 4% in Washoe Co which he won by 12.5% in 2008.

[Perdogg]

see my response below

[SeekAndFind]

RE: Latest Suffolk RV poll for Nevada had 0bama up only 4 points.

The first question I ask is always this -— What’s the D/R/I sample breakdown of this poll?

[Homer_J_Simpson]

In a close race, Obama wins it 281-257, or Romney takes it 285-253. But if Romney breaks this open — which he now has some small chance to do — then it’s 331-207, and a pretty complete repudiation of the last four years.

The Colorado guys that have been very accurate with their economic modeling system since 1980 came out with their final prediction: Romney 330, Obama 208.

[Perdogg]

Sorry, 0bama by 2

This is a registered voter poll with party ID D +8 (Dem 41, Rep 33, Ind 23). In 2008 it was D +8 (Dem 38 Rep 30, Ind 32) and in 2004 it was R +4 (Dem 35, Rep 39, Ind 26).

[T-Bird45]

FTA: All is proceeding exactly as I have foreseen.

Wonder if he realizes how much that sounds just like Emperor Palpatine in "The Empire Strikes Back"?

[devolve]

UH - THIS WAS WORKPLACE VIOLENCE

JUST A FEW BUMPS IN THE ROAD.....

WE'LL BRING IN TELEPROMPTERS

. . . . . . . . . .

[ColoradoTim]

This is still a turn-out election. You have to show up and vote!

Here in Colorado, voter registration went up from 3.2 million in 2008 to 3.6 million this year. The last minute registrations were pouring in to the Secretary of State’s web site at 100 per minute on the Monday. That is really fishy to me. How can they verify the legitimacy of hundreds of thousands of registrations by next Monday — when they start sending out ballots.

[Pollster1]

In contrast, Rasmussen has Obama up by one point in CO, but he has NV tied, where this gives NV to Obama. Rasmussen does support the Romney lead in NC and FL, but only by 3% and 2% respectively. Ras also has Romney by 1% in VA, not a sure thing that would justify an end to polling.

My take: This analysis suffers from excessive exuberance, but at least it’s not at the “irrational exuberance” level that worried Greenspan so long ago. I’m not saying the conclusions are wrong, nor am I commenting on the implications of any privately held polling data, but the public data from polling do not numerically justify the Suffolk University conclusion I see here. I am not commenting on the validity of the methods Suffolk is using, just that the polling numbers alone would not justify their position. Keep working, volunteering, watching the polls, and monitoring for massive fraud - a win is not a sure thing.