Undecideds will break for...?

[Likely Voter]

As we get closer to election day and with the polls being tight, people are going to start talking about undecideds. How they break (if they break at all in one direction) can tip the election, even making up for a deficit in the polls.

So how do you think undecideds will break?

[Ljube]

If in the final round of polling there are any undecideds left, they will break heavily for Romney (at least 75/25).

[Antonio the Sixth]

For lack of evidence on either side, evenly is the most logical answer.

[memphis]

If you haven't made up your mind by now, you're probably not interested in politics and are unlikely to vote.

[RI]

Undecideds only matter if both candidates are below 50%, and even then only sometimes.

[Joe Biden is your president. Deal with it.]

Most will go for Obama. you have to be crazy to think most will break for Romney...judging by the poll 7 people are crazy.

[Spanish Moss]

Most will go for Obama. you have to be crazy to think most will break for Romney...judging by the poll 7 people are crazy.

Anyone who hasn't decided by now is both crazy and unintelligent.

[Fmr. Pres. Duke]

The undecideds normally break against the incumbent. They've had 4 years to know Obama. If they haven't convinced themselves to vote for him by now they likely won't.

[Grumpier Than Thou]

Romney.

[Fmr President & Senator Polnut]

The undecideds normally break against the incumbent. They've had 4 years to know Obama. If they haven't convinced themselves to vote for him by now they likely won't.

There almost as many times where the undecideds break FOR the incumbent than against them.

FTR I think they'll break for Romney but not by anywhere enough.

[morgieb]

Not a lot of the undecideds will actually vote, but Romney will probably win a majority of them.

[Maxwell]

Undecideds break for the challenger ALWAYS its just a matter of how much. I think Romney is uniquely unfavorable in this situation, so I think it'll be 60/40 in his favor, which really isn't that great.

[Skill and Chance]

Undecideds break for the challenger ALWAYS its just a matter of how much. I think Romney is uniquely unfavorable in this situation, so I think it'll be 60/40 in his favor, which really isn't that great.

They broke 57% for Bush in 2004.

Regardless, I'd expect something like 58-42 for Romney.

[Maxwell]

Undecideds break for the challenger ALWAYS its just a matter of how much. I think Romney is uniquely unfavorable in this situation, so I think it'll be 60/40 in his favor, which really isn't that great.

They broke 57% for Bush in 2004.

Regardless, I'd expect something like 58-42 for Romney.

Oh wow, derp me! haha

[Fmr President & Senator Polnut]

Undecideds break for the challenger ALWAYS its just a matter of how much. I think Romney is uniquely unfavorable in this situation, so I think it'll be 60/40 in his favor, which really isn't that great.

They broke 57% for Bush in 2004.

Regardless, I'd expect something like 58-42 for Romney.

Oh wow, derp me! haha

In fact... there was a very good analysis by Nate Silver that showed incumbents are almost as likely to get a very late bounce in their direction, at the same time having the challenger get a majority of undecideds.

[Antonio the Sixth]

The undecideds normally break against the incumbent. They've had 4 years to know Obama. If they haven't convinced themselves to vote for him by now they likely won't.

This is largely a myth. Undecideds have broken toward the incumbent roughly as often as the other way around.

[Fmr. Pres. Duke]

The undecideds normally break against the incumbent. They've had 4 years to know Obama. If they haven't convinced themselves to vote for him by now they likely won't.

This is largely a myth. Undecideds have broken toward the incumbent roughly as often as the other way around.

Then my poli sci teacher was wrong!

But you're likely right. I think undecideds broke for Bush in 2004.

[Eraserhead]

Something like 54/46 for Romney probably. Won't matter much.

[Niemeyerite]

Undecideds have broken against the incumbent because they voted the other candidate 4 years before (remember that no reelected president has won with less EVs than he won the first time since Wilson, IIRC), and when the President is behind in the polls, they tend to vote for the challenger because he's doing a better campaign or because the President is unpopular.

2012 is a different story: we have a tie which will probably become a slim Obama lead before Sunday, and 3/4 of undecideds voted for him in 2008. I don't think they'll break 60-40 or 75-25 against the President. It'll probably be 50-50 to 55-45 for Obama.

[Torie]

A 120 percent of them will break for Mittens. I know that because Dick Morris told me using non-euclidian geometry. Nate Silver sucks. All he does is crunch historical numbers. That is almost inhuman. Just get HAL to do that before he turns on you.

Moving right along nobody knows, and anyone who claims they do is on crack. Thanks.

[The Vorlon]

There is a slim sliver of the electorate that pollsters call "The Good Citizens"

These people are weakly engaged to the political process, tend to be either independents or very softly aligned with one party, and are only marginally engaged.

They typically represent 4 or 5% of the folks who actually vote on election day, because they DO vote pretty reliably (That is why pollster call them "The Good Citizens", because "Good Citizens" vote)

They do tend to break fairly heavy to one side or the other, but it varies which way they break.  Most of them make up their mind very late in the process, often the last weekend.

You hear political folks speak of "The breeze" where things shift suddenly by a couple points right at the end... The "Good Citizens" are that breeze...

in 1980 they went massively for Reagan
In 1984 they went massively for Reagan
In 1988 they went more or less for Bush #1
in 1992 they broke for Clinton
In 1996 they broke for Dole, not that it actually mattered.
In 2000 they broke for Gore
In 2004 they more or less split.
In 2008 they modestly broke for McCain.

No real pattern as to HOW they break between the parties or the incumbent versus challenger.

HINT !

If the election is VERY close right to the end (it likely will be) trust the polls that use multi question likely voter screens that measure enthusiasm AND past vote behavior over those the use simpler screens based on interest and engagement - Typically the multi-question screens will perform a bit better.

[5280]

Voted option 2

[milhouse24]

Too Lazy to vote.  They will stay at home. 

One could argue that they might go for Obama, since Obama is the incumbent and generally garners the "soft, front-running, pop-culture voters" so if their friends are voting for Obama, these undecided voters will cave to peer pressure and vote for Obama. 

But if Romney gains a slight lead in the polls, these "front-running voters" will move to him. 

It will also be a question of "Branding to the midwestern swing voter."  Since this election will likely be decided by the white working class in Ohio, Iowa, and Virginia, it will be the candidate who best appeals to the "midwestern white working class" 

Obama has a strong midwest volunteer ground army.  Romney has Ryan who will also appeal to GOTV efforts. 

John Kerry was close in 2004, but he was never able to shake that "weak northeast liberal stigma" and GWB had a strong branding for white suburban christian voters. 

It will be a question of "Who knows me, and can understand my life and my problems"?

["'Oeps!' De blunders van Rick Perry Indicted"]

Undecideds break for the challenger ALWAYS its just a matter of how much. I think Romney is uniquely unfavorable in this situation, so I think it'll be 60/40 in his favor, which really isn't that great.

They broke 57% for Bush in 2004.

Where did you get that number from? The official exits have Kerry winning 52% of voters who decided in the last week and 54% of voters who decided in the last month.

[TJ in Oregon]

The undecideds normally break against the incumbent. They've had 4 years to know Obama. If they haven't convinced themselves to vote for him by now they likely won't.

There almost as many times where the undecideds break FOR the incumbent than against them.

IIRC his analysis showed that to be true early on in the campaign, particularly over the summer, but by mid-October the remaining undecideds tended to break toward the challenger.