Rasmussen: Chambliss +4

[Meeker]

Chambliss: 50%
Martin: 46%

This telephone survey of 700 Likely Voters was conducted by Rasmussen Reports November 18, 2008. The margin of sampling error for the survey is +/- 4 percentage points with a 95% level of confidence.

Republican incumbent Saxby Chambliss holds a four-point lead over Democratic challenger Jim Martin in Georgia’s closely-watched Senate runoff race, according to the first Rasmussen Reports survey in the state since Election Day.

Chambliss leads 50% to 46%, with the vote scheduled for December 2. Four percent (4%) are undecided. However, runoff elections typically have lower voter turnout than general elections and can be impacted in either direction by organized get-out-the-vote efforts.


Translation for the bolded part: We really have no fucking idea what's going to happen.

[Alcon]

I wish pollsters would report how many people get past their LV screens on run-offs, 'cause that would be hilarious.

[Lunar]

Are write-ins possible for a run-off election?

I want both to be under 50%.

[Sam Spade]

Ya, this is rather useless...  SUSA may the only polling company to halfway trust in this situation, since they don't use the standard turnout models.

I should note that among those "Certain to Vote", Chambliss leads 51-46, whereas those "Not Certain to Vote", Martin leads 47-41.

[Sam Spade]

Are write-ins possible for a run-off election?

I want both to be under 50%.

No write-ins.  That's why it's called a runoff.

[Lunar]

Are write-ins possible for a run-off election?

I want both to be under 50%.

No write-ins.  That's why it's called a runoff.

I figured, but Georgia is weird mmm

[Queen Mum Inks.LWC]

Are write-ins possible for a run-off election?

I want both to be under 50%.

No write-ins.  That's why it's called a runoff.

Yeah, I thought that was pretty obvious.

[Queen Mum Inks.LWC]

Database entry: https://uselectionatlas.org/POLLS/SENATE/2008/polls.php?action=indpoll&id=1320081118016

[Tender Branson]

What's bad for Martin:

"In Georgia, 52% of voters say they are less likely to vote for Martin if it means the Democrats will gain a 60-seat majority in the Senate. Thirty-eight percent (38%) say they are more likely to vote for Martin if that’s the outcome."

So, if Franken manages to pull ahead in the recount, Martin's probably losing by an even bigger margin.

[Хahar 🤔]

Doesn't matter.

Martin ran behind Obama. That means that McCain-Martin voters weren't a significant group.

[Sam Spade]

Doesn't matter.

Martin ran behind Obama. That means that McCain-Martin voters weren't a significant group.

You do realize, I hope, that most of the reason why Martin ran behind Obama is explained by black voters filling out the ballot for Obama and not paying attention to the other races.  Shockingly high in Fulton and DeKalb, for example, probably about 30-35,000 votes.

[Lunar]

I thought the same thing and almost posted it (minus county statistics) but I realized that black undervote doesn't necessarily make Martin-McCain voters a significant vote.  Well, there's nothing Martin can do -- he can't embrace McCain like Lunsford attempted without alienating himself stupidly.

The all-out, I'm-all-for-Obama strategy is probably the right one, although that does create the moderate risk of making the race more important for apolitical Republicans.

[Sam Spade]

I thought the same thing and almost posted it (minus county statistics) but I realized that black undervote doesn't necessarily make Martin-McCain voters a significant vote.  Well, there's nothing Martin can do -- he can't embrace McCain like Lunsford attempted without alienating himself stupidly.

The all-out, I'm-all-for-Obama strategy is probably the right one, although that does create the moderate risk of making the race more important for apolitical Republicans.

Oh, there are McCain-Martin voters.  But I suspect their number is less than Obama-blank voters (or at least somewhat similar).

FYI, an Obama-all-out strategy also creates the risk of alienating Republicans, as Al points out.