Thank you for the very useful information.

The rock-solid pair of polling questions are

“If the candidates on the ballot are McCain and Obama, for whom would you vote?”

and

“If the candidates on the ballot are McCain, Obama, and Barr, for whom would you vote?”

(with a series of obvious professional improvements: rotating the names, identifying parties, reminding voters that they are running for President, etc.)

all asked tot he same voters. I suspect that you would get the answer that you identify.

Or even if it is. You can only plausibly rig an election which is already close.

As to where Barr’s support comes from in Georgia, a NC poll by PPP (see Barr blog post) had in-depth cross-tabs that showed Barr with 12% among independents, and doing better among Democrats than Republicans (6% vs. 5%). His strongest support was among both 20-somethings and seniors, and he outpolled McCain among African-Americans. His support might lean more Republican in Georgia given his name ID as a former Congressman, but this indicates his affect on the McCain-Obama margin is small even though his total vote is high.

It’s far more likely though that Georgia will be a non-factor as either 1) Obama romps, or 2) Barr collapses, or 3) Georgia goes for McCain by double-digits like it did for Bush.

Obama may go down as the worst president since Wilson, but McCain would go down, period, and take America with him.

Of course, I don’t think McCain has any chance of winning this election unless the vote is rigged.

As usual, the polling data is not adequate to support the conclusion. One would need to determine where the Barr and whoever else is on the ballot votes are coming from in order to justify the claim.

Phillies is exactly correct. In the last Georgia poll I saw showed that roughly half of Barr voters had McCain as their second choice, and roughly half had Obama.

I would agree to this sentiment.

Most of the reports publishing Barr’s polling numbers seem to assume that virtually all of Barr’s voters would otherwise vote for McCain; historically, this has not been the case for LP nominees. It might be different this year, due to Barr’s background, but certainly not to the point where one can assume all LP voters would otherwise vote GOP…

Obama’s chances are generally overstated—Barr’s early poll showings are far higher than he will realistically get, and higher black turnout will be offset by even more conservative white Democrats moving into the R column. Leaning McCain.

Even though the reasoning is bad, I think it could be turned to the advantage of third parties. This is *not* to say that we shouldn’t challenge the “spoiler” reasoning whenever it crops up. We should. See:

http://www.logicforpolitics.com/2008/06/spoiler-fallacies-revised.html

Nevertheless, this could ultimately work to the benefit of Barr and Nader this year. If the Democrats buy the argument that Barr might spoil the election for McCain, they’ll have an incentive to include Barr in debates, etc. But if that happens, the Republicans may *insist* that the quintessential spoiler, Nader also be included.