Posted on 02/11/2008 1:10:17 PM PST by dangus
Friday, Rasmussen didn’t even poll Republicans on their primary choice; the national news media already announced McCain had won the nomination. Saturday Rasmussen did release their poll: McCain trounced Huckabee, 55-24. Sunday, as the nation learned Huckabee was still a candidate, Huckabee improved, but still trailed greatly, 49-29. Monday’s poll shows Huckabee pulling to within 12 points, 46-34.
This seemed in monumental conflict with Gallup, which showed McCain winning over those same three days, 56-25. But now another poll, AP/IPSOS shows the race just about as close as Rasmussen, 44-30. (The GOP establishment should be nervous that Ron Paul also polled at 9% nationally.)
But is it even closer than Rasmussen shows? All three polls were taken over the same time period. Only Rasmussen released results each day. With them having done so, we can extrapolate from the data roughly which percentage of respondents supported each candidate each day. Saturday McCain won 55-24. To have dropped to a 49-29 lead, McCain must only have led in Sunday’s poll, 43-34. And to have dropped further to a 46-34 lead, McCain seems to have actually lost Monday’s poll, 40-44.
Some caution is necessary: When splitting hairs like this, the effect of rounding and statistical error is amplified. Depending on rounding, McCain and Huckabee could have tied at 43 percent in Monday’s poll. And the margin of error for such a small sample is six percent… which means McCain could have gotten six percent more, and Huckabee could have gotten six percent less. Which means McCain could actually have led Huckabee by twelve percent in the most recent day’s polling data. (But, by the same logic, Huckabee could be winning by twenty percent.)
AP/IPSOS’ three day results are very similar to Rasmussen. Unfortunately, AP/IPSOS didn’t release day-to-day numbers. So we can’t tell if Rasmussen’s results are the result of an amazing tightening, or just bizarrely flukish data on Monday’s release cancelling out equally bizarre data the other way on Sunday. But it seems safe to say that AP/IPSOS confirms that Rasmussen’s overall three-day data is realistic.
So why is Gallup’s so skewed in comparison? Gallup polled “voters” not those who were were most likely to vote. It’s likely less politically active voters were less likely to realize that there was still a battle to be fought. Also, Gallup’s polls counted independents, even though independents won’t be allowed to vote in most remaining states.
It’s still worse for McCain coming up. Whereas California, New York, New Jersey, Florida, Delaware, and New Hampshire all have large numbers of moderate Republicans, and all voted for McCain, of the remaining states, that’s only Oregon, Maryland and Vermont tend towards moderates. (I’d also suppose McCain could do well in the military-heavy states of Virginia and Hawaii.) And what happens if McCain voters believe the media hype and don’t bother to vote, while Huckabee’s supporters swarm to the polls?
The notion that Huckabee can’t win simply isn’t at all true. Republican states are mostly winner-take-all, so Huckabee can still nearly every remaining delegate with 51% of the remaining votes. (And even that doesn’t account for the Paul voters.) Even winning scarcely more than half of the remaining delegates could force a brokered convention. Or, more likely, a negotiated victor with Romney supporters and state party leaders picking the winner. Whereas such a deal would haunt Clinton or Obama, who have already considerably tapped potential primary donors, Huckabee could campaign full-throttle on what are legally "primary" funds.
There is no race. McCain will win. The military will not support Huckabee nor will the Northern or Western Republicans.
What Northern and Western Red states will Huckabee lose that Bush won?
Its McCain’s to lose....but he seems to be tanking a bit after the CPAC convention...where Huck is gaining.
The GOP needs to tell McCain to step aside. Candidates are dropping out...and he is losing ground. Thats bad
You mean, like in Washington State?
>> Northern or Western Republicans. <<
The Northeast is done. Maine, Massachusetts, Connecticut, New York, New Jersey, Delaware, and New Hampshire have already voted. I’m not worried about Vermont. Washington looks to be a three-way tie (Saturday only chose delegates). Paul could even win Washington. California is done.
It is not a mathematical impossibility. That part is about all that is true. But it is so unlikely as to be obvious to all but the delusional at this pont.
...and as to most GOP states being winner take all...I believe that most of the ones left are proportional when taken as an entire state. Maybe winner take all by precinct, congressional district, or county...but not at all by state. Almost all of those have already voted.
(Saturday only chose delegates)
Meaning caucus delegates.
The military will support any Republican! They don’t a rerun of the Clinton era.
This is not a banana republic!
Considering he came within a whisker of McCain in Washington, I would not be so sure about that.
Many Conservatives and Evangelicals will not support McCain in November when it counts.
Huckabee and McCain in a horse race! Wow. This is just plain depressing.
It’s like my doctor announcing with breathless excitement that he’s not sure whether the crotch rot will destroy my genitalia before the gonorrhea does, or vice versa.
Sounds good. I just can’t stand the thought of McPain.
I believe that the poster is talking about just that...the military vote. Nothing I read hinted at any kind of military coup...just their voting block...which they are.
>> delusional <<
Name-calling hardly makes your argument.
vote for “Barf” or “Barfer”...
Huckabee could still win Washington. What happened Saturday was only the selection of unpledged caucus-goers.
And the more people learn about McCain, the worse he sounds.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.