Trouble For Joe Miller In Alaska?

A Hayes Research poll has Joe Miller in 3rd place in Alaska. They're the only ones showing that and have a very poor track record.

Via Politico comes news of a new poll that could indicate trouble for Republican Senate candidate Joe Miller after two weeks of bad press and campaign mis-steps:

After several rough weeks on the campaign trail, a new poll out of Alaska shows Republican Joe Miller has fallen to last place in the three-way Senate race.

A Hays Research Group poll released Thursday showed write-in candidates, presumably meaning Sen. Lisa Murkowski, in the lead with 34 percent, Democrat Scott McAdams with 29 percent and Miller with 23 percent. Murkowski is waging a write-in bid since losing the Republican primary to Miller.

Miller’s plunge in the new poll comes on the heels of several misfires in the local and national press. Most recently, Miller’s employment records from his job as a part-time attorney with local government were unsealed Tuesday, in which he admitted to using company computers for political purposes and then lied about it.

Miller had a high 68 percent unfavorable rating in the independent poll, which was taken Oct. 25 to 26. The poll appeared to ask the opinion of 500 respondents — not registered or likely voters, like most political polls. The poll appeared first on the local liberal blog The Mudflats.

Of course, a poll of anything other than likely voters may not mean much at this point but if Miller’s negatives among Likely Voters are anything like what’s reflected here, he could have a last minute problem on his hands.

UPDATE (James Joyner):  As of this morning, Nate Silver has a 66% chance of Joe Miller winning and has it Miller 39%, Murkowski 34%, and McAdams 26%.    RealClearPolitics has the Miller-Murkowski numbers as nearly dead even. Regardless, considering that Murkowski has already said she’ll remain a Republican if re-elected, the chances of a Republican winning are actually 100%.

FILED UNDER: Campaign 2010, Quick Takes, US Politics
Doug Mataconis
About Doug Mataconis
Doug holds a B.A. in Political Science from Rutgers University and J.D. from George Mason University School of Law. He joined the staff of OTB in May 2010 and contributed a staggering 16,483 posts before his retirement in January 2020.

Comments

  1. john personna says:

    The best news for Miller is that Murkowski can’t be on the ballet, and has to be a true write-in.

    And who can spell Murkowski?

  2. Anon says:

    The latest revelations are bizarre. First, if it was a serious poll, with real consequences, then he cheated. But if it was a inconsequential online, “fun” poll, then it was dumb.

    Did he think people wouldn’t notice when their browser history was cleared? And if the poll was tracking IPs (which it most likely was), why does he think clearing the browser history would work?

    And then the farcical denials after the fact reminded of some Monty Python sketches.

  3. sam says:

    Wonder how this will affect the election:

    Alaska judge bans lists of write-in candidates

    “Sen. Lisa Murkowski’s campaign and the state Division of Elections lose a bid to provide lists of write-in candidates to voters when they go to the polls Tuesday.”

  4. Sam,

    The Alaska Supreme Court has already stepped in to overrule that decision:

    http://articles.latimes.com/2010/oct/27/news/la-pn-alaska-senate-write-in-ruling-20101028

  5. sam says:

    Ah, didn’t know that, thanks.

  6. Chip Bennett says:

    500 respondents – not “likely” voters, not even registered voters. Topline only; no internals. The poll doesn’t show up on the Hays Research Group website.

    And it’s news… why, exactly?

  7. Nightrider says:

    Are there general legal standards (a hanging chad equivalent) for how closely a name must be spelled correctly to count? I guess it might be state-by-state. special presumptions where someone is known to be campaigning as a write-in candidate? what if someone has a mental slip and writes Frank Murkowski. Does Lisa M count?

    and what if Joe Miller were the write-in candidate — there are probably a bunch of Joe Millers even in Alaska, but I gather the election officials get to assume that all of the votes are for the same one?

  8. mantis says:

    Regardless, considering that Murkowski has already said she’ll remain a Republican if re-elected, the chances of a Republican winning are actually 100%.

    100%? Really? Did McAdams switch parties?

    Btw, Silver’s model hasn’t incorporated the Hays poll yet. Expect his projection to change a bit.

  9. James Joyner says:

    @Mantis: McAdam’s has zero chance of winning under Silver’s model. Lots of candidates are at 100% probability of winning in his system.

    Silver has it at “66% Republican” because there’s a 44% change Murkowski will win and she doesn’t have the Republican line.

  10. mantis says:

    McAdam’s has zero chance of winning under Silver’s model.

    No he doesn’t. He has a 4.6% chance of winning under Silver’s model.

    Silver has it at “66% Republican” because there’s a 44% change Murkowski will win and she doesn’t have the Republican line.

    Actually, Murkowski has a 28.8% chance in Silver’s model.

  11. James Joyner says:

    @Mantis: Thanks for the pointer. I was just going by the mouseovers from the big map; I didn’t realize these breakdown pages existed.

  12. mantis says:

    No problem! You’re right that there are some 100% projections, but none that have numbers as close as those in the Alaska Senate race.

    FWIW, I don’t think McAdams is going to win or anything. But it could happen!

  13. mantis says:

    I don’t usually link to National Review, but Geraghty did some looking back at Hays’ earlier polls:

    But I’m a little wary of this pollster’s results. In 2008, Hays Research’s final poll in the Senate race between Democrat Mark Begich and incumbent Republican Ted Stevens had the race looking like a near-blowout: Begich 48.3 percent, Stevens 40 percent.

    But on Election Day, Stevens fell just short, 46.5 percent to Begich’s 47.8 percent.

    They also had the GOP presidential ticket with the current governor barely ahead: McCain-Palin at 46.6 percent, Obama-Biden at 43.9 percent.

    On Election Day, McCain-Palin won easily with 59.4 percent of the vote, Obama-Biden took 37.8 percent.

    The Stevens/Begich numbers aren’t of much concern. They had about 8 percent undecided, and it looks like those mostly went for Stevens while their Begich prediction was accurate. The presidential election poll, however, was way off. So, you know, grain of salt.