Ann Selzer's final IA-Sen poll was right

I’m making good on my promise to declare Ann Selzer a polling genius. Selzer & Co’s last survey of the Iowa Senate race looked like an outlier. No one else’s final poll showed Joni Ernst above 50 percent or leading Bruce Braley by more than 3 points. I was also skeptical because of the big swing toward Braley in Selzer’s mid-October poll, followed by a big swing against Braley in her late-October poll. But in the final analysis, Selzer’s last Iowa poll for the Des Moines Register called the race 51 percent to 44 percent for Ernst. At this writing, unofficial returns put her ahead by roughly 52 percent to 44 percent.

I still think it was irresponsible for the Sunday Des Moines Register to put Selzer on the front page saying the IA-Sen race was over, knowing that several other polls in the field at the same time had found the race to be a dead heat. However, Selzer called her shot and stood behind her data and methodology. I respect her for that.

I don’t have the same respect for everyone on the Des Moines Register’s politics team, but that’s a topic for another day and another post.

About the Author(s)

desmoinesdem

  • Ann Seltzer - Supergenius

    Well done and nicely put.  I think we were all hoping you were right.

  • Selzer deserves kudos

    I dismissed her poll, with good reason, because most of the time in polling when “one of these things is not like the others,” that thing is wrong.

    But her outlier was right and the others were wrong.

    She now has two big feathers in her cap, the Iowa Dem caucuses in 2008 and IA-Sen 2014.

    I still don’t know what happened in the Iowa state Senate, and Google is no help because the federal race dominates search results.

    Help?

    • two seats flipped

      one in each direction, so it remains 26/24 D.

      I think the GOP picked up 4-5 IA House seats to make it 57.

    • Democrats held the Iowa Senate

      Unfortunately, Daryl Beall lost, but fortunately, Kevin Kinney picked up the seat Sandy Greiner vacated. We should have been able to expand the majority, but we needed better candidates in Sioux City and Ottumwa. Mark “Chickenman” Chelgren should never have been elected the first time, much less re-elected.

      Looks like Rs had net gain of four in the Iowa House, so 57-43.

  • Nate Silver's take on poll bias this cycle

    An interesting read on poll bias flipping from favoring GOP in 2012 to favoring Dems in 2014 as well as an admonition against drawing future conclusions from that fact.

    http://fivethirtyeight.com/fea…

Comments