Star Tribune eschews final-Sunday election poll for first time in years

If you hate political polls — and if you’re a GOP stalwart who really hates the Star Tribune’s Minnesota Poll – you got a gift of sorts this weekend.

For the first time in several years, the Star Tribune did not publish a poll on the final Sunday of the election year. One reason, according to editor Nancy Barnes: the newsroom did not want be perceived as influencing the election.

In fact, she goes one step further: “A poll can sometimes influence the outcome of an election.”

While DFLers have won four of the last five major statewide elections, the Minnesota Poll has overestimated the ultimate DFL vote in every one — though sometimes by a hair, and even though a new pollster took over in 2008. Here are the final Strib totals compared to the ultimate votes:

2004 presidential: +5 Democrat
2006 gubernatorial: +4 Democrat
2006 U.S. Senate: +1 Democrat
2008 presidential: +1 Democrat
2008 U.S. Senate: +4 Democrat

While the Strib editorial section has been full-throated in its Tom Horner cheerleading, Barnes has no role with them, and she is famously reluctant to have her newsroom perceived as biased. In 2008, Barnes told her Metro section columnists to “refrain from partisan political commentary in their columns in the news section” down the stretch. As I noted a few weeks ago, her new columnists tempt the fates far less often.

Money was also a factor here. Although the Strib has upped its polling budget this year, the paper decided to include cell-phone-only voters, who are more expensive to reach. Barnes say that meant doing three polls instead of four, and she decided to kill the final one on the schedule.

I think the columnists missed a chance to add more meaning to 2010, but I’m not as bummed about a missing poll. They’re fun for numbers junkies like me, but they can also suck all the oxygen out of campaign coverage. And I can understand if the newsroom wanted to distance itself from a number that is often mistaken as a prediction.

(MPR may want to do the same thing in a few days if its Humphrey Institute poll — currently an outlier showing a 12-point lead for DFLer Mark Dayton — doesn’t wind up closest to the pin.)

Still, it’s a bit odd to tout your polling all year and then fold your tent with 12 days to go.

Given that the Pioneer Press abandoned the political polling game after 2006, newspaper numbers now are absent from the final news cycles. (There were years when the Strib published tracking polls in the final hours of the election.) That leaves the field to TV stations, pollsters with partisan sympathies, and university researchers whose numbers are even more suspect.

Barnes, however, suggests 2010 might be a one-time thing: “We are likely to take a fresh look at how we poll next year, so I wouldn’t interpret that decision too broadly.”

Comments (4)

  1. Submitted by B Maginnis on 11/01/2010 - 12:06 pm.

    Best decision by the Strib in years.

    Emmer by 4-5%.

    Even the cheating won’t win this one for the libs.

  2. Submitted by Rosalind Kohls on 11/01/2010 - 03:13 pm.

    The Strib doesn’t want to draw attention to the fact that its endorsement of Horner didn’t influence many readers. A poll would make that glaringly obvious. Horner isn’t doing well.

  3. Submitted by B Maginnis on 11/01/2010 - 04:01 pm.

    So will an Emmer victory make it’s longterm water carrying for Dayton.

  4. Submitted by r batnes on 11/01/2010 - 06:38 pm.

    BD Maginnis says:

    “Emmer by 4-5%.”

    Not a chance. You’ll be crying on your pal swiftee’s shoulders on Nov 3rd.

Leave a Reply