Tim Bryant Hosted #GAGOP Chairman John Padgett

Chairman of the Georgia Republican Party John Padgett joined radio host Tim Bryant to discuss the 2014 elections. During the discussion, Padgett touched on polling. He explained that the Georgia Republican Party’s internal polling aligned closely with Tuesday night’s results. While the Party did not poll every race, staffers were confident that Deal and Perdue would win outright.

In what appears to be Party vs. Polling, Chairman Padgett took a swing at Mark Rountree of Landmark Communications and Matt Towery of InsiderAdvantage. “The polling missed it,” Chairman Padgett said. “Even at the end, it still wasn’t right.”

Here’s the relevant part of the interview:

It has become clear that polling Georgia is not an easy task. Here on Peach Pundit, our commenters questioned the accuracy of many of the pre-election polls. Chairman Padgett pointed out that bad polling effects the election process; it might not affect it at the end, but it affects the psychology and fundraising.

Landmark’s Rountree noted that his poll had a 2.5% margin of error which meant the poll was on the mark. Here is a comparison of Landmark’s final poll and the actual election results:

  Landmark Poll
Election Result
Deal (R) 51% 53%
Carter (D) 45% 45%
Hunt (L) 2% 3%
Perdue (R) 50% 53%
Nunn (D) 46% 45%
Swafford (L) 2% 2%

In the words of fellow Peach Pundit intern Chet Martin, “Presumably 5% of Georgians exist only to prove that David Perdue cannot be accurately polled.” We report; you decide. Who is right and who is wrong?


  1. This is Chairman Padgett acting silly to the point of embarrassing himself.

    It’s amusing, actually.

    Here’s the response I posted to Tim’s Facebook page when he posted the interview:

    “It’s always a fun show, Tim. Some people in politics simply get unhinged when they cannot control information or (the) source of it. And that’s the case here.

    It was humorous to learn that Mr. Padgett had not actually read the polls we, at Landmark, had produced — there’s really no other explanation for his lack of knowledge on the topic. I’ve provided a simple summary below, and the cross tabs can be found at http://www.landmarkcommunications.net

    Which raises a separate question: for purposes of transparency and honesty, where might *his* poll cross tabs be found? Let’s compare results.”

    Release the final three months of polling done by the GOP.

    Do it. Let’s clear things up.

    We are the Party of openness and transparency. I’m sure the Chairman would not want to appear to hide anything.

    People can see this for what it is: politics has people who have their own political agendas, and those people like to control information and the sources of information.

    Yet they cannot do that with Landmark. We are an independent company which, for 24 years, has remained independent. We are not puppet mouthpieces for anyone.

    An independent review of polling after the election was conducted by Tom Crawford of Georgia Report. His findings were that we published the most accurate polling numbers of the election cycle. I’ll post a few excerpts in a few minutes…

    Thanks for covering it, Will.

  2. John Konop says:

    Mark does not need me to defend him, but his polling was very good on the trend. He warned about the shift of libertarian votes at the end….had the votes been at 4 % it would of been a run off. At the current trend numbers next cycle will be a runoff if John Padgett keeps the same strategy….Spin anyway you want John Padgett, but math is math is…..8 to 11 to 7 is a clear trend, and Mark Rountree had the guts to be honest….If you buy Padgett BS….over Mark’s math, it is suckers bet next cycle…Number one reason businesses fail is they lie to themselves! Think about it JP.

  3. Incidentally, Landmark was the only firm in the entire country to publicly release a poll showing Democratic Congressman John Barrow being defeated by Rick Allen. We released it on RealClearPolitics.com, here:

    The same was true in 2010 with Democrat Jim Marshall vs. Austin Scott. And in 2010 our polls also hit that Deal would defeat Barnes. See the that interview here: http://landmarkcommunications.net/cbs-atlanta-campaign-coverage-mark-rountree-interview/

    And as mentioned above, Tom Crawford of The Georgia Report conducted an independent analysis of polling in Georgia, finding that Landmark’s results were the most accurate. Here is a link to findings;

    “Landmark Communications was one of the only polling firms that saw one of the Republicans breaking the 50 percent barrier. Their last poll for WSB-TV on Monday had Deal leading Carter by 50.8 to 44.5 percent, with Hunt at 2.5 percent, while Perdue bested Nunn by 49.8 percent to 45.6 percent.”

    “At the congressional level, there were two late polls of the 12th Congressional District race between Rep. John Barrow and Republican Rick Allen.

    Landmark Communications on Friday released a poll that showed Allen ahead of Barrow by 48-44 percent. Allen actually won by a 55-45 percent margin.”

    “Landmark came very close on late polls of two other congressional races.

    In the 1st Congressional District, Landmark’s final poll had state Sen. Buddy Carter with a 60-37 percent lead over Democrat Brian Reese. Carter won by 61-39 percent.

    In the 1st Congressional District, Landmark’s final poll had state Sen. Buddy Carter with a 60-37 percent lead over Democrat Brian Reese. Carter won by 61-39 percent.”

    The link is here: http://gareport.com/story/2014/11/06/polling-for-georgia-once-again-was-a-little-off-in-the-general-election/

  4. Edward Lindsey says:

    I will let my friends Mark Roundtree and John Padgett — both of whom are very good at what they do — defend themselves. The only thing I will add to the conversation is that Mark was on target in my race last May — much to my regret.

  5. Dave Bearse says:

    “but it affects the psychology and fundraising”

    Yeah, those contributing to purchase access are inclined to hold off to insure their investment is secure.

  6. The Landmark numbers were right, and to the extent they under-counted Republicans the wave nature of the election probably accounts for that. Their earlier more positive numbers for Democrats show what could have been, before the elections here were nationalized.

  7. Jon Lester says:

    There are other and greater reasons to recommend against re-electing Padgett to state GOP chair.

    As for the polling (a separate matter from the why of people’s votes, of course), one factor I haven’t seen mentioned much is that some folks just don’t like to be polled, no matter how much you assure them of its anonymity, and they might say anything because of it.

  8. rwlee2 says:

    Can’t we just have a good conversation about what polling is and what it isn’t? Polls are merely statistical snapshots at a particular point in time — taken alone, they’re pretty meaningless. However, taken together from all the various sources they show a trend.

    I don’t care what anyone says, the trend showed Perdue with some vulnerability at one point, but at the end the trends showed he’d fought that off. The trends also showed that Deal would win without a runoff. We have this hyper-consumerism of polling data for people that aren’t being properly told the story of what it means.

    It’s like selling someone a shotgun cause it’s cool to have a shotgun. Might wanna spend some time telling them how to use it, though…

Comments are closed.