Left-wing bloggers and activists declared war on an independent pollster, who dared to pick up on Romney‚??s positive momentum, both nationally and in the states.
‚??I don‚??t run fake polls,‚?Ě said Douglas J. Kaplan, who owns Winter Springs, Fla.-based Gravis Marketing, and who has been polling and releasing polls on the presidential election.
‚??It is ridiculous,‚?Ě he said. ‚??We have no power. We put out polls.‚?Ě
‚??They are trying to shut me down because they think I‚??m a pro-Romney pollster,‚?Ě he said. ‚??They said Karl Rove was funding me, I wish I knew Karl Rove.‚?Ě
This is not a victimless crime. Anybody who bought into Gravis Marketing and their status as a pollster is going to have their reputations tarnished.¬† Bloggers on the right that embraced every poll (and hundreds did) will be tarnished with the brush of being ‘gullible’….There is a straight line from Nixon’s plumbers to James O’Keefe. It appears, members of the jury, that we found a potential new O’Keefe in the making.¬† For that reason we are not going to simply discredit Gravis Marketing the polling company but we are going to use the disinfectant of transparency to expose Douglas Kaplan and all of his known associates. We are not going to stop when we have made our point. We are not going to stop when we have achieved absolute agreement by all observers that we have found a low level con trying to cheapen our political process.
Carmen Russell-Sluchansky, the host of the Washington-based radio show ‚??Capitol Affairs,‚?Ě said the war on Kaplan came as a complete surprise. Kaplan has been a regular guest on the program, which in addition to being carried by other stations, is also carried on Voice of Russia radio service.
Kaplan is not a part of any Republican or conservative scheme to manipulate polls or news about polls, said Russell-Sluchansky, who is a professional political consultant and pollster, who works exclusively on campaigns for progressive Democrats. Kaplan and Russell-Sluchkansky have partnered on election polling this cycle.
Kaplan said the attacks against him are personal and not just against his business.
In addition to smears online, for a 15-minute period Oct. 15, thousands of emails and comments were sent to his website in an attempt to shut down the site, he said.
Kaplan owns an auto-call facility and created a get-out-the-vote software program used by both Democrats and Republicans, he said.
The pollster entered into his current occupation when national name-brand pollsters contracted him to make their calls, he said. Looking at their mark-up on his calls, he decided to ask his own questions‚??making the transition from wholesale to retail.
‚??I hired my own statistician and started putting out my own polls,‚?Ě he said.
‚??Some of the first polls were not great,‚?Ě he said.
Kaplan said as he learned the polling business from scratch, he recognized that he needed help. He now has a political adviser who works in Washington, along with other personnel to help facilitate the polls.
Business was good and in addition to polling for clients, his polls became part of the rolling average of polls used by Real Clear Politics and were quoted by the Huffington Post and Nate Silver in The New York Times ‚??FiveThirtyEight‚?Ě column, he said.
While Kaplan is targeted as a gear in the GOP machine, one of his clients is the very partisan former Democratic congressman Alan M. Grayson, who is campaigning to return to Capitol Hill.
Everything was fine until September, when a Kaplan poll showed an advantage to Romney in Virginia right after the Democratic National Convention, he said.
‚??All of a sudden, the left-wing, they just went out to destroy me,‚?Ě he said.
Remember, Gravis was making the move from wholesale to retail, so nobody really knew who they were, except the pollsters and consultants who hired them.
The first serious mention of Gravis as a Republican front operation came Sept. 7 on the financial news blog ‚??AngryBear.com.‚?Ě The posting was by ‚??Robert,‚?Ě who was expanding on previous posts about polling and how polls by Republican-leaning pollsters need to be calibrated to remove the ‚??house effect.‚?Ě
Robert posted a link to the Aug. 29 FiveThirtyEight column, where Silver points out a Republican bias in Gravis polls. Silver’s case in point is the Gravis Ohio poll that showed Obama‚??s lead at less than one full percentage point. Taking into account that Gravis at the time showed leads for Romney in Florida, Colorado and Missouri beyond what the consensus of other polls showed, Silver argued that Obama‚??s one point advantage translated into a four or five point lead. ‚??Thus, Mr. Obama‚??s chances of winning Ohio rose somewhat based on this survey.‚?Ě
It is worth pointing out that the headline on Silver‚??s Oct. 19 column was: ‚??After Romney Gains, Should Obama Concede Florida?‚?Ě and the while most October polls of Ohio show Obama with a one to four-point lead, the Gravis poll for Oct. 18-19 has the Buckeye State tied.
Until the progressive activists got involved, the online discussion about Gravis was within the context of methods and talk of a GOP bias was just ‚??house effect‚?Ě that all pollsters deal with to one degree or another.
On the progressive website DemocraticUnderground.com, “DemocratSinceBirth” posted the following Sept. 27 in the article ‚??This Gravis Marketing Clown Is A Real Piece Of Work‚?Ě: ‚??He puts out a bunch of Republican leaning polls in the swing states and then puts out a poll for Washington that shows the president with a huge lead. Of course Washington isn’t a swing state and has never been seriously contested by Romney.‚?Ě
Then, again on DU,¬†grantcart’s post Oct. 7 began an attack with an expose. First, he compared the stock photo of a professional-looking man in his 40s to Kaplan‚??s own photo. Then, he pointed out that Kaplan had just showed up on the scene and had no professional or academic credentials to be a pollster.
grantcart also dug up records of business disputes between Kaplan and other companies, including Disney, who apparently sued him for using the word ‚??magic‚?Ě in the name of his travel promotion company.
Beyond the muckracking, there was a shift in tone with grantcart from the discussion with AngryBear, where Kaplan and a few others hashed out some of their differences in a comments section of the posting.
‚??So Nate, Huffington Post, Real Clear Politics have all been punked,‚?Ě grantcart wrote.
‚??Doug probably hasn’t made any money off of it yet but next time around the Republicans will be happy to throw some bucks his way to get some good PR numbers,‚?Ě he said.
‚??I know Gravis is a fraud,‚?Ě he said.
‚??The media can pick up this work and get a story out of it,‚?Ě he said. ‚??Now Doug Kaplan is going to scream like a stuck pig.‚?Ě
The DU posts were cross-posted at TheDailyKos.com.
grantcart posted at DU again Oct. 15 and the next day TheDailyKos.com ran its own summary of the case: ‚??UPDATED Gravis Marketing Exposed AND Eviscerated‚?Ě by AnnetteK.
AnnetteK includes new information including a series of IRS liens for payroll taxes from Kaplan‚??s previous businesses.
For those new to the story, AnnetteK explained that Kaplan runs Gravis Marketing, which she said publishes tilted poll results. ‚??This has helped to cause the narrative since the 1st debate in Romney‚??s direction.‚?Ě
It does not end.
Here are examples of tweets sent by Kaplan‚??s cyberbullies:
@JedStryke: .@GravisMarketing – THE FIRST BOMBSHELL UNWINDING THE WEB OF ILLICIT ACTIVITES BY AMERICAS GREATEST POLLSTER http://t.co/5BDLuvNz #Obama2012
@JedStryke: .@GravisMarketing You are being investigated and someone’s even been to your office: http://t.co/m1UlJs0f @demunderground #GOP #tcot #Obama
@JedStryke: .@GravisMarketing you know you’re being disected on @demunderground, right? http://t.co/NNiby6Bi #GOP2012 #Obama2012 #Dems #GOP
@DKElections: #FL09 Why is Dem Alan Grayson using sketchball Republican marketing firm Gravis to conduct his polling? http://t.co/1AHoZScG
@galacticasitrep: Daily Kos: UPDATED Gravis Marketing Exposed AND Eviscerated http://t.co/o3JgugA0 via @dailykos
@OleHippieChick: Why is #Gravis credible, Nate? @fivethirtyeight Daily Kos: Gravis Marketing Exposed AND Eviscerated http://t.co/x8FEkwfU via @dailykos
@pollbuster: Gravis Marketing Exposed AND Eviscerated http://t.co/uo5rNiFg via @dailykos
@RosannaLopez101: @fivethirtyeight Nate, one potential problem in your forecast: you keep including Gravis Marketing polls. They are a GOP-front organization.
@josephowusu: list of¬† fraudulent rightwing pollsters: Suffolk, Rasmussen, Gravis Marketing, WeAskAmerica, ARG, Susquehanna, McLaughlin/ACU (R).
Now, fast forward to Oct. 19, when grantcart posted Kaplan‚??s mugshots from 1998 and 2002 for drunk driving. The posting also included the Oct. 27, 2011 mugshot of Alan Giandinieri, an account executive with Gravis, for drug possession.
Kaplan said he has seen nothing like it.
‚??They said my poll was faked, that‚??s been debunked,‚?Ě he said. ‚??They called my family members, my friends‚??people I haven‚??t talked to in years. They‚??re trying to destroy my friends, any of my associations‚??it‚??s unbelievable.‚?Ě
Understandably, Kaplan said he did not want to get into the details of his past transgressions.
The IRS liens have been addressed and he said he is making regular payments, which have reduced his obligation to less than 5 percent of the original amount.
For the rest, he only said that he was younger and made mistakes. ‚??I take care of my family and I had businesses that failed.‚?Ě
Having said all that, the fact remains that the Gravis polls pointed out trends before other polls, he said.
There were two tracks for the attacks on Kaplan and Gravis Marketing. The first was personal. This amounted to public record searches for problems Kaplan and his employees had with law enforcement, other businesses and the IRS. While certainly awkward, nothing in the fire hose of embarrassing information the attackers doused on him helped anyone better understand his polls.
Kaplan could be both a drunk driver in 1998 and still have good polls‚??polls that caught the country‚??s pro-Romney move a month before it showed up in other polls.
The second line of attack is on the methods and legitimacy of the polls.
Kaplan said himself he has no training or experience in polling. But, you do not have to know how to make meatballs to own a restaurant.
The most direct criticism of Gravis polls was posted Oct. 14 on the polling blog logarchism.com by dawolf. In the post, dawolf points out a Sept. 29 to 30 Florida poll. The poll surveyed 914 likely voters, so each respondent was worth .1094 percent of the total, dawolf said. Yet, on page 12 of the poll‚??s crosstabs, there is a result entry: .07 percent.
Next, at 4:47 p.m. Oct. 14 dawolf sent his posting to Princeton‚??s Prof. Sam Wang, who runs the Princeton Election Consortium website:
Sam ‚?? this is something I put together. I believe its solid, but would welcome your thoughts (and, of course, if it is fraudulent, it stops you using them ūüôā )
Wang replied at 4:59 p.m.: I actually think they‚??re on the level. They could get their results by bad weighting.
Michael Weiss, the editor-in-chief of logarchism.com, promised readers his own review of Gravis Marketing polls based on his own interview with Kaplan. It has not been posted as of our deadline.
Kaplan said based on his conversation with Weiss, the editor will validate that the polls are real.
The key difference in Gravis polls is the methodology, he said. Pollsters have to choose a model for their poll that best anticipates the demographics of this year‚??s voter turnout.
Most are using a combination of turnout from 2004 and 2008, the last two presidential elections, he said. Gravis uses 2010 and 2008, which is unusual because mid-term elections tend to have lower overall turnout.
Despite the attacks, Kaplan said Silver continues to quote his polls, including the Oct. 19 national Gravis poll that showed Romney up by two points.
‚??At Real Clear Politics they told me it was part of the game,‚?Ě he said. RCP continues to use Gravis polls, too.