Good Pollster, Bad Pollster

Ken AshfordElection 2012, Polls1 Comment

Fordham University has published a ranking of the most accurate pollsters of the 2012 in terms of national trends, and (both) top spots were held by Democratic-leaning Public Policy Polling, the North Carolina-based firm.

Director of Fordham's Center for Electoral Politics and Democracy, Dr. Costas Panagopoulos, based the study on pre-election polling and compared it against the results from election day.

"For all the ridicule directed towards pre-election polling, the final poll estimates were not far off from the actual nationwide vote shares for the two candidates," Panagopoulos said in a statement. Here's the whole list:

1. PPP (D)*

1. Daily Kos/SEIU/PPP*

3. YouGov*

4. Ipsos/Reuters*

5. Purple Strategies



6. YouGov/Economist



11. Angus-Reid*

12. ABC/WP*

13. Pew Research*

13. Hartford Courant/UConn*


15. Monmouth/SurveyUSA

15. Politico/GWU/Battleground

15. FOX News

15. Washington Times/JZ Analytics

15. Newsmax/JZ Analytics

15. American Research Group

15. Gravis Marketing

23. Democracy Corps (D)*

24. Rasmussen

24. Gallup

26. NPR

27. National Journal*

28. AP/GfK

It should be noted that Rasmussen, the favorite pollster of Fox News and right wing blogs, was close to the bottom of the list.  They were bad last election cycle, which is why I ignored them this year.

As for poll aggregators, CNET reports that the best one was 538, i.e., Nate Silver. But TPM PollTracker, HuffPost Pollster, the RealClearPolitics Average, and the Princeton Election Consortium — successfully called the election for Obama, and save for TPM PollTracker and RealClearPolitics handing Florida to Romney, the aggregators were spot on across the board when it came to picking swing state victors.

Then, there is this:

Dean Chambers, the man who garnered praise from the right and notoriety on the left for his “Unskewed Polling” site, admitted today that his method was flawed.

 “Nate Silver was right, and I was wrong,” Chambers said in a phone interview.

Chambers’ method of “unskewing” polls involved re-weighting the sample to match what he believed the electorate would look like, in terms of party identification. He thought the electorate would lean more Republican when mainstream pollsters routinely found samples that leaned Democratic.

But as it turned out, the pollsters were right — self-identified Democrats outnumbered Republicans by 6% in election exit polls.

The bigger question is why on earth he ever thought just changing numbers in a poll until you liked the outcome was going to actually reflect the reality on the ground. That’s the power of the bubble working. When every source you hear, whether it be Clown Hall or Faux News or the National Review or whatever else ells you over and over and over that the media is biased or that statistics have a left-wing bias, chumps like Chambers start to really believe it. And then they do predictably laughable stuff like this- just rejiggering polls until the numbers put you in your happy place.

What’s even more astounding than some lone nitwit doing this is all the people who linked to it, repeated this nonsense until significant portions of the right wing believed it. That’s what is really astounding. You can find people doing really stupid things every where you go in the world. You don’t expect their stupidity to be elevated. And there was just no one, no one at all in the GOP feedback loop who stood up and said “Good lord, you people are morons.”