Also Vindicated in the Election: Public Policy Polling

This article is from the archive of our partner .

Update: A revised version of the study puts PPP in third place for accuracy below Ipsos/Reuters and YouGov. Gallup and Rasmussen are tied in fourteenth. 

Obama may have won last night, and so did Nate Silver, but another victor in the election was the Democratic pollster Public Policy Polling. PPP had raised eyebrows shortly before the election with a series of polls that showed Obama leading, but a study from Fordham University political scientist Costas Panagopoulos found it was the most accurate pollster in the election.

PPP's final polls put Obama in the lead in every battleground except North Carolina, Charles Mahtesian of Politico wrote Monday evening. "Against the backdrop of a highly competitive election, the uniformity of those results—and the fact that PPP shows 5- to 6-point spreads in Colorado, Michigan, Ohio, and Pennsylvania—puts the firm out on a ledge," Mahtesian explained. "So when the final state outcomes are known, PPP figures to be one of the election’s biggest winners or losers, poised either to spike the football—or to explain what went wrong." Save Florida where we still don't know the results, their numbers plotted in the map below, from Brett LoGiurato of Business Insider, looks a lot like our map (updated for Alaska).

LoGiurato's map:

Our map:

Of course PPP always had some supporters. On Halloween, former Bush pollster Matthew Dowd tweeted:

here is a prediction, and i have no dog in this fight: ppp will be way more accurate than rasmussen on this race.mark it down.

— Matthew Dowd (@matthewjdowd) November 1, 2012

On Panagopolous' list the Republican-leaning Rasmussen ranked 24th for accuracy on the list of 28 polls, tying with the much-maligned Gallup. But PPP had a lot going against them, Jason Zengerle wrote for New York's Daily Intel: "Public Policy Polling, a small polling firm in North Carolina, conducted a whopping 255 public polls in 2012, and it often seemed like polling skeptics (and even other pollsters) had a bone to pick with each one. This was partly because PPP uses automated dialers. It was also because PPP is a Democratic polling firm." The key to their success? Not underestimating turnout. "We just projected that African-American, Hispanic, and young voter turnout would be as high in 2012 as it was in 2008, and we weighted our polls accordingly," poll director Tom Jensen told Zengerle. "When you look at polls that succeeded and those that failed that was the difference."

PPP also had some jump to their defense following the election. Dave Weigel tweeted:

The much-deried final wave of PPP polls was basically accurate.

— daveweigel (@daveweigel) November 7, 2012

Let's look at some of those final PPP polls. PPP had Obama up by four in Virginia. He won by three. Rasmussen had Romney up by two. In Colorado PPP had Obama up by six. He was up by 4.5. Rasmussen had Romney up by three in the state. In Ohio PPP did overshoot Obama's success putting him up by five: he won by 1.9.

This article is from the archive of our partner The Wire.