Contrary to the many public opinion polls that showed Democrats and Republicans deadlocked heading into Election Day, most internal campaign surveys were correctly forecasting the GOP rout.

In interviews Thursday, strategists on both sides of the aisle told the Washington Examiner they weren’t surprised by the GOP’s Senate takeover, padding of its House majority and other gains made up and down the ticket in red and blue territory across the country.

Properly predicting the correct partisan and demographic turnout model was the difference. Campaigns and party committees got it right, while many, though not all, of the public polls were wrong.

A senior Republican strategist who helped direct the GOP’s midterm strategy speculated that media outlets weren’t willing to spend enough on their polls to produce a product that was as detailed as the private, more accurate partisan surveys. This Republican also believes that public pollsters in general were “vulnerable and susceptible” to the arguments Democrats were making about expanding the electorate, given what the party achieved in 2012.

“The Democrats kept telling us they had a gold-plated ground game,” added Republican pollster Wes Anderson, who worked on Sen.-Elect Tom Cotton’s race. “People kept weighting up Democratic turnout.”

The final average in the Arkansas Senate race showed Cotton, a sitting Republican congressman, leading Democratic Sen. Mark Pryor 48.2 percent to 41.2 percent, a margin of 7 percentage points. Cotton won the contest by 17 percentage points, defeating Pryor 56.5 percent to 39.5 percent.

It wasn’t just the public pollsters who assumed higher Democratic turnout in their 2014 surveys, however.

In 2012, Republican strategists approached Election Day confident that their party would capture the White House and pickup seats in Congress, based on data from internal GOP campaign polls. Just the opposite happened, with Republicans realizing that they misjudged the size and demographic of the electorate. This time around, Republicans took seriously the Democrats’ strategy to expand the midterm electorate.

In private conversations, Republican strategists working targeted House and Senate races often revealed that their own surveys showed a closer race than what was suggested by the public data. Democrats often accused Republicans of being overly optimistic about their prospects because they were under-sampling women, Hispanics, African-Americans and younger voters. In fact, Republicans were weighting up those demographic samples that tend to support Democrats, and making strategic decisions accordingly.

But in the homestretch of the campaign, Republicans started to notice that the voter data scores were revealing a crucial dynamic. The most likely Republican voters were also among the most interested in the upcoming elections, while the most likely Democratic voters were much less interested. On the eve of the midterms, previously cautious Republicans grew bullish. Democrats, whose data was just as impeccable as two years ago, saw the shellacking coming, too.

“What ended up happening very simply was that due to the national wave, all of the undecided voters broke for the Republican candidates,” said a Democratic strategist with regular access to internal polling.

On Election Day, the Republicans gained seven Senate seats, were leading in an eighth race (Alaska) and were positioned to win a December runoff (Louisiana) that would net them their ninth. The GOP won a net of 12 Democratic House seats, and flipped three governor’s mansions, including one in progressive Maryland and another in President Obama’s home state of Illinois.

The polling averages, and most of the surveys within them, did forecast substantial Republican gains in the midterms. But as in Arkansas, much of the public data understated GOP performance. In Kansas, GOP Sen. Pat Roberts trailed independent Greg Orman by 1 point in the final RealClearPolitics average; Roberts won by 11. In Kentucky, Senate Minority Leader Mitch McConnell led Democrat Alison Lundergan Grimes by 7 points in the final average; he won by 16.

Sean Trende, the senior elections analyst for, said there’s no clear answer, yet, for why some of public polls misfired. But Trende noted that the surveys seemed to exist in two clusters — one that predicted a Republican tsunami, and another that suggested less drastic Democratic losses. Some races, like Virginia, where Republican Ed Gillespie nearly pulled off an upset of Democratic Sen. Mark Warner, could have benefited from more polling, Trende said.

“I was not surprised by what happened in Virginia, because Mark Warner was polling at 50 percent all cycle, and if you compare poll to poll, it was clear that Ed Gillespie was getting every last undecided voter,” he said. “But they thought it was over, so [many public pollsters] stopped polling and didn’t get to see Gillespie close the rest of the gap.”

The contest, still yet to be called, shows Warner narrowly leading Gillespie 49.2 to 48.4 percent. The final RealClearPolitics average showed the incumbent leading 48.5 to 38.8 percent.

Qunnipiac University produced some of the most accurate public polls of the 2014 election cycle.

But one, the final Quinnipiac survey of the Iowa Senate race, was way off, and contradicted the Des Moines Register poll considered the gold standard for gauging Hawkeye State campaigns. The final Register poll, by Selzer & Co. of Des Moines, showed now-Sen.-Elect Joni Ernst leading Democratic Rep. Bruce Braley 51 percent to 44 percent. Quinnipiac’s final poll showed the race tied at 47 percent. Ernst won by a margin of 52.2 percent to 43.7 percent.

Peter Brown, assistant director of the Quinnipiac University Poll, noted that all of the undecided voters broke toward the Republicans over the final weekend of the campaign and emphasized that polling isn’t a perfect science. Other than the final Quinnipiac poll of the Iowa Senate race, Brown said, the other surveys released by the university during the final weekend of the campaign were solid.

“We can only assume that this was the outlier,” Brown said of his Iowa poll.