In the last few weeks, political strategists and prognosticators diligently followed pre-election polls, and many correctly predicted Republicans would reclaim control of the Senate, but most significantly underestimated the breadth of GOP midterm victories.
FiveThirtyEight's Nate Silver, who accurately forecast Obama's overwhelming victory in 2012, estimates that the pre-election polls in the 10 most competitive Senate races had a 6 point Democrat bias.
"The bias might narrow slightly as more votes are counted; late-counted votes tend to be Democratic in most states.
"Still, this is a big 'skew,' and it comes on the heels of what had been a fairly substantial bias in the opposite direction in 2012,"
posted Silver just after midnight.
The polls, he added, "are having some problems."
According to Silver, there was a smaller "bias" toward Democrats of 2 percentage points in the pre-election gubernatorial polls.
One of the elections Silver got wrong was the
Maryland gubernatorial race between Republican Larry Hogan and Democratic Lt. Gov. Anthony Brown, whom he predicted had a 94 percent chance of winning.
He was not alone. According to the
RealClearPolitics average of polls in the final weeks of the campaign, all but one showed Brown winning.
The Republican polling firm,
WPA Research, was the only survey which accurately predicted Hogan would take the governorship.
Hogan ended up defeating Brown 52 percent to 46 percent,
according to Politico.
In Iowa, however, the final Des Moines Register poll was right, but its findings were largely dismissed by Democrats.
On Saturday, the Des Moines Register released its
final survey, which showed the GOP's Joni Ernst with a 7-point lead over her rival, Democrat Rep. Bruce Braley.
Democrats and the Braley campaign were quick to dismiss the survey as an "outlier" that did not reflect their own internal polls.
"That’s not the number we are seeing on the ground," state party Chairman Scott Brennan
told MSNBC.
Retiring Sen. Tom Harkin, who infamously counseled voters not to vote for Ernst just because she was "attractive," also downplayed the Des Moines Register poll, saying "I’ve been in politics long enough, folks, and in enough races to know that there’s always one poll — they just catch people on the wrong night, or they just don’t do their due diligence in their calling, and we have an outlier. This is one of the outlier polls. It just doesn’t fit into anything we know."
By the end of the evening, The Associated Press
was calling the race with Ernst holding a 51 percent to 45.1 percent margin of victory, just a 1-point difference from the Des Moines Register results.
Responding to Democrats' claims that the pre-election polls were underestimating their party's strength, Sean Trende of
RealClearPolitics said biases toward Republicans in 2012 did not necessarily mean Democrats would retain control of the Senate.
"The only thing we can say with any sort of certainty is that in some years the polls have had a slightly pro-Republican skew, while in other years they have had a slightly Democratic skew, while in still other years they've had no skew," he wrote.
Polling, however, might have been
the biggest loser of the night, says Breitbart's Joel Pollack.
"Regardless, it was a very poor night for polling overall — and not just for Democratic pollsters. Conventional wisdom was that the individual Democrats were doing better than President Barack Obama. The story of how that failed should be fascinating," says Pollack.