The weekly 2016 Electoral College Projection will begin upon the conclusion of the primary season next spring. Until then, here is Race’s final projection for the 2012 campaign:
Since we began the Race42012 electoral map updates back in May, Governor Mitt Romney has never led when there are no tossups. The question as we approached Election Day over the past week, then, was whether or not he would finally take the lead this morning in our final prediction. Our final prediction is this: President Barack Obama will succeed in his bid to be re-elected tomorrow.
On Wednesday or Thursday of last week, I began really thinking through what this final prediction was going to look like – and as I began looking at all the polls, advertising, campaign schedules, and other data available, I began to realize it all pointed in one direction: unless something changed in favor of Governor Romney, President Obama was going to win this race. And since Thursday, things seem to have broken not in favor of Romney, but against him. State polls shifted to Obama. National polls which once showed a 5-7 point Romney lead shrunk to show a tied race. Stops in states like Virginia and Florida were added to Romney’s schedule late when those were states he was already supposed to have sewn up. Perhaps most of all, President Obama’s job approval rating has rebounded to sit at or above 50% in five different surveys (51% in Rasmussen and CNN, 50% in ABC/WaPo, FOXNews, and CBS News). In fact, in every recent national survey, Obama’s job approval rating is now at 49% or higher.
A sitting President with an approval rating of 49% or above has never lost re-election in modern history.
In addition to this, more states are falling further out of Romney’s reach as he tours the country on his closing argument tour. Colorado, which sported an R+1.0 lead among five polls last week, has now fallen to O+1.0 with five new polls. Wisconsin, which was just O+1.0 last week, has now slipped to O+3.5 among seven polls — with Obama hitting the magical 50% mark in the average. In Ohio last week, four polls showed the race at an exact tie. This week, only one does while nine others show Obama taking a 2-6 point lead.
Any momentum Governor Romney had gained from his debate performances throughout October seems to have completely stalled as the calendar page turned to November – and not only stalled, began falling in the wrong direction.
At this point, you may be thinking, “You’re forgetting one key thing: the state polls are no good because of crazy party ID splits.” Normally, I would agree with you. However, we are now less than one day out from Election Day. If pollsters were going to adjust their samples to protect their fragile reputations, this past week would have been the time to do it — and they didn’t.
Think about it this way. In 2004, all of the Democrats were crowing and complaining about the party ID splits in public surveys. There is no way the party ID split could be even, they said. John Kerry will win because all the polls are wrong. But they weren’t all wrong, and Bush won.
In 2008, Republicans were raising cane about the party ID split of public surveys. There is no way the Democrats could have such a huge advantage in party ID split, they said. Barack Obama won’t win (or won’t win by much) because all the polls are wrong. But they weren’t all wrong, and Obama won big.
This campaign cycle, we have spent months upon months ridiculing, scoffing, and complaining about party ID splits. But when it counted most, when the pollsters had it all on the line, nothing changed. That tells me we are in trouble, both short-term for this election, and long-term for the future of this country.
If that doesn’t convince you, look at it another way. In order for Mitt Romney to win Colorado, 4 out of 5 polls taken in the last week have to be flat out wrong. In order for Romney to win Ohio, 9 out of 11 polls have to be wrong. (One shows Ohio tied, and the last one is done by the firm who showed Todd Akin still winning in Missouri.) In order for Romney to win Iowa, 4 out of 6 polls have to be wrong (including the Iowa gold standard DMR poll). In order for Romney to win Wisconsin, 6 out of 7 polls have to be wrong.
I am not in the business of declaring that all the polls are wrong. Not after 2004 and 2008. And guess what — in CO, OH, IA and WI, the poll in each state that was actually tied or showed Romney winning was Rasmussen.
Either Rasmussen is wrong this time around, or everybody else is.
If these were all media pollsters with agendas to drive, that would be one thing. But these include private polling organizations such as SurveyUSA, We Ask America, Pulse Opinion Research (Rasmussen), and Gravis Marketing. These organizations aren’t out to “get” Romney or propel a narrative. And the fact that they are (largely) agreeing with all the media polls does not bode well for tomorrow night’s results
This isn’t the final word on the subject. A surprising GOTV machine and incredibly high turnout for Republicans could end up winning the day after all. And the enthusiasm gap is there that could propel us toward that result. If I am incorrect and Governor Romney ends up winning tomorrow night, here’s how it would happen: Colorado and Iowa go red, and then Romney wins one of either Wisconsin or Ohio. Those are the four most likely states to flip to light red on this final map — and in that order of probability as well.
Updated November 5th, 2012 – Matt Coulter.