On Election Day 2012, the RealClearPolitics polling average found incumbent President Obama “leading” Republican Mitt Romney by seven-tenths of a point — a statistical dead heat. The actual final result, an Obama victory of 3.9 points, was very different.

Of the nine surveys that were part of that final RCP calculation, three (including CNN’s) found the race tied and two, Gallup and Rasmussen, had Romney up by a point. Only four of the nine surveys showed Obama ahead.

Both the Washington Post-ABC News survey and Pew Research found Obama leading by three points, while the NBC News-Wall Street Journal and Investor’s Business Daily polls showed Obama up by a single point. Not even one survey conducted within a week of Election Day found Obama up by as many as four points.

AD

I cite these numbers not to challenge the RCP average or disparage the pollsters who got the race wrong. Four years earlier, the RealClearPolitics average was spot on. It estimated an Obama win by 7.6 points, almost hitting the actual 7.3-point margin.

AD

Still, the 2012 “miss” should remind us that polling isn’t perfect and even “averaging,” a reasonable strategy, can mislead. Averaging may smooth outliers, but what if one poll always seems to be an outlier, like the Los Angeles Times-USC tracking poll?

Some additional data to consider: Six weeks after the 2012 Democratic National Convention, the NBC News-Wall Street Journal poll showed the Obama-Romney race tied at 47 percent, while the Washington Post-ABC News poll found Obama up by three and the CBS News-New York Times poll found Obama ahead by one. But the Associated Press found Romney up by two points, while Monmouth-Survey USA had the Republican leading by three.

AD

Poll numbers bounce around, and things can get particularly confusing when pollsters switch from registered voters to likely-voter samples, as they are doing now. It’s best, I have often advised, not to get too caught up in the daily release of poll numbers. Of course, even I fail to heed my own advice from time to time.

AD

The problem everyone faces is knowing exactly who will vote.

Fox News’ Sept. 11-14 poll found Hillary Clinton leading Donald Trump, 46 percent to 43 percent, among registered voters in the two-way ballot, down a hair from its late-August survey, which found Clinton leading 48 to 42. But among likely voters, Fox put Trump ahead, 46 percent to 45 percent.

AD

The four-way ballot was different. The mid-September Fox survey showed Clinton leading by four points among registered voters, 41 percent to 37 percent, an improvement from her two-point lead in late August. But among likely voters in the Sept. 11-14 poll, the former secretary of state held a one-point lead over Trump.

Interestingly, the two-way registered-voters ballot confirmed the “race is closing” narrative of the previous 10 days, while the four-way registered-voters ballot refuted that same narrative. Go figure.

AD

So where does the race stand now?

It appears to have closed more than I expected. Clinton has suffered through some recent criticism, while Trump has been on script more than he was initially. The Republican nominee often sounds like a none-too-bright middle-schooler when reading from the teleprompter, but that’s still better than when he simply says whatever pops into his brain.

AD

Trump has had one-point leads in two polls, from CNN and Fox, while Clinton has held leads of two to eight points in other polls conducted over the past two weeks. Clinton’s lead has shrunk from late August, but she continues to have the advantage.

Trump tends to shoot from the hip more when he feels good, as he now does, and the media’s focus may well swing back to the Republican nominee after the last few weeks. Moreover, Clinton and her surrogates seem poised to boost their efforts to energize younger voters, African Americans and Latinos in November.

AD

Trump has been unable to broaden his appeal, which is why he has not grown his support in the minority community or among younger white voters and college-educated whites. Clinton, of course, has her own problems, particularly with white men. But as Obama demonstrated four years ago, if Democrats can turn out their voters, they will win.

AD

Clinton isn’t likely to replicate the Obama coalition, but she doesn’t need to. She has a 3.9-point cushion, and Trump is suffering many more defections than Romney did.

Democrats have more reason to be concerned today than they did five weeks ago. But while the polls have closed, the larger fundamentals still favor Clinton. At this point, she has a much greater opportunity to grow her support than does Trump.

The first debate, on Sept. 26, could be a significant event.

Stuart Rothenberg writes about the politics of the presidential and congressional races.

AD
AD