It’s the strangest thing. The media already declared this week that Mitt Romney lost the election, but the polls still seem to show the race tightening. First, from today’s USA Today/Gallup poll, which has Romney trailing Obama by two points in the swing states:
Registered voters in key 2012 election swing states remain closely divided in their presidential vote preferences, with 48% supporting President Barack Obama and 46% Mitt Romney. Other than a nine-point lead for Obama in March, the two candidates have been essentially tied in the swing states throughout the campaign.
Gallup’s daily tracking poll also finds Obama leading Romney by one point nationwide. Note that both of these polls were conducted among registered, not likely voters, which means they are more likely to favor Obama:
Gallup Daily tracking of registered voters nationwide now finds Obama at 47% and Romney at 46%, suggesting a fading of Obama’s post-convention bounce. At this point, it is too early to tell what impact a newly released video of Romney’s unflattering characterization of Obama supporters from an early 2012 fundraising speech might have on the race.
Today’s AP/GfK poll of likely voters has a similar finding. Obama leads Romney by a single point:
Obama is supported by 47 percent of likely voters and Romney by 46 percent, promising an all-out fight to the finish by the two campaigns to gin up enthusiasm among core supporters and dominate get-out-the-vote operations. That’s an area where Obama claimed a strong advantage in 2008 and Republicans reigned four years earlier.
That’s not to say there aren’t signs of trouble for Romney. We still don’t know how the potentially-damaging “47 percent” video will impact the numbers above. He’s down by five points in the NBC/WSJ poll of likely voters, which came out last night but was conducted before the video dropped. He also appears to be losing ground on economic issues, at least according to the AP/GfK poll.
But as the media has been saying for the past couple of months, there aren’t many undecided voters out there to win. The race will largely come down to turnout, and while Romney’s “47 percent” remarks could potentially depress some of his support, the narrative that he’s losing by a landslide (when polls show he’s not) isn’t helpful for Obama either. Much of Obama’s get-out-the-vote efforts rely on frightening his base about the prospect of a Romney presidency. If they falsely believe that Obama has the race locked up, there’s also less incentive to turn out at the polls.
There's no tightening & no looseining: nFor the last 18 weeks,things have been stable for the Electoral College projections: nObama/221 nRomney/191 nBattleground/122 nNo change in this 47% versus 47% dead heat.
The race tightens…around Romney's neck. n nToo bad for Obama that we don't have a system that elects presidents by an electoral college with certain electorals from each state. That would make state polls important and national polls with no relevance to the precise outcome would be irrelevant. n nOh wait…
In finance there's a saying: don't fight the Fed; and, in relation to the stock market, don't fight the tape. Well, in politics we should say: don't fight the polls. Parse as we will, the polls consistently show Obama ahead. To be sure, some are tighter than others, but all show that trend, particularly the state polls of the swing states which are really the only meaningful ones to consider. What's my point with such negativity? We conservatives should be focusing upon saving some of our good people who are running for the House or Senate by contributing to their campaigns.
Sorry, but most of the polls are bunk. The NBC/WSJ poll showing O up 5 oversamples Dems by 5% over their national identification (44-37) and undercounts Reps by 10 (27-37). The AP poll showing an O lead of 1 is split 31 D, 30 R and 30 independent (the balance being "other", whatever that means). Overwhelmingly, the polls in the swing states also are tilted toward Dems and away from Reps. If you believe that the samplings are reflective of what the turnout will be, fine. However, there is every indication that like 2010, turnout among Reps/conservatives will be stronger than Dems, in which case these pollsters have a rude awakening. n nOne also might ask why the polls are skewed. Might it be to depress Rep voters into not bothering? Just asking. n n
That may be the 64000 question. There seems to be too much of dissonance between the polls. Rasmussen has Romney consistently ahead (albeit by a smidgen) and NBC/WSJ gives Obama a big five point margin. What gives? And it seems to be counterintuitive that Obama leads Romney in Virginia yet tralis in Colorado! What gives? Poll are all over the map and they may not be meaningful till two weeks before the election.
Rasmussen counts "likely to vote" instead of Registered voter for its projections. n
Methodologies do indeed differ, as do the polling firms "guesses" as to what the electorate is likely to be on election day. What frustrates me is when we conservatives deny what the bad polls tell us and look for those that please us(Rush and Sean are notorious for this). That's why, in my above post, I argue it's best to look at trends–and, in a way, that's what RealClear Politics does when it gives averages. When we do this, Obama is clearly ahead. Now, if the trend or averages change in favor of Romney, I think we'll have something to crow about. But these major polling firms are not skewed. What one can argue is that their results are hyped when Obama is ahead. This is clear when we don't see Rasmussen too often quoted–the most favorable numbers for Romney—except in Drudge.
None of us are looking for polls 'that please us". We may quote them differently. The dichotomy arises when they are just too dissonant. I will not believe that Virginia will go for Obama- and that George Allen is behind Kaine (Rasmussen!). is it the same Virginia that elected O'Donnell – after Obama was elected!- in a quasi landslide? What makes the electorate turn around and vote for the opposite of what they just voted ?? Virginia is doing well under O'Donnell, where is the reason for this turnaround? I am just using the Virginia polls to illustrate that polls can be so misleading. Because, indeed, if Obama carries Virginia , he will win in a landslide which I just don't believe.
John has a good post related to this issue. As far as Virginia is concerned, more specifically, I do agree a bit that there is some dissonance. Romney is much more behind(3-8 pts.) than Allen(1-2). It hardly makes sense that someone would vote for Obama and turn around and vote for Allen. As far as the near-landslide for McDonnell(a great governor), Virginia holds their state-wide elections in off-years, so turnout is down sharply. So, 2009 and 2012 are not perfectly accurate comparisons. Also in the 2011 elections for the state legislature, the Republicans, led by McDonnell, spent heavily to capture the Senate. They were able to get only a tie, as they kept the House.