Recently a site called Datatech Guy's Blog examined election data and came to this conclusion: unless Democrats have a greater than 6% edge on voter identification, they lose national elections. He shows data back to 2004 which demonstrates this effect, and he makes a solid case. Whether that is true further back than 2004 I do not know, but what I do know is that there's no way Democrats have a 6% lead in voter identification this year.In fact, the Rassmussen polling agency (the guys that got 3 of the last 4 elections closest to their real results) has Republicans up by 4%, which may be true. Certainly the data I've seen shows that people are abandoning their party membership for independence, but more are dumping the Democratic party than Republican. I have also read several stories which indicate more people self identify as Republican than Democrat this year.
Polling is a fixation during major elections, especially as the days get closer to the actual vote. Generally speaking pollsters are a bit more sloppy and willing to give the results their customers want to see further out from an election, but tighten up significantly in the last few weeks for more reasonable results. The cause of this is simple: a polling agency is a business that survives by being more right than wrong, and while they can be used to promote a certain less-than-accurate picture for a while, they cannot keep getting clients if they are way off-base when the actual vote takes place.
But honestly I don't care for polls. They are at best only useful for examining long term trends, and they tend to be presented to get a result rather than to give the best data. I'll give you two examples.
Recently the Public Policy Polling company put out a poll which showed President Obama leading Mitt Romney in the presidential race in North Carolina by 1 point; 49-48. But to get that result they gave Democrats a 13% edge - they said Democrats led Republicans by 13% of voter identification.
CBS recently did a poll which showed President Obama leading among registered by 3 points 51-49, but they also used a 13% Democratic party edge.
According to reports, pollsters are also heavily weighting polls toward female and single voters, downplaying male and married results. This also tends to give Obama a slight edge, but you'll not even those efforts aren't very optimistic for the president's chance.
But none of that is why I think Obama will lose in November. The first bit I've covered before; Obama only got 53% of the vote in a year with massive opposition to Republicans, by fooling people into thinking he was different and not the usual politician, with incredibly fawning, worshipful media coverage, and he was considered the first real chance for a black president, which people - including me - wanted to see happen. None of that's true this time around, plus he's got a horrible economy and a miserable performance in office.
But the main reasons I see Obama losing are a bit more obscure. I can't think of a clever name for the effect, so I'll just list a few of its indicators.
Major corporate and fat cat donors are giving to Romney more than Obama. Of course, you argue, he's a Republican, and they all stick together! Except, that's not true. When Clinton was running for the presidency, he got the bulk of the fat cat money. When Obama ran in 2008, he got the corporate dollars. Big shots donate not to parties or cronies when it comes to the presidency, but to who they think is going to win. Its about buying access and favors, not helping out a buddy. Its strictly a business investment, and they think Romney will win. They're always right. Never in my lifetime has this indicator gone to the loser, not once.
Another indicator, a bit less obvious, is the social media effect. In 2008, the social media was full of glowing worship and adoration of Obama. You couldn't whisper something negative without a full frontal assault of bitterness and rage. These days, people giggle at the guy and don't disagree much with criticism, and that's across the board on all platforms.
The same thing holds true for another related, less obvious metric. Non political sites like IziSmile and I Can Haz Cheezburger were filled with pro-Obama stuff and bitter rage against Republicans in 2008. You couldn't criticize Obama then, and nobody, anywhere did it, even on these non political places. These days? Its all over the place, mocking Obama, even some pro-Romney stuff is up. Of the top 10 best sellers on Zazzle.com (a tee shirt, mug, etc home made site), 8 of them are pro-Romney. That was all Obama in 2008.
People point to the old indicators like polling and yard signs and bumper stickers, who is getting the most ads out, what gaffe will affect which group and so on. I think they're missing the point. Few people even watch ads any more. When was the last time you heard of a single person who was swayed to vote by yard signs? There really aren't that many out any more, and few bumper stickers except leftovers from previous elections. It seems like the same ten guys buy all the bumper stickers and cover their car like an old steamer trunk, not the rest of us.
Its little stuff like this, and they're all over like the last few Newsweek covers, that show the country is tilting away from Obama. As I wrote a few days back, elections don't happen because of advertising, buttons, yard signs, and bumper stickers. They happen because of a general mood, a cultural shift. Presidents are elected not on policy so much as just the zeitgeist, the spirit of a given time period. McCain had no chance, no matter what he did. It was just the mood of the country.
Its kind of like how protests and guys holding signs don't change a thing. But they may represent what the country wants and is working to change, and are the visible edge of that mood - like the Tea Party Movement was in 2010.
Obama could still win, but I just don't see it.