Obama's Numbers Drop...Quick, Cook a Poll!

This time it's NBC/Wall Street Journal. (I know, I can't wait for someone to argue that it can't be biased because the Journal is fairly conservative.) The poll of Registered Voters (a surefire way to get a Democrat a couple extra points, as we've discussed before) claims that if the election was held today, President Obama would receive 49% of the vote while Governor Mitt Romney would get 43% of the vote.

This is questionable if only based on the rest of the Real Clear Politics average, minus the poll we're analyzing, shows a tie of 46% each; a straight average of the four registered voters polls minus the poll we're analyzing shows Obama at 46% and Romney at 45%; and a straight average of the two likely voters polls shows Romney at 47% and Obama at 46%. So a 6% swing in Obama's favor begs the question of the poll's sample.

And well it should, because this poll's sample is as follows:

Including Independent leaners, 45% Democrat, 35% Republican and (presumably) 20% true independents.  This includes 24% of those polled who consider themselves "Strong Democrats" and only 16% who consider themselves "Strong Republicans." Considering the poll asked people about their personal affiliation and not their party registration, this is absolutely outrageous.

I've had people ask me how I can call this outrageous, because it's "just a difference in the way they create their samples...but valid to the voting block in one way or another." This claim does not hold water for one reason: NBC is trying to give a prediction of the 2012 Election, which means they ought to be using a sample that will closely line up with the actual electorate.

To claim that only 35% of the voters who will turn out in November will be Republican is simply ridiculous. This is an election where Republicans are fired up, but NBC wants you to believe these voters will stay home and just let Obama win? That goes against every ounce of logic in this election. It simply does not add up. Especially when you consider that, according to my favorite statistical poll, 40% of Americans consider themselves conservative and that doesn't include the very reasonable assumption that at least a goodly amount of the 35% of Americans who consider themselves moderate are also Republicans. So to assume that Republican voters are just going to stay home is preposterous.

The reality is the same as always: Obama's poll numbers look bad when you look at a sample that actually reflects the electorate, so let's make up a phantom electorate that we WISH was the electorate (even though not even the big victories in 2008 present that low of a Republican turnout, and 2010 of course is a far more accurate sample given that it was more recent than 2008).  So if I was Mitt Romney, I'd see this poll and be in a very good mood.

On a final note: I will be on vacation the entirety of next week, so I will be posting "Best of Biblical Conservatism posts for the week. Enjoy!"