Doping: Marist Juices Numbers to Help Obama
Never in my life have I seen such obvious, blatant and one-sided skewing of media polls. I've noted before that the media have weaponized polling this election, using their voter surveys to buoy the struggling Obama campaign. Marist and Quinnipiac have been the most egregious polling propagandists this cycle. But, in their latest poll of Ohio, Marist has achieved a new depth. It will go to hell for its cardinal polling sins.
Marist's latest survey of OH finds Obama with a comfortable 6-point lead, 51-45, among likely voters. Among independents, however, Romney is leading Obama by 8 points, 49-41. If you reread those two sentences you'll see that something odd is going on.
In Marist's previous survey of OH, conducted before the presidential debate, Obama had an 8 point lead, 51-43 and was leading among independents by 4 points, 47-43. According, then, to Marist's own polling, Romney picked up 12 points among independents, but only picked up 2 points in the overall match-up.
How did Marist achieve this odd result? They packed a lot more Democrats into their poll.
Pre-debate, Marist's poll of OH was D+5, 36% Democrat, 31% Republican. Post-debate, their poll is D+11, 40% Democrat, 29% Republican. Marist would have us believe that, in the immediate wake of Obama's widely panned debate performance, the electorate in OH had a 6-point swing towards the Democrats. Really?
Marist really thinks that Democrats in 2012 will have a larger turnout than in 2008? And that the Democrat surge came in the wake of last week's debate? Apparently, Obama's epic debate fail was just what he needed to make Ohioans flock to the Democrat party.
A quick look at early voting in OH shows the lie in Marist's numbers. In 2008, Democrats had a nearly 15 point advantage over the GOP in early voting. This year, their advantage is down to just 6 points. Republicans have already almost reached their total number of absentee ballots from 2008, while Democrats are just around half the level reached in 2008.
It isn't even remotely defensible to argue that this year the Democrat turnout advantage will eclipse 2008. This is nothing more than pure propaganda.
Immediately after last week's debate, I joked on twitter that the media would have to start deploying D+20 samples to maintain Obama's lead. Marist is well down that road.
Follow me on twitter