Five Thirty-Eight has a problem with Real Clear Politics polling methods. So do we.
Five Thirty-Eight admits their site is not devoid of partisan slant, but they contend that RCP's "partisan leaning is infused into their numbers." That's where Five Thirty-Eight draws the line.
"If RCP disclosed their methodology -- articulated their rationale for excluding or including certain polls -- I would give them the benefit of the doubt. But they do not, so I do not."
http://www.fivethirtyeight.com/2008/09/real-credibility-problems.html
We also have a problem with Stu Rothenberg of the Rothenberg Political Report who shows recent movement in House seats with no analytical or statistical reference.
Where did the movement come from? For instance what is the movement based on for each elected official? Is there a science to it? Or is it all about who has the most in the bank? There is less detail given in this blog entry than other prognostications Rothenberg has made-- let's say for instance the Mass race, which clearly, was moving in Brown's direction for lots of reasons. Rothenberg's prediction on the Mass race was backed up by polls. No eight-ball busting there.
Way back in 2006 David Sirota said " in a world where 98 percent of incumbents get reelected. I mean, really - how difficult can predictions be when all you have to do is say "the incumbent is going to win" in order to be 98% accurate?
David asked the more important question when he said, " The question is: when will the media stop calling (him) Rothenberg "nonpartisan" and start labeling him as a conservative?"
http://www.davidsirota.com/2006/04/stu-rothenberg-cant-hide-his-right.html
Friday, April 16, 2010
Subscribe to:
Posts (Atom)