PoliticsIf you've been following Pollster.com or FiveThirtyEight, you probably think that Barack Obama is a shoe-in. On the one hand, Kathryn Lopez's cherry-picking of polls in The Corner is not impressive. But, I've found some data which strongly suggests that there is a major systematic bias and flaw in all the polls which is grossly overestimating the support for Obama. My results and analysis are below....
BOOOHHHH!!!!!!!!!!!!!!!!!!!
SHOCK POST!
Sorry, I just like scaring people, and there are many spooked & nervous liberals & Democrats right now. Too tempting & too easy....
Update: If you're stupid, you might find this persuasive, Turners.
Comments
note to self: stop clicking on your posts on culture 11...
Posted by: yupsterism | November 3, 2008 10:37 AM
It isn't beyond the bounds of possibility that the polls are systematically biased towards Obama if their sampling of independents doesn't take into account that this time around a substantial number of them are ex-Republicans who were just fed up of Bush and the congressional GOP.
I'm not saying that it is the case or that it will be enough to close Obama's 6 point advantage even if it were true.
Posted by: Ross | November 3, 2008 10:37 AM
Cruel. Very cruel. But funny!
Posted by: ctenotrish | November 3, 2008 11:08 AM
I said to myself: April first already?
Posted by: John Emerson | November 3, 2008 11:51 AM
Before we get too complacent, let's recall that no one predicted the Reagan landslide in 1980.
http://www.time.com/time/printout/0,8816,924541,00.html
Posted by: atyl | November 3, 2008 1:00 PM
Before we get too complacent, let's recall that no one predicted the Reagan landslide in 1980.
history is important, but it needs to be strongly modulated by context and noting differences in the "moving parts." analogies which are only superficially similar simply mislead. e.g., partisan polarity has increased a lot since 1980, there are many more polls than there were then, and the information about these polls flows really freely. finally, remember john anderson. and so forth. obama could lose. but the volatility of the electorate has decreased since the late 1970s (the post-60s realignment arguably didn't finish until the '94 elections) and polling has gotten more robust.
p.s., and reagan really caught in the final 72 hours from what i recall. we aren't seeing that right now, there are 24 hours left....
Posted by: razib | November 3, 2008 1:10 PM
such a dick.
okay, i laughed...
Posted by: tevebaugh | November 3, 2008 6:21 PM
Reagan landslide in 1980
If 51:41 is a landslide then a landslide is withing a reach for Obama.
Posted by: DK | November 3, 2008 7:41 PM
not really. look at the map. john anderson + less regional skew in party strength = substantively different dynamic. analogy: clinton wouldn't have won montana in '92 if not for perot.
Posted by: razib | November 3, 2008 8:09 PM