NYT Nate Silver Doubles Down: 75% Chance Obama Win

I think Silver will be right about 49/50 states again, just like 2008. Virginia has way too many absentee red ballots to stay blue. Unfortunately, that won't be enough for Romney to pull it off.
 
As I have learned to say in my job, predictive algorithms are right until they are wrong. When they are wrong, you rarely hear from them again.

I am interested in seeing whether this model holds or not. It has one admittedly very impressive cycle of testing. If it repeats, it will be interesting to see how the various parties react. If it fails, no one should be suprised as model with this many factors and smoothers relying on third party data invariably blow a gasket and pisses everyone off.
 
Pundit accountability: The official 2012 election prediction thread (Washington Post)

Link

From Romney at 325 (Dick Morris, Fox News) to Obama at 440 (Jim Cramer, CNBC), 19 pundit predictions for your perusal

(with links to each)
 
"Perham1, They are in the tank for the Dem party. Why would any rational American take anything they say serious."

It's kind of a leap to say that Nate Silver's statistical prediction method is "in the tank" with the Democratic Party. He uses EVERYONE ELSE'S POLLING for his models, not some sort of left-wing think tank stuff. If his models showed that the GOP was going to win, then he'd report it and blog about it.

Then there's guys like Dick Morris, who just ignores polls and spouts off EV numbers from his gut. If it takes more balls to put yourself out there with an outlandish prediction, so be it. But that doesn't make it "rational."
 
is the entire point of this thread that Nate Silver will lose credibility if his projection is off......well Duh!!

and thus far it looks like his projection is holding
 
You must log in or register to reply here.

Recent Threads

Back
Top