2016 Postmortem
Related: About this forumThe YouGov Polls moved the meta margin to 5.1 and The Bayesian to 95% -HMMM
http://election.princeton.edu/writes3000
(4,734 posts)Demsrule86
(68,576 posts)RAFisher
(466 posts)Obama won Florida. I'd say 538 did a better job then. Also on election day 538 was closer to the actual popular vote than Princeton.
http://election.princeton.edu/2012/11/06/presidential-prediction-2012-final/
http://fivethirtyeight.blogs.nytimes.com/fivethirtyeights-2012-forecast/
https://en.wikipedia.org/wiki/United_States_presidential_election,_2012
Demsrule86
(68,576 posts)"Neuroscientist Sam Wang at the Princeton Election Consortium made a bold call last night.
Wang indicated that his model found that Obama would close with 332 Electoral votes.
Wang also called the following popular vote spread:
Final predicted popular-vote margin: Obama +2.2 +/- 1.0%.
Two-candidate vote share: Obama 51.1%, Romney 48.9%.
This morning, what did we find? Well, if Florida goes to Obama, he'll close out with a strong showing of 332 electoral votes. And as for the popular vote, here's what Wang says in his "morning after" post:The popular vote is Obama 51.1% to Romney 48.9%. Again, this exactly matches my prediction, which was state polls with a little Bayesian help from national polls."
http://www.businessinsider.com/meet-sam-wang-the-neuroscientist-who-beat-even-nate-silver-in-his-election-prediction-2012-11
Nate's model was not as accurate: 313 Obama and 225 Romney. Wang was also right about the popular vote.
http://fivethirtyeight.blogs.nytimes.com/fivethirtyeights-2012-forecast/
Coyotl
(15,262 posts)oberliner
(58,724 posts)The electoral map is definitely a favorable one.
We just need to hold the Obama states.