Gelman on the Election ctwardy | blog | SARBayes


My Weblogs
SARBayes blog
Prior Analytics
SCI 410 news
Blogger(?)

Other Weblogs
D4H
ISRID blog
Nuit Blanche
ACFR
Statistical Modeling...
Machine Learning Theory
pfh
njh'
Older njh
jpl-rpl
...More coming...


Gelman has a nice quick analysis of election results, suggesting that state-by-state there was overall a uniform blue shift of about 4% (+- 3%) compared to 2004. However, since 2000, the youth vote has become increasingly blue shifted:

And link Nate Silver did well , predicting 52.3% for Obama and 46.2% for McCain (actual appears to be 52.4% and 46.3%), and correctly predicting all but 1 or 2 states. And although correctly predicted, Alaska was way outside the margins of error (+25% vs +12%), though many votes are yet uncounted.

Many comments at 538 echo this response to McCain's concession speech: The McCain that gave this speech tonight would have done well running for President.

Thanks to CrookedTimber for pointing me to Gelman's analysis.




[æ]