4 ms·

Whoa, there nelly! First things first, Project Orca (you can read two first hand user accounts here: http://www.breitbart.com/Big-Government/2012/11/08/Orca-Ho

Whoa, there nelly! First things first, Project Orca (you can read two first hand user accounts here: http://www.breitbart.com/Big-Government/2012/11/08/Orca-How-the-Romney-Campaign-Suppressed-Its-Own-Vote http://www.breitbart.com/Big-Government/2012/11/08/Orca-How-... and here: http://ace.mu.nu/archives/334783.php http://ace.mu.nu/archives/334783.php ) appears to be a Get Out The Vote and volunteer management app. Second, 1 and 2 are not incompatible results. Nate Silver and others have pointed out that the polls appear to have underestimate democratic performance at the polls... is that due to Republicans failing to turn out? That seems like a likelihood that's at least worth considering.

ajross 14y ago
I don't believe Nate Silver pointed out that the polls undercounted democrats, do you have a cite? In fact his analysis was basically that the aggregate polling "got it right", coming very close to the final counts in almost all cases. In the four victory-critical states I still have in my head, Silver's polling averages (which weren't very much different from Wang's or TPM's or RCP's, by the way) nailed it dead on in Florida, undercounted Democrats in Colorado and Virginia, and overcounted them in Ohio. But in no case were the errors more than 1.5% or so.
- pflats 14y ago
  He said as much in an interview. I'm pretty sure it was when he was on the Daily Show, but I could be misremembering. http://www.thedailyshow.com/watch/wed-november-7-2012/nate-silver http://www.thedailyshow.com/watch/wed-november-7-2012/nate-s...
  - knowtheory 14y ago
    This was the interview I was recalling, thanks pflats!
albedoa 14y ago
Nate Silver and others have pointed out that the polls appear to have underestimate democratic performance at the polls This is what Nate Silver said on election night: So far, however, Democrats also have an edge in the party identification numbers in the exit polls. In Ohio, 38 percent of voters identified as Democratic in the exit poll as compared with 31 percent of Republicans. And in Virginia, Democrats had a 37-to-33 advantage in party identification. These numbers are similar to what many pre-election polls showed. http://fivethirtyeight.blogs.nytimes.com/2012/11/06/live-blog-the-2012-presidential-election http://fivethirtyeight.blogs.nytimes.com/2012/11/06/live-blo... Can you point to where he adjusted this analysis?

ajross 14y ago

I don't believe Nate Silver pointed out that the polls undercounted democrats, do you have a cite? In fact his analysis was basically that the aggregate polling "got it right", coming very close to the final counts in almost all cases. In the four victory-critical states I still have in my head, Silver's polling averages (which weren't very much different from Wang's or TPM's or RCP's, by the way) nailed it dead on in Florida, undercounted Democrats in Colorado and Virginia, and overcounted them in Ohio. But in no case were the errors more than 1.5% or so.

pflats 14y ago
He said as much in an interview. I'm pretty sure it was when he was on the Daily Show, but I could be misremembering. http://www.thedailyshow.com/watch/wed-november-7-2012/nate-silver http://www.thedailyshow.com/watch/wed-november-7-2012/nate-s...
- knowtheory 14y ago
  This was the interview I was recalling, thanks pflats!

albedoa 14y ago

Nate Silver and others have pointed out that the polls appear to have underestimate democratic performance at the polls This is what Nate Silver said on election night: So far, however, Democrats also have an edge in the party identification numbers in the exit polls. In Ohio, 38 percent of voters identified as Democratic in the exit poll as compared with 31 percent of Republicans. And in Virginia, Democrats had a 37-to-33 advantage in party identification. These numbers are similar to what many pre-election polls showed. http://fivethirtyeight.blogs.nytimes.com/2012/11/06/live-blog-the-2012-presidential-election http://fivethirtyeight.blogs.nytimes.com/2012/11/06/live-blo... Can you point to where he adjusted this analysis?