TimG Posted October 9, 2008 Report Share Posted October 9, 2008 fivethirtyeight.com seems to be doing a pretty good job analyzing the numbers this electoral cycle. We won't really know whether their analysis is any good until we see the actual results, will we? Quote Link to comment Share on other sites More sharing options...
hrothgar Posted October 9, 2008 Report Share Posted October 9, 2008 fivethirtyeight.com seems to be doing a pretty good job analyzing the numbers this electoral cycle. We won't really know whether their analysis is any good until we see the actual results, will we? The proof is definitely in the pudding. At the same time, it is possible to look at the recipe and draw some early conclusions. 538 does a very good job describing the methodologies that they are using. Quote Link to comment Share on other sites More sharing options...
TimG Posted October 9, 2008 Report Share Posted October 9, 2008 Is that the same Nate Silver who developed the PECOTA system for Baseball Prospectus? Interesting that he has taken his projections from sports to politics. Quote Link to comment Share on other sites More sharing options...
Mbodell Posted October 11, 2008 Report Share Posted October 11, 2008 Yes it is the same Nate Silver and yes it is the same sort of methodology. They define landslide as 375+ or more of the EV. And note they aren't modeling what would happen if the election were held today, they are modeling what today's numbers predict about the election in November (which means they regress it so the person trailing does better as they predict all elections are more likely to tighten than widen). And the numbers they have now predict a 375+ landslide a full 35.24% of the time for Obama. Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.