2016 Postmortem
Related: About this forumPrinceton Election Consortium projects 69% probability of Dem House takeover
These guys are not as polished as Nate Silver, and I'm leary of relying on generic party polling on a national scale, but for what it's worth...
Caveat #1: The quality of House district data is unlikely to be as good as it was for the last two cycles because polling is down. But perhaps our data source (HuffPost/Pollster.com) or another aggregator will know better. This is a situation where a more complex approach for filling in missing data might help.
...snip...
Considering the possibility of movement in opinion between now and November and the fact that seat outcomes dont precisely reflect popular vote, the total 1-sigma uncertainty is about +/-4%. This leads to a prediction of R+2% to D+6% (68% CI). This corresponds to anywhere from a 12-vote Republican majority to a 36-seat Democratic majority. The probability of retained Republican control is 31%, a knife edge situation.
All of the analysis above comes from data taken before the addition of Paul Ryan to the national ticket. Since then, the two most recent polls show the Democratic lead expanding to 4-8%. If that held, of course it would change the outlook. Ill have more on this as it develops.
http://election.princeton.edu/
TroyD
(4,551 posts)Are they Independent, or do they lean Dem?
brooklynite
(94,350 posts)About the Princeton Election Consortium
This blogs mission is to provide informed analysis of the US 2008 election by members of the Princeton academic community. It is open to scholars in the Princeton area from all disciplines, including (but not restricted to) politics, neuroscience, psychology, computer science, and mathematics.
For now, much of the sites information is about polling. As the campaign season progresses we will expand to other interesting topics, and expect for diverse contributions. To write for us, please contact Sam Wang.
This blog began in 2004 as a meta-analysis directed at the question of who would win the Electoral College. Meta-analysis of state polls provides more objectivity and precision than looking at a single poll and gives an accurate current snapshot of the state of play. Over the course of the campaign, this site attracted over a million visits. In 2004, the median decided-voter calculation on Election Eve captured the exact final outcome (read this article and the follow-up). The 2008 calculation provides results based on decided-voter polling from all 50 states.
The Management
Prof. Sam Wangs academic specialties are biophysics and neuroscience. In these fields he uses probability and statistics to analyze complex experimental data, and has published many papers using these approaches. He is also the author of Welcome To Your Brain, a popular book about his field.
He originally developed the Meta-Analysis in 2004 to help think about how to allocate campaign contributions. He was originally motivated by the fact that in a close race, one can make the biggest difference by donating at the margin, where probabilities for success are 20-80%. The Meta-Analysis has subsequently been found to be useful as a highly sensitive tracking tool over time. To read a discussion click here. In 2004, the Meta-Analysis of State Polls got tens of thousands of hits per day. He was helped that year by Drew Thaler. Sam can be contacted as sswang at princeton dot edu.
Andrew Ferguson is a 2008 Princeton graduate and responsible for the sites data processing and infrastructure. While in college, he studied probability, statistics, and computer science. He is currently a computer science graduate student at Brown University, where he works on Software Defined Networks and frameworks for analyzing Big Data.
freshwest
(53,661 posts)brooklynite
(94,350 posts)...because its not modelable.
Everyone's analysis is based on what people intend to do. Whether they can and do vote is a separate issue.