6 thoughts on “The nonpuzzle of the close opinion polls

  1. Derek,

    Yes, I rank ordered them in the display because I thought that would be the most readable way to present the data in the book. Follow the links to see some alternative versions of the graph.

  2. I noticed that the 2 exceptions (where the incumbent party lost despite good growth) were 1952 and 1968–war years. 2008 is also arguably a war year, and I think that explains part of the 'puzzle' of the close polls.

  3. Steve,

    The Abramowitz et al. article you link to is fine, but it's a bit too deterministic for my taste. It's probably a bad sign that the website it's on is labeled "Larry J. Sabato's Crystal Ball." Larry Sabato knows a lot, but nobody has a crystal ball.

    Obama is forecasted to win, but there's always uncertainty. They cite lots of recent polls but that's not always a perfect guide to what will happen in November.

Comments are closed.