Chris – I expect to be live blogging the debate Friday, unless I am asked to moderate the debate on campus that night. (Presumably I can do both).

Regarding the impact of previous polling on current polling – that’s an issue that many people have raised. Are voters’ preferences influenced by media coverage and/or polling that suggests one candidate is doing better (or worse) than expected? I know of no hard data on this question (anyone out there have any studies on this?) But my gut reaction is that the impact of polls is probably less than one might think for the simple reason that most voters don’t pay much attention to polling. Remember as well that fully 80% of voters already know for whom they will vote. Of course, in a close election, it is possible that polling may affect the outcome, but I don’t have any data suggesting this is true.

I greatly enjoy reading your blog!! The comments are also interesting, I can pick out the ones who have had you before and who haven’t!! I was just wondering if you were planning on doing a running commentary on the upcoming debate? Also, are you going to be doing the election night thing at the Grille this year? I know its your vacation, ehem…hmmm… sabbatical year, so I thought I’d ask.

As for a few comments on your analysis… With the economy turning the way it has been (I believe it was the Times or USA Today whose headline read “Greatest bailout since the Great Depression) can you see either of these candidates trying to emulate FDR? I believe we have seen Obama multiple times envoke the historical memory of Lincoln. So do you think this is possible? As I was typing I thought of the acceptance speech of Obama, which was like an acceptance speech of FDR correct? At least in the fact it was outside in a large stadium? But I thought that this type of connection might particularly interest you, if you believe the candidates will try.

Also, and I’m sure you’ve gotten this alot, but we’ve heard the others comments and predictions, what about yours?

Lastly, one of the things you talk about with polls often is that more than potential winner come November they measure who the public feels is getting the best of the media now? At least, I hope thats right. With that being said, and your discussion of the vast amount of polling data available, is it possible to begin to believe that the polling data gathered has been impacted by previous polls? Meaning that Candidate A’s percentage will continue to climb because previous polls show that he is doing well in the public perception?

Thanks again. We’ll have to gather over good scotch and continue these and more discussions!

Thanks for the insightful reply to our questions – mine in particular. It’s quite clear: the models assume that campaigns are going to do what they usually do right now, which is frame the election. (Or try to.)

After chewing on this a bit, my thoughts turned to how and where campaigns try to frame elections, and the debates seem like the best opportunity to do so with a national audience. Can an extremely poor (or good) debate performance affect a campaign’s ability to successfully frame an election? If so, can it affect a given model’s ability to measure “reality”?

(Also, do you read fivethirtyeight.com?? Your opinion?)