How did all that energy end up working out for you?

Of course, no rule in elections is inviolate. But there are patterns as you well know. The one I chose is the last election, because the voters are stile in the same frame of mind, even more so, in my humble opinion.

I am on record as being way out there in deep right field, I know. I assure you, that is not my usual situation. I am usually much more cautious about predictions until a time much closer to the election itself.

But I was active in 1994 and I see the same writing on the wall now that I saw then. I predicted the results of 2010, but I had inside information. The now Majority Whip is my pal, and I knew what he was doing to win the House.

The swing states that have Senate races and hot Congressional races add another dimension. I see the ads, I get all the literature on both sides, and I am seeing an energized Republican base like I haven’t seen since 1994, except in 2010. Obama is pleading for more money and more support. The wind is not at his back as it was in 2008. Nor does he have to enormous monetary advantage he had in 2008.

I may be all wet, but I think after the Republican Convention, Romney will pull away from him and never look back. He only needs to tell the people how he will fix the economy and get them to believe it. He is believable.

Obama can’t match that.

I’m drawing on perhaps faulty memory here, but my recollection of the research is that the late deciders are NOT more informed about policy issues, or the fundamentals more generally. Indeed, I think they are LESS informed, in general, although I need to check this. But again, our forecast models don’t presume deeply knowledgeable voters well versed in the policy platforms pushed by each candidates. All we ask is that they understand some very basic issues, such as which candidate is the incumbent, and who is more liberal/conservative, and whether the economy has improved under the incumbent’s watch or not. And almost all voters can discern these basic differences, and they vote accordingly.

It’s not that they stop being frivolous. It’s that as Election Day looms larger, they begin paying attention to the race and thinking about who they really want to vote for. You have to remember that just because a potential voter gives a sincere answer to a pollster’s survey question in early August, if the respondent’s answer isn’t really based on an informed assessment of the race, it is not as likely to be as stable as an answer based on a careful consideration of the fundamentals. Now, this doesn’t apply to every voter – many are pretty well locked into their vote as early as August by virtue of long-standing and deep-seated partisan predispositions. I’m mostly focusing here on the minority of voters who simply are paying almost no attention to the race and who might have weak partisan attachments.
So, political scientists aren’t really using two different voter models, in which voters first pay attention to media coverage and then later discount it. Instead, our forecasts are based on one model: that of a voter whose knowledge of the election fundamentals gradually increases through the course of the campaign and who votes primarily on the basis of those fundamentals, as interpreted through their preexisting partisan lens.

As a follow-up, do you regard it as a kind of weakness of the standard political science model that it has to posit two different “mechanisms” to explain how people respond in such polls? (This is perhaps a weakness that is trumped by various strengths.)

Here’s what I mean. If you’re right, then during the summer months the people who respond to pollsters are responding to the latest gaffe, or media coverage, or whatever. But then, come early October or so, they stop being so frivolous and put their economic fundamentals hat on, and vote on the basis of 2nd quarter GDP (or whatever). Does that really seem independently plausible, that such a shift occurs?

An apparent advantage of the sort of approach you’re arguing against in this post would seem to be that it can say that the same single mechanism is in play all throughout the polling process. In July, voters are responding to the latest gaffe or media coverage or whatever (just as you concede), and in October the very same thing is going on. The pundits’ view achieves a kind of theoretical parsimony that your view doesn’t. Or is that wrong?