Mondale 2 (and a Bit of Palin)

There's a lot going on that I didn't get to yesterday...campaign finance, presidential nomination process, and more.  I hope to get to them later.  But I gotta go back to get my licks in one more time on a subject from ancient history, Walter Mondale in 1984.  That's because Steve Kornacki responded to my earlier response.  I do want to set this in context...Kornacki is clear that he's not saying that Mondale was responsible for sinking the Democrats that year; he's just saying that another candidate, and specifically Gary Hart, would have done somewhat better.  That said: I think he's dead wrong, and it's worth revisiting because of the general question of the importance of candidates in presidential elections.

Kornacki's evidence of Mondale's weakness is, first, the logic of the situation -- as Jimmy Carter's VP, Mondale was easy to tie to the rejected Carter administration.  And, second, he cites Mondale's low favorable ratings, especially with independents, throughout the campaign cycle (I assume he's correct about that; I haven't looked it up).  The anecdote he produces to support that, however, reveals more than he thinks:

When, after a decent debate showing against a very wobbly Reagan, Mondale's favorable score momentarily outpaced his unfavorable in polling, his campaign considered it a minor miracle.

OK, now, think about that for a second.  Mondale's debate performance wasn't anything special; what was remarkable about that debate was, as he says, that Reagan did a terrible job.  But, we're told, Mondale's favorable ratings rose.  Hmmm....

The problem here is that Kornacki is following the wrong model of how voters decide things in elections.  The model he's following assumes that voters have opinions about the candidates, and assess the arguments of the campaigns, and vote based on that.  But most of the evidence (for general elections) runs the other way: people decide how they're going to vote, and then develop opinions about the candidates and the arguments presented based on their vote choice.  When people suddenly thought that Reagan might be too old for the job, and briefly considered voting for Mondale, suddenly their opinion of Mondale turned favorable.  Once they reassured themselves that voting for Reagan was okay, they remembered that they didn't like Mondale after all.

So if Gary Hart had been the nominee, some of the people who, in June, were inclined to think well of him (and to think of themselves as independents) would have by September learned to associate him with Carter, and they wouldn't have liked him very much.

How do we know that would happen?  Because it always happens; that's why the statistical explanations, which only know about objective factors such as the economy, war, the terms that the incumbent party has held the White House, and candidate ideology, work as well as they do.  And we just had a terrific test of this in 2008.  After all, Republicans in 2008 nominated a candidate who polled very well with independents, and probably had more distance from the White House than any other significant Republican.  But by November, independents had found all sorts of reasons to dislike John McCain and to support Barack Obama.

Look, when we're going through the campaign day to day, we're terribly aware of the mistakes that the losing candidate makes, and how the winning candidate is connecting with the nation.  Those things are true -- but they're also an illusion.  The winning candidate is making plenty of mistakes, too...it's just that it's easier to notice and remember the foolish things the loser did.

Candidates aren't completely irrelevant.  The statistical models don't capture everything that happens (and don't claim to); it's certainly very possible that a good candidate could do a bit better than a lousy one.  But absent something spectacular, that's apt to be in the range of a percentage point or two.

Now, the one thing to add to this, and getting back to Palin, is that so far all the major party nominees -- at least in the era for which we have good enough data to run the models -- have been more or less adequate.  There are no Sharron Angles who get nominated for president.  Gary Hart was certainly no Sharron Angle; he knew the issues well enough.  But we have no idea how he would have held up to the demands of a national campaign.  On the whole, I don't think the record is very encouraging.  Sarah Palin, on the other hand...we've never had a nominee anything like what she would be.  We've had nominees with weak polling numbers, but nothing anywhere close to her unpopularity.  We've had candidates who didn't seem to have learned much beyond the basics of public policy, but she appears (still, at this late date) to be a step or two below the worst we've seen on that score.  I don't recall anyone who was nominated after the sorts of ethics difficulties she's had; I don't recall anyone who was nominated with such little experience (and, yes, Obama in 2008 was relatively inexperienced, but not as much as she is).  Despite all that, she certainly could wind up (were she nominated) doing about as well as a generic Republican as conservative as she is would do, but it does seem to me that there's a huge amount of potential downside here that Republicans would be nuts to risk. 

Of course, Republicans would also be nuts to nominate her because, should she actually win, they'd have a disaster in the White House.