Via the new Pew Poll comes evidence that McCain's broad coalition of People Who Like Him is beginning to polarize by party as he moves towards the Republican nomination. His favorability numbers among Democrats have tanked, and his ratings among independents have fallen, even if only to 51 percent. Meanwhile, Republicans seem to have discovered a newfound affection for the guy, and their recognition that he too supports endless war and upward redistribution have sent his numbers skyrocketing. None of this is particularly shocking -- in fact, expect to see the same phenomenon with Obama before very long -- but it suggests that McCain is vulnerable in exactly the ways everyone is saying he's vulnerable. As Matt noted the other day, both McCain and Obama have faced their toughest battles in primaries, and so very few voters have heard McCain attacked from the left, or Obama assailed from the right. But they will soon.
This is exactly the process that I predicted would take place if McCain won the nomination:
McCain spiked upward in popularity at the time he made his run in the 2000 presidential primaries, and has barely declined since...
The reason, I think, is that Democratic politicians don't criticize McCain (some even ask him to be their vice presidential nominee). He's more or less the only partisan politician who Democrats and Republicans generally praise. As a result, the public likes him across the board -- the Quinnipiac poll shows his favorabe ratings as 37% favorable, 8% unfavorable among Republicans; 31% favorable, 8% unfavorable among Democrats; and 40% favorable, 9% unfavorable among independents.
But this will inevitably change if McCain runs in 2008. The reason is that he's never received significant Democratic criticism. He was defeated in 2000 before the Democrats felt the need to open up on him, and since then they all praise him because they want to look bipartisan and co-sponsor bills with him in Congress. But he won't get the Republican nomination unless he starts unloading on the Democrats, and if he does get nominated (which I think is unlikely), things will turn around really quickly. Pretty soon Democrats will start pointing out that he's a pro-life, ultra-hawkish, government-cutting conservative, and his favorability profile will start to look like most other Republicans. And even though McCain's numbers look great today, few Democrats would actually vote for him -- if he's only getting 15% against Hillary Clinton right now, imagine what he'd draw against a more moderate Democrat like John Edwards at the end of a vicious presidential campaign.
The counterpoint to this is the finding in the same poll that Barack Obama's support lags Hillary's among some Democratic constituencies. But as Matthew Yglesias notes, these people are very likely to come into the fold by November:
Obama is winning even though he's doing unusually poorly among self-identified Democrats. In particular, older white working class Democrats seem drawn to McCain in pretty large numbers. But you've got to consider that at this point almost ever older white working class Democrat in America has been the target of a lot of messaging from Hillary Clinton arguing that Obama is too inexperienced and too dovish. They haven't, meanwhile, heard any messaging from anyone about how John McCain wants to privatize Social Security and cut Medicare benefits. Obama, in other words, is currently winning despite weakness with this demographic, and is also almost certain to look less weak among this demographic in November than he does today.
The tension here is between the predictability of partisanship and election outcomes on the one hand and the prevailing belief in candidate quality as a major factor in elections. During campaigns, the fundamentals end up pushing people in relatively predictable directions, but we tend to attribute many of those changes to political skill (or the lack thereof). In reality, the candidates themselves probably only matter on the margin, as Andrew Gelman argues (though you'd still rather be McCain than Romney and Obama rather than Clinton):
The predictability of election outcomes from fundamental variables suggests that different presidential candidates from the same party don't differ much in the votes they will receive in the general election. It's better to be a moderate than an extremist, and it's better to be a better campaigner etc., but all these things together probably only count for a couple of percentage points of the vote.
...Now, don't get me wrong: a couple of percentage points of the vote can make a big difference--just look at the tied elections of 1960, 1968, 1976, and 2000, as well as the very close election of 2004. Also, who knows how things will go with the unprecedented "woman or young black guy vs. old white guy" dynamic. But, based on past elections, I'd say the whole "electability" thing is overrated. Once Election Day comes around, people will find a reason to vote for the party they want to support.
For more, see the article Gelman and Harvard's Gary King wrote on this phenomenon, which shows how voters tend to eventually end up where we would predict over the course of the campaign:
Gelman, Andrew and Gary King. "Why Are American Presidential Election Campaign Polls So Variable When Votes are So Predictable?" British Journal of Political Science, Vol. 23, No. 1 (October, 1993)...
As most political scientists know, the outcome of the U.S. Presidential election can be predicted within a few percentage points (in the popular vote), based on information available months before the election. Thus, the general election campaign for president seems irrelevant to the outcome (except in very close elections), despite all the media coverage of campaign strategy. However, it is also well known that the pre-election opinion polls can vary wildly over the campaign, and this variation is generally attributed to events in the campaign. How can campaign events affect people's opinions on whom they plan to vote for, and yet not affect the outcome of the election? For that matter, why do voters consistently increase their support for a candidate during his nominating convention, even though the conventions are almost entirely predictable events whose effects can be rationally forecast?
...We show that responses to pollsters during the campaign are not generally informed or even, in a sense we describe, "rational." In contrast, voters decide which candidate to eventually support based on their enlightened preferences, as formed by the information they have learned during the campaign, as well as basic political cues such as ideology and party identification. We cannot prove this conclusion, but we do show that it is consistent with the aggregate forecasts and individual-level opinion poll responses. Based on the enlightened preferences hypothesis, we conclude that the news media have an important effect on the outcome of Presidential elections---not due to misleading advertisements, sound bites, or spin doctors, but rather by conveying candidates' positions on important issues.
This plot in particular does a nice job of showing how the estimated effect of ideology and race increased over the course of the 1988 campaign, while the effect of other variables declined dramatically (see page 445 of the paper for details):
In addition, this plot shows how actual support for George H.W. Bush converged with his predicted level of support using regression weights from polls immediately before the election (see page 448 for details):
If John McCain ends up losing this election, expect a flurry of stories claiming he lost because he shifted to the right and lost his so-called "maverick" appeal. The state of the economy and other factors will get much less attention.