Posted 10:38 AM by Robert

Taking Gallup Seriously For A Moment

Polls are not elections and the current Gallup Poll showing President Bush leading John Kerry by 13% is just one poll taken before the candidates have even held one debate. I think it is fair to say that few professionals believe that Mr. Bush will lead by 13% in the November election.

But that 13% number does make considering some consequences of a really big Republican presidential win - a win, say, north of 8%. It is commonplace wisdom that turnout is driven by the top race on the ticket, and late-in-the-election-day voters therefore can be expected to stay home if the election is already decided early in election day. But this election has an additional dynamic which may profoundly interact with that commonplace wisdom: Several polls show that Mr. Kerry's support is disproportionately voting against Mr. Bush, not for John Kerry.

What happens to that kind of negative Kerry "support" if the media announces early on election day that Mr. Bush wins big, which is likely to be the case if the President is reelected by a big margin? One normally expects (again, a commonplace belief) the supporters of the loser to become disheartened and to stay home. In contrast, the supporters of an early big winner only have to overcome the sense that their vote will not further the success of their candidate - but they have their feeling of being "up" from his early win to accomplish that overcoming. So in any presidential election an early win generally tends to suppress turnout of the loser's supporters more than turnout of the winner's supporters.

But what about that curious additional factor noted above? Kerry-Edwards "supporters" have few positive feelings about the Democratic candidate in this presidential election. There seems to be little precedent for a race colored by that factor to the extent this one is - so predicting this factor's effect on the election is necessarilly even more speculative than usual. But it seems likely (and at least plausible) that Kerry-Edwards' supporters would be unusually and disproportionately disheartened by an early and big Bush-Cheney win because so many supporters of Kerry-Edwards don't like "their" candidate anyway. A post-win vote just evidences the voters affirmative support for the loser, and there's not much of that to evidence on the Democratic side. Once Mr. Bush wins, there's no point in voting against him. If that perspective is correct, Kerry-Edwards supporters should stay home much more than is usual (compared to other elections) and much more than Bush-Cheney supporters (in this election). It would be interesting to see focus group and private polling on this question. One might ask a likely California voter who has already chosen a candidate: "If on election day you were to hear on the news at a time before you voted that your candidate had already lost (or won) the election, would you still make the trip to the polling station?"

As noted above, elections are normally driven by the race at the top of the ticket. If Kerry-Cheney supporters stay away in droves, Democrats will likely be adversely and seriously affected all down the line. A big and early Bush-Cheney win would have particularly spectacular effects in the West. Of course, Western voters have more of a chance to evaluate Eastern voting patterns before deciding whether it's worth a trip to the polls - which should tend to exaggerate the turn-out consequences. But there's more. The polls to which I have access do not break out "internals" on a state-by-state or regional basis. But substantial anecdotal evidence has convinced me that John Kerry's aloof personality, a personality that is already hard for a lot of New Englanders to take, is even less sufferable in the West than it is in the rest of the country. Indeed, despite polling that indicates that Senator Kerry has a measurable and positive "likability" rating, I cannot find a single Southern California voter who actually likes him personally - not even among people who have given huge amounts of money to his cause and Democratic 527's. I therefore strongly suspect that Western Kerry-Edwards supporters are even more disproportionately voting against Mr. Bush, and not for John Kerry, than is true in the rest of the country. If my suspicion is correct, Western Democratic turnout should be even more reduced by a big early Bush-Cheney win.

My expectation is that a big and early win for Bush-Cheney of the type suggested by the Gallup Poll would translate into a particularly large Republican "coat tails" effect, especially in the West - even in California.

Of course, we don't yet know that there will be such a big and early win.