Perspectives, News & Opinions From The Researchers At Edison

Where the Presidential Election Stands Today and Why This Election is So Hard to Predict

Entry by Tom Webster | Monday, August 30th, 2004 | Permalink

by Joe Lenski, Edison Executive VP

The latest Time Magazine Poll came out this weekend and it showed Bush with a two-point lead over Kerry (46% to 44%) if Nader is included on the ballot and a dead-even race (46% to 46%) without Nader included. This poll confirms what all of the recent national polls have shown – there has indeed been a 4 to 5 point net move to Bush on the horserace question in the last two weeks or so. In the last six polls released nationally Bush is only behind in one – the Fox Poll, which has Kerry up by only one point.

Besides the horserace numbers, the Time Poll has a couple of interesting trends which seem to explain this trend. First, Bush’s “re-elect” number (the percentage of voters who say that Bush deserves to be re-elected) has gone from 42 to 46 over the last three weeks. If Bush can get this number over 50 after the convention the Democrats are really going to be worried. At the same time the Kerry “favorable” number has gone from 53 to 44 in the last three weeks. Part of this is the Swift Boat controversy, but a more important reason for Kerry’s drop is that he has not succeeded in putting anything else out there in the last three weeks to take the media’s attention off of the Swift Boat controversy.

It is also amazing how fast the conventional wisdom has moved from “this is Kerry’s race to lose” to “Boy, has Kerry already blown it.” Both statements were and are complete exaggerations of the situation. We are headed for political territory with which there is no past experience, and we can’t look at any past campaign in order to predict what is going to happen in the next 64 days until the election.

First, we have never had an incumbent right at or just under 50% job approval at election time before. Every incumbent over 50% has won (Eisenhower, LBJ, Nixon, Reagan and Clinton), and every incumbent under 45% (Ford, Carter, Bush the elder) has lost. No one can predict from past experience how an incumbent between 45% and 50% job approval will do.

Second, we have never had a campaign before where the both candidates have targeted the same few states and concentrated almost their entire effort on those few states FOR MORE THAN SIX MONTHS leading up to the election. What are the repercussions of all of the television ads and campaign visits being concentrated in the same 18 states? Again no one knows from past experience. In past elections the list of battleground states has typically changed during the course of the campaign – remember Dukakis picked Senator Lloyd Bentsen from Texas as his running mate in 1988 because the polls around the Democratic convention indicated that Texas could be competitive.

This year will the voters in these 18 battleground states begin tuning out because of campaign ad overexposure? Or will the sheer volume of campaign ads lead to increased voter attention and thus increase voter turn out in these states? The conventional wisdom is that if there is a national wave at the end of the campaign (such as the wave for Reagan in the last week of the 1980 campaign) it will send all of the close states one way. But has all of this advertising and campaigning in the battleground states made voters there immune from the effects of a last-minute national wave? Or is a national wave not going to happen – any wave that we may have to look for may be concentrated in the 18 battleground states and not show up in the national polls? Again no one knows.

Third, more votes will be cast before Election Day than ever before. There are eight states (Arizona, Colorado, Nevada, New Mexico, Oregon, Tennessee, Texas and Washington) where more than 40% of the vote will be cast either by mail or in-person during the weeks before the election. Another eight states (Alaska, Arkansas, California, Florida, Hawaii, Iowa, North Carolina and Vermont) expect 20% to 30% of the total votes to be cast by early or absentee voters. Nationwide it is likely that at least 20% of all votes (over 22 million) will have been cast before polling locations even open on November 2nd. We have never seen a national campaign operate under these rules before. Last minute news like the revelation of the Bush DUI arrest during the last week of the 2000 campaign or some sort of terrorist attack such as the one in Madrid this spring a few days before the Spanish elections can only affect the 80% of voters who are voting on election day – the other 20% of the votes will literally already be in the mail.

In the end this election might not hinge on any of the old conventional wisdom – “the economy, stupid”; presidential job approval; rallying around the flag in war time etc. – but rather on which of these two men a majority of voters want to trust with the future of the country. And here the nation remains essentially evenly split.

Here is one last set of numbers, this time from the latest AP/Ipsos poll:

Does George W. Bush make you feel optimistic about America’s future or not? Yes 51%; No 48%

Does John Kerry make you feel optimistic about America’s future or not? Yes 50%; No 49%

Looks like it is going to be a long election night.

Comments are closed.