Tricks of Trade Don't Predict '04 Race

Armed with the belief that looking at the past is the best way to predict the future, election analysts still use old rules of thumb to try to gauge the success of the presidential candidates. But many say even those are as good as useless this year.

"I’ve stopped using them," says Terry Madonna, public affairs professor and director of the Keystone Poll (search) at Franklin & Marshall College in Pennsylvania. "Part of the problem we’re facing is this is such an unusual election."

Based on seemingly sound pattern recognition, several "rules of thumb" are frequently used by analysts to determine how the incumbent will fare:

Right Track/Wrong Track — The incumbent president is bound for defeat if the majority of Americans polled say the country is on the "wrong track," rather than the "right track."

Approval Ratings — The president is also bound for defeat if his approval rating is below 50 percent. President Bush (search) is on the losing side on both "approval ratings" and "right track/wrong track" measures in recent polls.

Change of Direction — An incumbent can’t get re-elected if the majority of Americans support change in the direction of the country.

Late Undecided Votes — Undecided voters typically "break" for the challenger on Election Day.

— Popular Vote — A president who did not win the popular vote the first time around will not be re-elected. This is based on the three one-term presidents who lost the popular vote and won the electoral vote: John Quincy Adams, Rutherford Hayes and Benjamin Harrison. Bush won the electoral vote in 2000, but lost the popular vote.

Some rules aren’t based on statistical history, but on external events and voter behavior. And even they can contradict themselves. For instance, some theorize that an incumbent cannot win re-election in the middle of an "unpopular war." But this theory doesn’t jibe with another rule of thumb: "Never change horses in midstream."

Then again, none of these may mean a hill of fallen chads come Tuesday.

"Every year, an iron law gets broken," Jim Pinkerton, political columnist and senior policy official in the first Bush administration, told "They all make perfect sense, until you realize they don’t."

Analysts say the approval rating and right track/wrong track assumptions appear thin this year because Bush’s approval numbers are hovering around 50 percent. And although most of the country appears to think the country is on the wrong track — between 48 percent and 56 percent in recent surveys — according to the same polling, Bush's opponent, John Kerry (search), doesn’t seem to have firmed up overwhelming confidence with voters on making things better.

Juan Williams, correspondent for National Public Radio and FOX News contributor, says tried and true rules mean nothing because the war has divided people like never before.

"In this election, all of that gets thrown in the air," Williams said. "It may be that people have problems with the president, the direction of the economy or whatever, but it comes down to the war."

And while voters may be disenchanted with what is going on in Iraq today, "They are divided about who would be better to get us out," Williams said.

Conventional wisdom says the presidential election is "all about the incumbent," but even that has been turned on its head this year, as the Bush administration and its surrogates have endeavored to make Kerry look weak on both the domestic and foreign policy fronts, said Williams.

"We have never had this level of spending and advertising where the incumbent has made the point that his challenger does not have the spine to win this war. They have made this a referendum on the challenger," he added.

Jim McLaughlin, a New York-based Republican pollster, said he too believes historical precedence is being waylaid by the dynamics of this campaign.

"People will say ‘wrong track’ on the war in Iraq (search) and the economy, but they are not necessarily blaming Bush, and they think he’s doing a good job of keeping them safe," he said. "And they don’t think raising taxes is the best way to improve the economy."

As far as undecided voters go, McLaughlin pointed to the 1996 race between Bill Clinton and Bob Dole — Dole got two-thirds of these voters, but certainly not enough to help him beat the incumbent president.

Tom Smith, an analyst with the National Opinion Research Center at the University of Chicago, said there is "good theory" behind the rule on undecided voters, but he’s not sure it fits this year. "There seems to be more undecided who are conflicted rather than uninformed," meaning they like each man on different issues and it is not clear yet which one of those issues will ascend to priority.

This is different, Smith said, from previous years when the undecideds were leaning toward the challenger but needed to "feel comfortable" before closing the deal.

Pinkerton said the risk one takes by predicting future results based on retrospective analysis only goes so far. "Any time you reason that because it happened retrospectively, then therefore it will happen in the future, then you are committing intellectual fallacy," he said.

He noted the 18th century historian and philosopher David Hume, who said that just because one has seen only white swans doesn’t mean black swans don’t exist. In other words, look out for the random black swan.

In the 2000 election "there were models which took into account a combination of presidential popularity ratings, change in gross national product, unemployment rates and approval rates — and they all said Al Gore would win by a comfortable margin," Smith said. "This shows how good, well-tested rules can go wrong."