Decisions, Decisions

Every hour on the hour, a new presidential poll comes out. With the clock winding down toward Election Day, a tick up in the polls for one candidate is news. A tick-tock is a trend.

It isn't easy to make sense of this spasmodic sprint to the finish line of this stultifying marathon. But one, necessarily anonymous, strategist in the Gore organization has, I think, got his finger on what is going on. "Anytime somebody breaks away, there is this incredible buyer hesitation," the source said.

As soon as the undecided voter makes a choice, his feet get cold. As soon as the undecided voter has a sense that one of the horses is pulling away, the other horse starts to look a little better. Or safer. Or like the lesser of the two evils.

It's like when you finally decide on a Taurus and head out to the Ford dealer, only to find that the Camry idling next to you at the red light really seems more solid. In this case, the choice seems to be more between a Pinto and a Corvair. Or Bridgestone and Firestone.

"The country is going well and one guy (Bush) wants to change direction," the source observes. "On the other hand, people want to get rid of this crowd (Gore), but they're not sure of the new guy."

An independent voter who came to Washington from Ohio put it more bluntly: "I can't stand Gore, but I'd be ashamed to vote for Bush." Would you like root canal, sir, or shall we just extract the tooth? And I am so sorry, we're out of Novocain.

Undecided voters have more complicated feelings about Gore than Bush. But they aren't that complicated, according to our source, who studies these things for a living. "They've decided they don't like him (Gore) very much. Telling the truth ranks pretty high with them."

The Democratic National Convention cast Gore in a different, more positive light for many voters. But his decline from that convention bounce began not with the debates, as is commonly thought. It started a bit earlier, when the vice president lobbied his current boss to end the "oil crisis" by opening the Strategic Petroleum Reserve.

Focus groups in battleground states conducted immediately after President Clinton opened the spigots showed this incident had a "dramatically" negative impact on undecided voters' views of Gore. Not only did they see through it as a clumsy stunt, it revitalized the image of Bad Al Gore, of Buddhist temple and "controlling legal authority" fame. And then there was the Medicare, mother-in-law and dog medicine mini-flap.

That was part of the setting for the first debate. All the instant polls showed voters thought Gore won handily. But then, after two days of stories about Gore's debate exaggerations dominated the news, his poll numbers went south. "He won the debate, but then people found out he cheated," a pollster quipped.

The vice president's response to this pressure, of course, was a desperate "Reinventing Al Gore" campaign. Acused of exaggerating, he apologized in the second debate. Accused of being too soft, he got aggressive in the third debate. Accused of being for big government, he swore an oath on his Palm Pilot that he wouldn't add a single person to the federal workforce, ever.

Not surprisingly, the further away Gore gets from the debates, the better his poll numbers look.

Throughout all this upheaval, there's only one group that has been universally confident that Gore's going to win the election: academic political scientists who make their living devising mathematical models that predict elections based on mega-factors like consumer confidence, GDP growth and interest rates. These models have proven very accurate in the past. Michael Lewis-Beck of the University of Iowa was the 1996 champion, predicting Mr. Clinton's margin of victory within one-tenth of one percentage point, months before the election was actually held.

At the convention of the American Political Science Association this summer, seven such professors proffered their prognostications. All seven predicted a Gore victory. His share of the vote ranged from 52.3 percent to 55.4 percent in six of the models. The seventh gave a Gore 60.3 percent win. None thought the election would be particularly close. These academicians may understand Gore's advantages better than the folks who plot his daily strategy.

So, if Al Gore wins, it will be a triumph of grand economic and historical forces. If George Bush wins, it will be a triumph of personality.

And if either is elected with an overwhelming mandate to govern, it will be a miracle.