As the tasters of public opinion chew up and spew out the daily menu of political polls that profess to say who's ahead and who's behind in the presidential race, unforeseen events continue to intrude, mocking their certitude.

The vice-presidential nomination of Sarah Palin supposedly was the game-changer of the 2008 campaign — until the financial crisis on Wall Street suddenly elbowed her aside. Overnight, John McCain's assurance that the economy is "fundamentally strong" gave Barack Obama an opening to get back on the offensive.

As George Gallup Sr. used to caution, a poll is no more than a single snapshot of a single moment of time. But the temptation to treat polls as crystal balls is increasingly irresistible to armchair oracles who endlessly gaze at them as prophesies.

The worst offenders are cable television swamis, who lump them together as a "poll of polls," blurring or ignoring the distinctions of sample size and method and other refinements, and stirring them into an inconclusive potpourri of data.

This is not to say that public-opinion polls are useless; they can be invaluable — to campaign strategists. They look to the best of them to gauge the strengths and weaknesses of their candidates with certain voter groups and on specific issues, and to adjust their campaigns accordingly.

But in most presidential elections, the flow of events injects a changing imponderable element that defies prediction. Although the current financial crisis appears now to be a tidal wave flowing over all previous calculations, there's no telling there won't be some other unforeseen event jumbling the outlook and the fortunes of the candidates.

That kind of game-changing event need not be unforeseen either. Three presidential debates between McCain and Obama, the first of which is now only a week off, and the vice presidential debate between Palin and Joe Biden a week later, could reshuffle the cards.

In the first presidential debate of 1960, John Kennedy's more aggressive and forceful discussion of a range of issues, coupled with Richard Nixon's disinclination to hit back hard at him and often ill-at-ease appearance, was later appraised as the turning point of that campaign.

In another very close election in 1976, President Gerald Ford's erroneous contention in his debate with Jimmy Carter that Poland was not under Soviet domination arrested his late-campaign momentum. Many said it cost Ford the election, although his pardon of Nixon for alleged Watergate crimes probably did more damage.

In 1980, Ronald Reagan in debate with Carter deflated voter concerns about his age and lack of national experience, and four years later he again dismissed the age question with a quip. In 1992, Bill Clinton in debate demonstrated a much greater grasp of the economic concerns of average voters than the first President George Bush and went on to win the election.

Late-breaking events can also be critical. In 1968, Hubert Humphrey was closing the gap on Nixon when Vietnam peace talks by President Lyndon, which might have turned the election Humphrey's way, collapsed. Nixon was later suspected of trying to scuttle the talks, but it was never proved and he was narrowly elected.

Because election outcomes are so subject to the uncertainties of events, whether formally staged events like debates or unpredictable turns of fortune, the news media focus on polls is of dubious merit, for all the hoopla over the ups and downs.

The volatility of the campaign is also intensified by the tremendous mushrooming of news media outlets of every variety, poised to circulate and in many cases magnify what the candidates say and do, around the clock. They are under closer scrutiny than ever not only on the campaign trail but also in endless television network and cable interviews as well as radio chatter.

Voters will better serve their own interest, and that of the country, by concentrating on the substance of the unfolding campaign day to day, without paying much attention to who's supposedly ahead and who's behind in the polls. Election night on Nov. 4 will provide the answer soon enough; that is, unless the result is so close that there's another 2000 fiasco, in Florida, Ohio or somewhere else.

Reach Jules Witcover at juleswitcover@earthlink.net.