Tom Wolf, so the public opinion polls tell us, has a commanding lead with less than a week to go before the Democratic gubernatorial primary. According to the Franklin & Marshall College Poll, he is 19 points ahead of his closest opponent, U.S. Rep. Allyson Schwartz.

Wolf has maintained his surge in the polls for weeks - ever since he began his intensive advertising campaign, long before the other three Democratic candidates. A year ago, Schwartz was considered the inevitable Democratic nominee, but c'est la vie for her.

But as good as it looks for him, it would be presumptuous of Wolf to start measuring the windows in the governor's office for new drapes. Wolf, though way ahead, is the first choice of only 33 percent of Democratic voters.

That means two-thirds of the electorate either favors someone else or, even at this late date. is still undecided. The undecided figure this close to the election - 39 percent - is huge. And with all the buzz Wolf has gotten, and with all his willingness to spend scads of his own money, wouldn't you think that anyone who was going to fall in love with him would have done so by now?

In a primary that could set records for low voter turnout, it is not difficult to envision a scenario where the undecided vote breaks at the last minute for Schwartz, state Treasurer Rob McCord or Katie McGinty. The returns Tuesday night could show Wolf winning in a "squeaker" or - who knows? - either Schwartz or McCord, with their mastery for "getting out the vote," staging a "surprising" upset.

Polls are interesting for political wonks and, of course, absolutely indispensable for campaigns. They have been known to be wrong at times and, of course, they are always subject to interpretation, but, by and large, they are the accurate reflection of public opinion at a particular moment in time. Of course, there is always a chance people's minds could be changed tomorrow.

But no one has ever convincingly answered that age-old question: Do polls predict the outcome of an election, or are they self-fulfilling? To put it another way: How many people walk into the voting booth and vote for someone just because he or she is "ahead" in the polls? It's human nature to aspire to be on the winning side.

Is Wolf poised to win the Democratic primary because his nomination appears more likely or because people like what's he saying and how he says it? Are people turned on by his ideas or by an irrestistible urge to jump on the bandwagon?

Volumes have been written and multitudes of hands have been wrung raw due to agonizing over the buying of elections. Candidates with deep pockets or with a knack for befriending others with deep pockets generally wind up cheering on election night. An even more insidious influence - at least potentially - is the ever-present public opinion polls. Thanks to polls, and their high levels of accuracy, election campaigns are treated more like baseball games. "Who's ahead" dominates media coverage of campaigns, with thoughtful discussion of candidates' stand on issues taking a back seat.

Campaign contributions are free speech, and so too is the publication of poll results. So much as it would be in the public interest to prevent news organizations from sponsoring and then publicizing the results of campaign polls, it won't pass constitutional muster.

Poll results affect potential donors' decisions on who to support and how much to give. Polls give the impression that likely outcomes are inevitable, thereby fostering the impression that voting is pointless. The cause of democracy would be better served if people changed the channel or turned the page every time the word "poll" is uttered or written.

(Jake Betz is an assistant editor of The News-Item.)