Al's ahead. Dubya's ahead. Ralph's ahead (well, maybe not). Pat's ahead (God help us all).

Every day some news organization (doncha just hate that phrase?) has a new presidential election poll. For a long time, it looked like Mr. W would just ride his butt right into the White House. Now we're being told that he just might not.

Here's the deal: A landslide is not an interesting story. Reporters don't like them, editors don't like them and most people don't like them. We want our Super Bowls to be 20-19, wide left (or was that wide right? I get confused). We want our elections to be close so that Dennis Miller won't refer to Walter Mondale's performance in the 1984 election as getting "stomped like a narc at a biker rally."

Plus, you know, close elections are more fun.

So I've got a plan. The next time some pollster calls your house, tell them you're a committed lesbian for Buchanan. Or a tree-hugger for W. Or an oilman for Gore. Or a GM executive for Nader. Anything that will totally screw up the poll.

A few years back, the government did a study on heroin use. Some grandma in a nursing home thought it would be fun to play with Uncle Sam, so she said that she shot up twice a day and had her drugs delivered to the home via UPS or her grandson, whoever was cheaper.

You might recall the headlines proclaiming rampant heroin use among the elderly. I do. Turns out that the grandma in question represented more than a million senior citizens in the study's methodology. And all of a sudden, the drug problem ramped forward a couple of generations. There were stories written about retirement centers doubling as smack houses, and elderly people so desperate for a fix that they signed over their social security checks.

I'm not making this up. I know I'm a bit hyper on this issue, but all this really happened. One story on the study came out, and lots of "localized" stories followed. It didn't even matter that the study was wrong, though all of the storied did cause the researchers to go back and take a second look at the data.

Well, they finally figured out their mistake. And even if Granny rode the white horse, I think it would be silly to assume that hordes of old folks are slapping their arms, trying to find a vein. The thing is, it took six months for common sense to kick in. By that time, no one wanted to hear that the study was a bust.

One nicely warped grandma totally screwed up a federal drug survey. Imagine what a few thousand bizarro answers on polls would do to the presidential race. These guys have 9 1/2 weeks, not six months. They'll never notice the strangeness until it's too late.

So from now until November, keep the pollsters and the candidates guessing. If we mess with the numbers enough, we might convince W to endorse universal health care, Gore to come out in favor of gay bashing as an Olympic sport or Nader to drive a Corvette.

Alright, maybe that last one's a bit dicey under any circumstances. But if politics are as poll-driven as they seem, a big wad of nutty answers in the data just might be the ticket to straightening out the system. And if they're not, well, at least it was fun trying.

If you can't have fun at election time, you're just not trying hard enough.

Jon Worley plans to vote. Don't ask about particulars right yet.