So who’s going to win the election?

I don’t know. You don’t know either. Oh, and if you think you know, well, you’re full of it. [I know, I know!!–ed. No, no you don’t.] There are now a lot of sites providing Electoral Map projections, and all of them showing a close race in way too many battleground states. But these are all based on polling techniques that, in recent years, have elevated margins for error. Over at Slate, William Saletan, David Kenner, and Louisa Herron Thomas have a summary of the various bells and whistles each polling service has — but none of them can correct for the problem of declining response rates. Richard Morin makes this point in today’s Washington Post:

Two consecutive Election Day debacles have shaken public confidence in exit polls, once viewed as the crown jewel of political surveys. Cell phones, Caller ID and increasingly elaborate call screening technologies make it harder than ever to reach a random sample of Americans. Prompted by the popularity of do-not-call lists, a few state legislatures are considering laws that would lump pollsters in with telemarketers and bar them from calling people at home. Costs are soaring as cooperation rates remain at or near record lows. In some surveys, less than one in five calls produces a completed interview — raising doubts whether such polls accurately reflect the views of the public or merely report the opinions of stay-at-home Americans who are too bored, too infirm or too lonely to hang up…. No surveys are immune. “Phone surveys are suffering, but so are response rates to mail surveys and even mall intercept surveys” in which people are interviewed while shopping, says Tom Smith, director of the General Social Survey at the University of Chicago, the best source of data on social trends in the United States. “All of the dominoes are being knocked down because the whole table is being shaken.” Currently cooperation rates hover at about 38 percent for the big national media surveys conducted over several days, but can dip down into the teens for surveys completed in a single night, says Jon Krosnick, a psychologist at Stanford University who has completed a groundbreaking study of response rates. Even exit polls are feeling the pinch. In each of the past three presidential elections the proportion of people who agree to be interviewed after leaving the voting booth has dropped — from 60 percent in 1992 to 55 percent in 1996 to 51 percent in 2000. For decades, the conventional wisdom has been that high response rates equaled high-quality, more accurate surveys. Generations of pollsters-in-training were told in graduate school that the people who decline to participate in a poll, or cannot be reached, could be different than those who are contacted, in ways that would affect results. Two converging trends — the rise of telemarketing and growing time pressures in the home — have frayed America’s nerves and left many people unwilling or downright hostile when it comes to talking to pollsters. But a bigger problem seems to be that people are simply harder to reach. They’re working longer, going out more and using call-screening devices when they’re home, Krosnick says.

Keep this in mind when someone trumps a one or two point lead by their candidate. And check out Mark Blumenthal on the cell phone issue. There is one wild card, however, that I haven’t seen discussed all that much. While much of the concern about third party tickets is whether Ralph Nader would get votes for Kerry, this Electoral Vote Map points to another potential third-party spoiler:

A Rasmussen poll taken Oct. 26 in Arizona puts Libertarian party candidate Michael Badnarik at 3%. When the pollsters actually ask about him, he does surprisingly well. He might end up canceling out the Nader factor by appealing to disgruntled Republicans who support a balanced budget and small government and are appalled by the current deficit and power the Patriot Act gives the government to snoop on people’s lives.

I’ve largely tuned out on the polls, but I don’t think I’ve seen many of them with Badnarik included. With the number of states within the margin for error, that three percent could matter. UPDATE: The Weekly Standard‘s Rachel DiCarlo runs with the Badnarik meme, observing, “In September, a Rasmussen poll gave Badnarik three percent of the vote in Nevada, and in August Rasmussen showed him taking five percent of the vote in New Mexico–both considered potential swing states.” Readers are invited to suggest the biggest factor that is not showing up in the polling data but could decide the election — as well as who you think will actually win. UPDATE: Another question: how big will the Schilling factor be in New Hampshire? UPDATE: Never mind.