Today is Election Day, the day when all of the disputes about the who will vote and in what numbers get resolved -- at least until the next election.

There has been no more bitter fight between partisans in the two parties than the battle over party identification in state and national polls leading up to today's vote. Republicans have consistently argued that pollsters are tilting their samples too Democratic and, therefore, producing results that not surprisingly tend to favor Democrats.

We've written a bunch on party ID -- and how badly it's misunderstood -- but since it's Election Day and all we can do is sit around and wait until results start coming in, we thought it might be worth revisiting the argument now.

Let's start with where the party ID wound up in the unleaned final polls conducted by some of the big media organizations and the best independent pollsters.

* Washington Post-ABC News: D+6

* NBC News-Wall Street Journal: D+5 (D+2 when independents are pushed to state preference)

* Pew Research Center : D+4

* Politico: D+3

And now for a look at what the party identification has been in the actual elections, according to exit polling dating back to 1972.

In nine of the 10 elections since 1972, Democrats have had a party identification advantage ranging from 16 points in 1976 (hello, Watergate!) to just two points in 1984. (The one outlier was 2004, when 37 percent of the electorate identified themselves as Democrats and an equal 37 percent identified themselves as Republicans. 

In that same time period, Republicans won six of the 10 presidential elections.

What gives?  A few things.

1. People often badly under-estimate the role that independents play in an election. It's why, in the 1984 landslide in which Ronald Reagan carried 49 states, Democrats had a two-point party ID edge in exit polling. Reagan simply won a major victory among independents and carried many so-called Reagan Democrats. Also remember that so-called "independents" tend to line up with one party or another in exit polling -- a phenomenon explained by the fact that they have just committed the decidedly partisan act of voting. 

2. Party ID in pre-election polling is subject to the same margin of error that the rest of the numbers in a given poll are. That goes for each of the numbers in the party ID calculation; so if the margin of error in a poll is plus or minus 2.5 percent, that margin applies to the Democratic, Republican and independent numbers too.  Here's a chart that makes the relative wideness of the margin of error plain:

All of which gets us to the most basic question asked by partisans -- particularly on the Republican side: Do you really think the 2012 electorate will be as Democratic-leaning as the 2008 one?

Simple answer: We (and the polling community) don't know. Assuming that the party ID will be somewhere between 2004's neutral split and 2008's seven-point Democratic advantage might seem to make all the sense in the world -- except that we are either four or eight years removed from those two electorates and, well, things change.

Making assumptions about what the partisan shape of the electorate will be today based on what it was four years ago (or eight years ago) assumes absolute stasis in the electorate. And we all know that the only constant in politics (as in life) is change.

Will the above argument satisfy the party ID deniers? No. And will this argument resurface in four years time? Yes. But, remember that party ID is a moving target, and assuming that it can or ever will be a fixed target is a mistake.

Romney banks on Ohio or Pennsylvania: Romney is spending the final day of his campaign in two states that he's likely to lose, the Post's Felicia Sonmez reports.

Romney is spending the day in Ohio and Pennsylvania after initially planning on finishing the campaign in New Hampshire, where he has a vacation home. He will stop in both Cleveland and Pittsburgh for his final swing.

The move is both a reflection of how important Ohio and Pennsylvania are to Romney and how tough the Electoral College math is for him.

His campaign appears to recognize that there will be no victory without Ohio or Pennsylvania, and better to focus on those states at the end rather than the pure tossups.

Romney will be in Boston on Election Night, while Obama will be in Chicago.


Make sure to keep The Fix's Election Night Viewer's Guide handy tonight.

late Gallup poll shows 56 percent of Americans disapprove of Obama's handling of the economy.

David Axelrod takes on Dick Morris.

A pro-Obama voter guide in Virginia references Romney's Mormon faith.

Some are worried that Florida's lengthy ballot could pose problems.

More signs of tension between Romney HQ and New Jersey Gov. Chris Christie (R), with a Romney aide saying Christie declined to appear at a Romney event in Pennsylvania. Christie's camp denies it.

We should know by early Tuesday night just how the early vote in Ohio broke. Polls have shown Obama winning the early vote there by 20 points or more.

Sen. Dean Heller (R-Nev.) basically predicts Obama will carry Nevada, saying “I clearly need the Obama voters to win.”

New York voters who were displaced by Sandy will be able to vote anywhere.

new poll in Massachusetts shows the race is tight again, with Elizabeth Warren at 49 percent and Sen. Scott Brown (R-Mass.) at 48 percent. Warren had been pulling away in other polls.


"Obama, Romney campaigns brace for battle after Tuesday’s vote" -- David A. Fahrenthold and Bill Turque, Washington Post

"Dueling Bitterness On Cable News" -- Jeremy W. Peters, New York Times

"Election ‘Day’? With early voting, it’s more like Election Month." -- Ann Gerhart, Washington Post

"What kind of president would Obama be in second term?" -- Dan Balz, Washington Post

"In Case of a Recount, a Long Wait for Ohio" -- John M. Broder, New York Times

"Secret money funds GOP door-knockers" -- Kenneth P. Vogel, Politico