With only three days left, where does the race stand in the nine swing states that will determine the election? The best way to figure this out is to focus on the polling averages calculated for each state. There has been a torrent of polls released in the last two weeks, and—collectively, never individually—they give us an accurate picture. Rather than use one average, we’ll average all of the averages—from Real Clear Politics, Pollster, Talking Points Memo, and FiveThirtyEight—in order to get the fullest picture. Since the swing states are divided into four regions—Northeast, Southeast, Midwest, and Southwest, we’ll tackle them in that order.
One thing to keep in mind is that it’s unusual—if not rare—for candidates to lose states where they lead by two points or more this close to Election Day. It can happen, but it’s far from likely.
There’s only one swing state in this heavily Democratic region of the country, and that’s New Hampshire, with its four electoral votes. Its racial homogeneity (92.3 non-Hispanic white) keeps it from being a lock for Democrats, but its large population of highly educated people gives it a light-blue tint. The polling averages show Obama with a decent lead: Nate Silver gives him a margin of 2.9 points (48.8–45.9), Real Clear Politics has him up 1.8 points (49.2–47.4), Pollster has it at 2.2 points (49–46.8), and Talking Points Memo's average gives Obama a lead of 4.3 points (49.5–45.2). Meta-margin: Obama by 2.8.
This region is absolutely crucial to Mitt Romney’s chances for the White House. Virginia and North Carolina are longtime red states that—thanks to demographic shifts—flipped in the 2008 election, while Florida is a historic swing state that has traditionally had a Republican tilt because of its large population of conservative-leaning Cubans. Since October, Florida and North Carolina have been slightly in Romney’s corner; the meta-average gives Romney a 0.7 lead in Florida and a 2.4 (rounded up) lead in North Carolina. In Virginia, Obama holds a narrow 1-point lead, built on his overwhelming support from African Americans, young people, and women.
In practical terms, this makes North Carolina—and its 15 electoral votes—solid for Romney. Virginia (13 EVs) and Florida (29 EVs) are toss-ups and will almost certainly be decided by turnout and the strength of each side’s ground game. Romney has to hope that his is up to the challenge; while losing Virginia puts him at a huge disadvantage—he'd have to win every other swing state, plus Ohio or Florida to reach 270 electoral votes—losing Florida makes it impossible for him to win.
The campaigns have invested the most resources in Ohio (18 electoral votes), Wisconsin (10 EVs), and Iowa (6 EVs), under the (correct) assumption that the race will be decided here. For Obama, it’s where the auto-industry bailout has offered the greatest political dividend. Keeping the industry afloat in Wisconsin and Ohio has contributed to his solid standing with white working class and pro-labor voters in each area. What’s more, this is where Team Obama has done the most to damage Mitt Romney’s credibility—his favorability remains underwater in Ohio, even as it's improved nationwide.
At the moment, according to the meta-average, Obama holds a 2.9 point lead in Ohio, a 5-point lead in Wisconsin, and a 2.9 point lead in Iowa. These are all close to decisive—it’s hard to imagine Romney making up this much ground in a few days—but it’s the Ohio lead that’s most valuable. Obama has led in the state for most of the year, and that's what has kept him a favorite for reelection, even after his standing nationally saw a swift, post-Denver debate decline.
Without Ohio on the table, Romney has few paths to the presidency. To win without Ohio, he has to sweep almost every other swing state, including Wisconsin and Iowa, where the president holds durable leads. If Obama can hold them, he’ll win the election; alone, without any other swing states, they give him 271 electoral votes.
As in Virginia and North Carolina, demographic change—and in particular, Latino immigration—has turned Nevada (6 EVs) and Colorado (9 EVs) into swing states. At no point in the last two years has Obama lost his lead in Nevada; he has overwhelming support from Latinos in the state, which has kept him above water throughout the election cycle.
Colorado is a little different. There, Latinos provide an important base for Obama, but he needs a solid margin among independents and college-educated whites to seal the deal. His position slipped somewhat after the first debate, but since then, he’s recovered. The meta-average gives Obama a 1.5 point lead in Colorado, and a 4-point lead in Nevada. Romney can still overcome Obama in Colorado, but Nevada is out of his reach.
With that said, there’s one important thing to consider. Matt Bareto of Latino Decisions—among others—has suggested that pollsters are underweighting and undercounting Latino voters. If that’s the case, then Obama’s margin in Colorado and Nevada is larger than it looks. Indeed, there’s precedent for this: In 2010, pollsters missed the huge Latino support for Harry Reid in Nevada and Michael Bennet in Colorado, and thus failed to predict their victories.
There are several other states that you could add to this list if you were taking a more expansive view of swing states—Michigan and Pennsylvania stand out. But these are reliably Democratic states in recent history, even in years like 2004 where the GOP is favored. Team Romney is spending money in both states, but it’s unlikely to make a difference.
If you were keeping count, this is what the picture looks like: Obama holds a greater than 2-point lead in five states—New Hampshire, Ohio, Wisconsin, Iowa, and Nevada—for a total of 281 electoral votes. He holds smaller leads in Colorado and Virginia. Florida is slightly red, and North Carolina appears to be solidly in Romney’s column. If you simply give Obama the states where he holds leads in the meta-average, he wins 303 electoral college votes to Romney’s 235.
As for the popular vote? It's tied, with both candidates just under 47 percent. Real Clear Politics gives President Obama a 0.1 percent lead in its average, Pollster gives him a 0.2 percent lead, and Talking Points Memo’s Polltracker gives him a 0.4 percent lead.
In all likelihood, this is an artifact of the methodological differences between national and state polling. When the votes are tallied, it's almost certain that the winner of the Electoral College will have won a popular vote victory as well.