If you're a Karl Rove conspiracy theorist, get ready for some head spinning as you try to figure out this one: maps produced by the consulting firm Rove heads show that, right now, in a hypothetical head-to-head matchup, Hillary Clinton would be beating John McCain in the Electoral College. In the same situation, Barack Obama would be losing to McCain. (You can download the maps in PDF form here.)
More specifically, this latest edition of the Rove firm's maps -- they've been done before -- shows Clinton making a win all but impossible for McCain. She wins 19 states, plus the District of Columbia, for a total of 259 electoral votes. (There are 538 total Electoral College votes; 270 are needed for victory.) In the same analysis, McCain wins 25 states and 206 electoral votes. Six states -- New Mexico, Missouri, Iowa, Minnesota, Michigan and New Hampshire -- and 73 electoral votes are shown as tossups (defined as being within 3 percentage points either way). In this situation, if Clinton were to win almost any combination of two tossups, or either Michigan or Missouri alone, she'd win the presidency. On the other hand, the only states McCain could afford to lose would be Iowa, New Hampshire and New Mexico, and even then, if he lost Iowa he couldn't lose either of the other two.
The Obama/McCain map, by contrast, shows a very close race with McCain ahead slightly. In this map, McCain wins 27 states and 238 electoral votes; Obama wins 15 states, plus the District of Columbia, for 221 electoral votes. Eight states -- New Mexico, Colorado, Nebraska, North Dakota, Iowa, Michigan, Ohio and Virginia -- and 79 electoral votes are tossups.
ABC News' Jake Tapper, a former Salon reporter, obtained the maps. In a post on his blog, Tapper takes a look at the big differences beyond the numbers themselves and notes:
Clinton would make competitive some states that Obama would lose -- such as Missouri and New Hampshire -- and she would win others outright, such as Arkansas, Florida, Ohio and West Virginia.
On the other hand, McCain would handily win beat Clinton in some states that Obama made competitive, such as Colorado, North Dakota and Virginia.
There are some good reasons to doubt the accuracy of these maps. First, as Tapper observes, "States are allocated in Rove's exercise based on an average of public polls, which many pollsters would tell you is a rather unscientific way to look at the data." Also, examining both maps I noticed some things that just don't pass the smell test.
First of all, the Obama/McCain map shows both Nebraska and North Dakota as tossups. That's hard to believe. In 2004, Nebraska went 66-33 for George W. Bush over John Kerry, and North Dakota went 63-36. There's nothing in either state's demographics to suggest such a big change is in the offing. And in the same map, Wisconsin is shown as a McCain state, by 4 points. But in 2004, John Kerry won a narrow victory there, and the general trend among voters nationally since then is obviously anti-Republican. Also, though primary elections really aren't a very good predictor of general election success, Obama certainly seemed to have at least some support in Wisconsin when he trounced Clinton, 58-41, in that state's Democratic primary. Clinton, too, is shown as losing Wisconsin -- that's also questionable, despite her primary loss there.
Additionally, both maps show Michigan as a tossup, which is debatable, at best. It's true that in the 2000 presidential election, Michigan's voters were separated by only 3 points, and that the margin in 2004 was only 4 points. But in both cases the state went Democratic, and in 2000, Al Gore's victory there might have been even bigger were it not for Ralph Nader, who got 2 percent. And with the current state of the economy and Michigan's acute sensitivity to such economic troubles, it's fair to expect that it will trend even further away from the incumbent president's party. Also, if Obama were the nominee, the state's larger-than-average African-American population (granted, only slightly above the national average, but every bit counts, especially considering Obama's dominance in that demographic, and remember that some outlying states pull down the average) would be an additional advantage.