Fun with Electoral College maps!

A new set of Survey USA 50-state polls on McCain-Obama and McCain-Clinton general election matchups show McCain losing narrowly to either candidate (280-258 to Obama, 276-262 to Clinton), but on the basis of very different electoral maps.

Survey USA has Obama winning nine states that Hillary doesn't: Washington, Oregon, Nevada, Colorado, North Dakota, Iowa, Michigan, Virginia, and New Hampshire. They also say he'll win two of Nebraska's three congressional districts, but lose the state.

Meanwhile, they show Clinton winning five states that Obama doesn't: Arkansas, Florida, West Virginia, Pennsylvania, and New Jersey.

Now, I don't believe for a moment that Obama would lose New Jersey, nor that he would win North Dakota or any part of Nebraska (barring a much more sweeping popular-vote landslide than the map implies). I also don't think Hillary would lose Washington or Oregon (unless things are going badly enough for her that she's also losing swing states like Florida, New Mexico, Pennsylvania, etc.). But the rest of those differences actually make a lot of sense.

Incidentally, if you "flip" the above-mentioned states that I find intuitively suspect -- NJ, ND and NE in Obama's case, WA and OR in Clinton's -- both candidates win by slightly larger margins, with Clinton doing marginally better: Obama 290-248, Clinton 294-244. Personally, I think that, in the end, Obama would do better (I bet he'd get Florida, Pennsylvania and Missouri) and Hillary would do worse (I bet she'd win Michigan but lose Arkansas and Florida... and maybe more). But these results are very interesting, and, on their face at least, pretty plausible.

Comments

Interesting....I am curious about one thing you said....namely that you think Obama would win Florida while Clinton would lose it. In looking at the demographics of Florida I would think they are much more favorable to Clinton than they would be to Obama....Not sure either will win there, but I would suspect that Clinton would have the better chance.

What would also be interesting is to see a survey of what polls might show for an Obama/Clinton ticket versus a Clinton/Obama ticket.

[Brendan Loy]

I just think that the dynamics of the race would be such that Obama would end up doing better nationally against McCain than Clinton would, and that national success would translate into Florida success. But I'm talking through my hat, of course, and past history of my political predictions suggests I'm probably wrong. :)

[Andrew]

Explain to me why geriatric Florida wouldn't prefer geriatric McCain over Obama...?

[JD]

Nebraska's second district is Douglas County (Omaha) and half of Sarpy (suburbs). I think there is potential for Obama there. Not much, but some. I seriously doubt he'd win either of the other districts.

I can also easily see Iowa going for Obama over McCain but McCain over Clinton.

[pegleg]

I suspect Obama will suffer from "the hidden backlash": an unwillingness of polled participants to openly state they won't vote for a Black candidate. I suspect it will be studied for years to come--IFF Obama gets the nomination.

Pegleg

[Scott]

Pegleg is right. No way Obama wins in Ohio, where I live. There are a lot (and I mean a lot) of people who will say they support him but when it comes down to it, will not be able to pull the lever. Sad.

[jlr]

A Democrat losing NJ, PA, and WV but winning VA? That's a bit hard to swallow.

[Sandy Underpants]

This poll, or whatever it is, is a bunch of bunk. How credible is something like this when we're 8 months from the general. Who knows how many more Army recruiting stations will be attacked by "terrorists" throwing firecrackers in the meanwhile. And everyone knows the best way to combat teenagers (..er terrorists) with firecrackers is a grumpy old man.