CAN EDWARDS WIN?….Over at Slate, William Saletan has some intriguing amateur statistical analysis about John Edwards’ chances against John Kerry in the upcoming primaries. I was all ready to blow it off, but as I kept reading I realized that his numbers actually made a bit of sense (although he presents them in the wrong order, which made it harder to follow his argument).

Here’s the short version: exit polls indicate that a large number of people are picking a candidate not on the basis of the issues, but simply on the basis of who’s most likely to beat Bush. So far, these people have voted very heavily for Kerry.

But if you look at exit poll results for independents and crossover Republicans, it turns out that Edwards is running stronger than Kerry among these groups. Since these people are likely to make the difference in the general election, you can make a pretty good argument that Edwards is actually more likely to beat Bush than Kerry.

So Saletan’s question is this: what happens if the voters who care only about beating Bush figure out that it’s actually Edwards who is better positioned to do this? Will they abandon Kerry and go for the more electable Edwards?

I don’t know, especially since most voters don’t pay attention to the minutiae of exit polls. But it’s an intriguing thought. And since I happen to like John Edwards a lot (he was my second choice candidate after Clark), it’s a comforting thought too. The race isn’t over yet.