Pennsylvania

Final numbers don't seem to be up yet, but the result is roughly 54.5% Clinton to 45.5% Obama, with the nine-point difference representing over 200,000 votes. The percentages are similar to those for Ohio, suggesting that the result reflects Pennsylvania's similarly Clinton-friendly demographics rather than any sudden understanding of Obama's profound vulnerabilities in the general election.

Nevertheless, it shows once again that Clinton is more appealing than Obama to the swing voters in the swing states where the general election will actually be decided.

The popular-vote margin has narrowed dramatically. If Florida and Michigan are counted (which they certainly should be in evaluating the candidates' popularity -- those are real voters, whatever the squabbles between their state party leaders and the national party), then RCP's count actually has Clinton ahead by 121,943. This is misleading since it omits the 238,168 "undecided" votes in Michigan, the majority of whom must have been Obama supporters. But even if most of those votes are tallied to Obama, his popular-vote lead is significantly under 100,000 -- out of more than 30,000,000 cast. (Yes, thirty million. One out of every ten Americans has voted in a Democratic primary or caucus, and out of people old enough to vote, it's more like one out of every seven.)

So the popular vote -- counting everyone who actually voted -- is practically a tie. And Clinton is more popular than Obama in the swing states. That's her case to the superdelegates. Hope is still very much alive.