The Democrats (33 delegates) - Barack Obama suffered, arguably, his worst news cycle of the campaign in the run-up to the March 4 primaries. You can follow it on the Gallup tracking poll or on the polls of Texas and Ohio, or in the gap between early Texas voting and election day voting: Obama's trajectory shifted at the 11th hour, and he lost. Every indication now, however, points to an Obama recovery. He dispatched Clinton in Wyoming by 23 points (comparable to his Feb. 5 victory in just-as-white, just-as-barren North Dakota), and he should crush Clinton in Mississippi.
This is the deep, deep South, largely rural, and there is no state on the radar at all like Mississippi. But we can compare it to Alabama to get a sense of whether Clinton has gained any momentum in these less "relevent," less-hotly-contested states that she lost badly in the run-up to March 4. In Alabama, 51 percent of voters were black, and they went for Obama over Clinton 84 to 15. But the white vote gap was the largest it's been in any state since John Edwards decided to spend more time with his family—72 to 25 for Clinton. Those numbers, and the numbers of Democrats who 1)rule out voting for the other candidate or 2)rule out a "dream ticket," will be worth watching. Also worth watching will be the white/black proportion of the vote. In 2004, a majority—56 percent—of primary voters were black. And check out how Republican voters break. Will the "Limbaugh" effect continue, or are Republican cross-over voters going to play it straight out of fear of a resurgent Clinton? (The fact that both of the GOP districts, the 1st and 3rd, are holding very competitive primaries, should cut down the crossover vote.)
I'm predicting an Obama win of about 17 points to win 6 or 7 of the 11 statewide delegates. In the heavily black 2nd district (Jackson and the Mississippi Delta) Obama will win in a landslide, scoring 5 of 7 delegates. The rest of the congressional districts, with 5 delegates, will split 3-2: Obama will probably win the 3rd district, Clinton will win the 1st (northeast) and 4th (Gulf Coast).
The Republicans (36 delegates) - Will John McCain win? Oh, obviously. But will he cross the crucial Bush line, the political yardstick I carved while I was writing this post? In 2000, Mississippi held its vote at the exact same time in the cycle as it did this year, right after the front-runner locked up the nomination. That year George W. Bush won 88 percent of the vote against the still-active Alan Keyes campaign (5.6 percent) and the inactive McCain insurgency (5.5 percent). This year Huckabee is still on the ballot, and Paul is still leading his rEVOLution. Will McCain, who has lost every Deep South state save South Carolina to Huckabee, hit 88 percent? I doubt it: I expect him to break 80 percent but watch a sizable chunk of the vote go to Huckabee, then to Paul. (The high watermark vote for an inactive candidate in this election cycle has been John Edwards's 10.2 percent in Oklahoma.)
Indiana-7: There's a smidgen of suspense in this race to replace the late Rep. Julia Carson. While the Indianapolis-centered district voted by 16 points for John Kerry (up, actually, from Al Gore's 12-point margin in 2000), Indiana Republicans were buoyed by the surprise 2007 election of Mayor Greg Ballard. The Democrats nominated Carson's inexperienced, Islamic-convert grandson Andre: The GOP nominated a rising star from the state House, Jon Elrod. If the Democrats can take Denny Hastert's seat, can the GOP defeat a nepotistic dud in Indianapolis? Probably not. Libertarian candidate Sean Shepard will take protest votes from anti-Carsonites who aren't impressed by Elrod, and enough Democrats will turn out to get Carson to 50 percent at least.
UPDATE 7:26: It's too early to say anything, but Carson is leading the Indiana race. From 0 to 2 Muslim congressmen in less than two years: Virgil Goode was right! Except that Carson, like Keith Ellison, is an African-American who converted to Islam later in life, not a scary immigrant.
UPDATE 8:05: The networks won't call it yet, but with about half of the vote being black and going 91-9 for Obama, and Obama scoring an Alabama-like 27 percent of the white vote, he'll win going away. It might be closer to 56-43 than the 58-41 I expected. Tim Russert's comment that there's a "growing racial divide" here was probably off the cuff, but it's still wrong. This is how white Deep South voters roll.
UPDATE 8:18: John McCain might not break 80 percent of the vote, or even 76 percent. The first exits show a protest vote of about 11 percent for Huckabee and 7 percent for Paul. Ninety percent of voters say they like him, but among the (shrinking) anti-illegal immigration vote he only scored 62 percent.
UPDATE 8:29: Here's a fun number: 37 percent of Democrats have a favorable impression of John McCain. Clinton wins those voters 62-37.
UPDATE 8:32: Maybe the Limbaugh effect is alive. In Alabama, Obama split the independent vote with Clinton and lost Republicans by only 7 votes. Here he's winning the independent vote, narrowly, by 3 points. The Republican vote is going to Clinton by about 50 points. It would be specious to pin that all on Limbaugh, though. The bottom line is that Obama's "I can win the deep South because I'll turn out black voters" is bunk. He has no shot whatsoever at Mississippi or Alabama: His 99 percent of the black vote will be swamped by a landslide of white votes for McCain, many of them from Democrats.