Cross-posted at Politics and Other Random Topics
It's funny, in some ways this has been a bad few weeks for Democrats politically (Dino Rossi's entrance in Washington State against Patty Murray and the thing with Blumenthal in Connecticut) but at the same time, the Senate picture actually looks better for the Democrats.
My most recent changes are to move Connecticut back to Likely Democratic from Leans Democratic and to move Nevada from Leans Republican to Toss-up.
The Connecticut thing should be pretty obvious, the New York Times screwed up pretty bad on their several stories regarding Blumenthal (plus Linda McMahon's idiotic bragging about giving the Times the story basically killed any chance of it seriously damaging Blumenthal).
Nevada's an interesting one, because Harry Reid hasn't magically become more popular than he was, but his polling against all three challengers has definitely improved. While I had been classifying the race as Leans Republican for my purposes, I'd always believed that Harry Reid was the incumbent who was most likely to come back from the grave and win simply because his opposition is so weak and his war-chest is really nothing to sneeze at ($9 million Cash on Hand, compared to his opposition who have a combined Cash on Hand amount of about $400,000, with that coming largely from Lowden with $200,000).
Now then, with the official caveat that the election is still several months away and there are any number of things that could happen in the meantime, let me give you my first preliminary prediction for the Senate races:
Democrats take the following seats from the Republicans: Ohio, Missouri, and Florida (I think Charlie Crist wins and that he caucuses with the Democrats, thus I consider it a Democratic gain).
Republicans take the following seats from the Democrats: North Dakota, Delaware, and Arkansas.
Honestly, I think for all the hoopla about Democrats getting routed in the fall, there's a very good chance that the Democrats break even for Senate races (to get this out of the way, I believe that Democrats will hold Indiana, Colorado, and Illinois despite polling to the contrary).
The best-case scenario for the Democrats right now is probably keeping their seat losses limited to North Dakota and Delaware (some Democrats are holding out hope that New Castle County Executive Chris Coons can pull off an upset, but I doubt it) and somehow hold Arkansas (frankly, Arkansas is bordering on being a lost cause as well), and then taking Ohio, Missouri, New Hampshire, Kentucky, North Carolina (this one's definitely a sleeper for the Democrats), and maybe catch Chuck Grassley off-guard in Iowa (to be fair, this is a bit of a stretch, as Grassley, despite showing some slight weakness, is still a pretty damn popular incumbent who isn't likely to lose). This scenario gives Democrats somewhere between 61 and 63 seats with the Republicans at between 39 and 37 seats.
Conversely, the best-case scenario for the Republicans is to hold onto to their competitive open seats (Ohio, New Hampshire, Kentucky, and Missouri), protect North Carolina (which is probably going to be pretty easy if the Republicans hold all of their open seats), take all of the Democratic open seats (save for Connecticut), knock off Reid, Lincoln (or the open seat, depending on what happens in the run-off), and Bennet, and then beat Barbara Boxer in California (frankly, despite their candidate recruitment coup, I don't think the Republicans really have a prayer of defeating Patty Murray). This scenario gives the Republicans 50 seats (which basically means that Democrats will maintain control of the Senate unless Lieberman decides to screw the Democrats and switch, which I wouldn't put past him).
My current prediction might seem a bit optimistic for some, but it's still worth mentioning that even now, it's still reasonably possible that the Democrats can break even or even gain a seat or two in these senate elections.
(To reiterate, this is a preliminary prediction of the status of a series of elections that won't take place for another five months, so these predictions are very much subject to change).