Races are ranked in order of their likelihood of changing parties by November 2010, accounting for all factors such as potential retirements, primary challenges, and so forth.
Likelihood of party switch has increased since October's rankings.
Likelihood of party switch has decreased since October.
1. Delaware (D-Open) -- The reason I keep this one in first position is because Beau Biden has yet to declare and there's a small chance he won't run, in which case it's a walk for Mike Castle. If Castle wins close to 100 percent of the time that Biden doesn't run, and 55 percent of the time that he does, then that puts his odds above 60 percent overall.
2. Connecticut (D-Dodd) -- You probably know how skeptical I tend to be of internal polls. So when an incumbent candidate publishes an internal poll that shows him behind his leading Republican opponent, you know that he's in trouble. With that said, this race is a long way from over; Republicans could still wind up nominating a goofy candidate like Linda McMahon, and there likewise remains a chance that Dodd won't be the Democratic nominee.
3. Missouri (R-Open) -- Robin Carnahan's polling has been very stable, continuing to show her with a very slight lead. That's probably because her opponent, Roy Blunt, is a rather vocal member of the sitting Republican Congress, which is losing popularity just as quickly as the Democrats in Congress are. He's the wrong candidate for this type of cycle.
4. Nevada (D-Reid) -- The reasons why Reid could lose are fairly self-evident, both on a polling and a narrative basis. So here's why he could still win: (1) $$$$$$, (2) it'll be harder to brand him as ineffectual if he passes health care, (3) Nevada is a machine state and Democrats control the machine, (4) Republican candidates are untested.
5. Colorado (D-Bennet) -- We need more polling here.
6. Arkansas (D-Lincoln) -- No doubt Blanche Lincoln has the fight of her life on her hands, but perceptions that she's dead meat are formed principally by Rasmussen's polling, whereas most other polls have continued to show her with a slight lead. Following my advice, she also managed to stay out of the spotlight at the end of the health care debate as Ben Nelson and Joe Lieberman led the three-ring circus instead.
7. Ohio (R-Open) -- Some of the polling has started to move Rob Portman's way. Fundamentally, though, this will still be about whether Portman is cast as the Responsible Deficit Hawk or the Bushie Cabinet Hack, a dynamic that should keep this one close straight through to November.
8. Kentucky (R-Open) -- Upgraded because it looks like Rand Paul will be the Republican nominee. Kentucky is not a very libertarian-friendly state; it tends, rather, to be in the opposite corner, conservative on social issues but supportive of an economic safety net. So -- although Paul will surely raise millions from activists all over the country -- there's a chance this plays out something like NY-23, where a national message proves to be a bit tone-deaf in an idiosyncratic corner of the country.
9. Pennsylvania (D-Specter) -- Democrats are going to want to make this one about Pat Toomey, who in an ordinary cycle would meet Rick Santorum's fate. But 2010 may not be an ordinary cycle -- the larger the national "wave" the less the particulars of the candidates may matter (this was how Santorum got elected in the first place). And Arlen Specter's flip-floppin' gives Toomey plenty of ammunition and might allow Toomey to control the tempo of the campaign. So, upon further reflection, I've become a bit more inclined to believe the polling here, which shows the race to be a toss-up. But the polling also shows that more voters move into the undecided camp in the event that Joe Sestak is the nominee, in which case the strategy of attacking Toomey could be more effectual.
10. New Hampshire (R-Open) -- New Hampshire tends to be a very good bellwether for national trends, perhaps because it's a well-educated state where people watch a lot of news and really enjoy the sport of politics. And since the national momentum is likely to favor the Republicans in November, that means Paul Hodes is running into a bit of a headwind.
11. Illinois (D-Burris) -- Rasmussen actually shows a tiny bit of momentum toward the Democrat Giannoulias. Mark Kirk is in a somewhat awkward position; he was elected repeatedly in Democratic-leaning IL-10 by running unashamedly as a moderate, but that's not where the GOP zeitgeist is this year and he's accumulated a more conservative voting record. I don't care how bad the national environment is -- I'm not sure you can win by reflexively opposing Barack Obama's agenda in Barack Obama's home state.
12. North Carolina (R-Burr) -- If there's one thing I'm certain of it's that North Carolina belongs in the #12 position; it marks the dividing line between those races that can basically be considered to be toss-ups and those that clearly lean toward the incumbent party.
13. Florida (R-Open) -- I'm pretty sure that Marco Rubio now has to be considered the favorite in the Republican primary. I'm not a huge believer in "trends" -- meaning, I don't think you should necessarily assume that just because a candidate has been gaining ground means that he'll continue to do so. But in this case, the trends have been pretty bleepin' overwhelming and have been accompanied by a disastrous narrative that has developed around Crist. In any event, whether it's Rubio who emerges or a wounded Crist, this race has come somewhat back on the radar screen as an opportunity for Democrats.
14. North Dakota (D-Dorgan) -- There are several polls showing John Hoeven with a huge lead over Byron Dorgan if he decides to enter, but all were conducted by Republican-leaning outfits. I'm a little bit suspicious of this sort of polling when the national party is trying to recruit a candidate into the race. Also, as a sitting governor whose term doesn't expire until 2012, Hoeven would have to explain to the electorate why he quit the governorship to challenge a fairly popular incumbent, which could instantly take a bite out of his numbers. So, I'm upgrading this race slightly on upside potential for the Republicans, but Hoeven remains an underdog to enter (having already blown through his self-imposed deadline of September) and is hardly a shoo-in if he does.
15. Louisiana (R-Vitter) -- Nothing in the way of new polling here. Vitter's failings are personal rather than political and are somewhat old news. I'm just not sure this is a winnable seat for Democrats in a state that has gone very red very quickly.
16. Iowa (R-Grassley) -- Democrats have gotten a semi-interesting candidate in the form of Roxanne Conlin, who has 73 percent name recognition in Iowa. Although Grassley, who had built up a large reservoir of goodwill, has probably not hurt himself quite enough to lose, the health care fight undermined his brand as a bipartisan dealmaker. This race deserves more polling.
17. New York (D-Gillibrand) -- Would Rudy have made the same decision not to run for national office if the Eunuch Bomber had struck on Thanksgiving rather than Christmas?
18. Wisconsin (D-Feingold) -- The buzz that Tommy Thompson might run seems to have died down. Nor do I know why Thompson would challenge a fairly popular incumbent for the Senate when there's an open seat in the governor's mansion instead.
19. California (D-Boxer)
20. Indiana (D-Bayh) -- Republicans have gotten themselves a candidate here in the form of six-term Congressman John Hostettler. Although Hostettler is a notoriously weak campaigner and Bayh has historically won by huge margins, this race could conceivably become interesting at the upper bound conditions of Republican momentum.
21. Arizona (R-McCain) -- It's getting a bit late for J.D. Hayworth to primary John McCain and Democrats would have to scramble to find their own candidate even if he did. But one contrary indicator: a Republican poll sponsored by a McCain-backing group tested negative messages about Hayworth, which suggests that somebody is at least a little bit nervous.
22. Texas (R-Open?) -- Bill White will run for governor rather than Senate, which robs Democrats of most of their opportunity -- especially since it has become less clear that Kay Bailey Hutchison will resign her seat in the first place.
23. South Carolina (R-DeMint) -- A PPP poll shows that Jim DeMint could potentially be vulnerable. But Democrats have almost no infrastructure in the Palmetto State, which probably makes this an academic question.
24. Georgia (R-Isakson) -- I'm going to start to downgrade some of these races where there's no fresh polling and no credible opponent.
25. Hawaii (D-Inouye) -- Ditto.
26. Massachusetts (D-Open) -- It's worth polling the special general, which will occur in just three weeks, but I'm pretty sure that Massachusetts will soon have a Senator Coakley.
27. Oklahoma (R-Coburn)
28. Alaska (R-Murkowski)
29. Kansas (R-Open)
30. Maryland (D-Mikulski)
31. Washington (D-Murray)
32. Alabama (R-Shelby)
33. South Dakota (R-Thune)
34. Vermont (D-Leahy)
35. Oregon (D-Wyden)
36. Utah (R-Bennett)
37. New York (Sr.) (D-Schumer)
38. Idaho (R-Crapo)