House +70, Senate +9

I've already laid out my House predictions, now it's time to lay out my Senate predictions. The topline numbers are +70 House and +9 Senate.

Arkansas -- R. This race has actually been decided for about 4 months, yet shockingly a few pundits kept it on a less than safe Republican rating.

Connecticut -- D. Margin of error could be surprising.

Colorado -- R. Michael Bennett has probably run one of the better Dem campaigns, but I don't think it's going to be enough to hold off Buck.

Delaware -- Republicans nominated a bad candidate. Even a generic conservative R might have had a chance here -- which is truly remarkable -- but O'Donnell doesn't. D

Florida -- R. Crist probably made the right move by trying to engineer a Meek dropout, even though it was a long-shot. I think it may have hurt his final numbers though. I think it's a closer question whether Meek beats Crist than whether Crist beats Rubio.

Illinois -- R. Polls show Kirk ahead, though it's a blue state and voters are very unhappy with their choices. I think Kirk pulls it out, though some Green support in polls will likely melt into Democratic support just enough that it might put the Dem nominee over the top. Again, the enthusiasm gap probably gets Kirk there.

Nevada -- Harry Reid v Sharron Angle really tests the theory of just how far voters will go to vote against an incumbent they despise. Angle is historically incompetent as a candidate, with a plethora of gaffes and her current avoid-the-media-at-all-costs strategy. Harry Reid is despised by Nevadans, but while he is not a candidate who is made for TV, he does know how to punch and campaign. If this were any other election -- even 1994 -- I think Reid might pull it off. But this is 2010, and I predict Angle squeaks it out. R.

New Hampshire: R. Paul Hodes rode the wave into the House, and now he's going to ride the wave out. Investors often talk about beta, I think Hodes is a great example of beta in a politician: he moves with the market.

Ohio: R. How amazing is it that one of the closest House members to W (and later joined the administration) basically just had his D opponent give up against him? Wow.

Pennsylvania: R. How amazing is it that Pat Toomey is the likely winner? Watching Toomey's moves over the next six years will be very interesting. He wins in 2010, he wins in 1994, but in very few other environments can a politician of Toomey's reputation win a statewide federal race in Pennsylvania.

Wisconsin: R. Feingold is just a very weird fit for Wisconsin who had gotten a little lucky until now. If I were him I'd move to New York and try again.

* Finally, the really difficult races to predict. *

Alaska: R. This is the rare race which is less affected by national trends. Miller seems to be trying to lose; meanwhile at a local talk show host's urging, a few hundred Alaskans put their names on the ballot as possible write-ins, including several Lisa M. That step probably increases Miller's chances to win...but also increases McAdams's chances, which is a net gain for Democrats. Strategic voting might change the voting decision of a small but significant portion of the swing and Dem electorate. The most likely case is still a Miller win (60-65%?), but Dems probably have like a 10-15% chance of fluking this seat, just like they did with Alaska's other Senate seat in 2008. Give Murkowski a 25% chance or so, even given that legitimate write-in candidates probably run 10-15% behind what they would if their name was on the ballot. A truly awful campaign from Miller is what gives her a remarkably high possibility for a write-in candidate.

California: D. If the marijuana initiative wasn't on the ballot, I might be inclined to think that Fiorina could beat Democratic incumbent Barbara Boxer. Boxer might be the most underwhelming member of the Democratic Senate caucus who has essentially won previous races in California by yelling "Democrat! Roe v Wade!" ad nauseum. Her closing TV adbuy is Obama's endorsement speech of her. A recent poll had Obama under 45% job approval in California, but even if it were in the 50-55 range, I think it's a weak ad. The average undecided or least-committed-to-Boxer voter probably neutral to lukewarm on Obama, so it doesn't really persuade them. The argument is that Boxer is ineffective, and the ad doesn't fight that. I think the race is likely to be closer than polls indicate, although I think the marijuana initiative might bring out some younger voters who will vote for Boxer.

Washington: R. In a recount? If history is a guide, Rossi probably has a disadvantage in a recount, and he's now on the ballot statewide for the 3rd time in 6 years without winning, so I subtract a few points for that too. Obama won by 17 points here in 2008 while Rossi lost his rematch by 7. Patty Murray is a pretty undistinguished incumbent, probably a point or two behind a generic D. This is truly a tossup, and if I could fudge and predict 8.5 seats instead of 9 seats, then I probably would. But forced to pick, I think the enthusiasm gap will just barely tip the scales in Rossi's favor. That's an awfully tough pick to make in an Obama +17% state.

West Virginia: D. Dem Manchin's approval rating is something like 72/24, while Republican Raese's is something like 47/47. WV is a unique state in that it's really not a Republican state -- a generic D is a favorite over a generic R in a gubernatorial race. However, WV is a very anti-Obama, anti-cap-and-trade state. Manchin is super popular as governor -- does Manchin = Obama Rubber Stamp work? I'm inclined to think not, as Manchin has very clearly delineated himself from Obama, although Raese is airing some clips of Manchin supporting ObamaCare. The average WV swing voter at this point strongly dislikes Obama, thinks Manchin is a good governor, and is lukewarm about Raese. At one point in my predictions, I wrote "Democrats will win one and Republicans will win one of the West Virginia and Washington races. I just don't know which one." This race could surprise and be an R, but my best guess is Manchin squeaks it out.

Senate: +9. As I already called, House +70.

Races that are most likely that I called incorrectly: West Virginia and Washington. California if the wave is truly large. For idiosyncratic reasons, Nevada and Alaska. If the wave is smaller -- which I doubt -- Illinois and Colorado too.

By the way, what's with my stringing-all-my-words-together recently? No idea.

Posted by Evan @ 11/01/10 07:18 PM


Previous Entry | Home | Next Entry


Re: Delaware. As I've said before, Castle ran a lackluster campaign and didn't deserve to win the nomination over O'Donnell. It seems to me that one of the features of the anti-incumbency wave is that is doesn't discriminate: it isn't strategic, no matter how much anyone might wish it weren't so....

Posted by yara @ 11/01/10 10:29 PM

One might have thought that Delaware Republicans were tired of losing and would want to take advantage of their only opportunity to win a statewide race for the next 30 years.

Posted by Rick Perry vs World @ 11/01/10 10:37 PM

Add Comments

No flames or impolite behavior. HTML will be stripped. URLs will be transformed into hyperlinks.