Updated Senate race prediction as we get closer to Election Day 2014. <br /> Overall I believe that the Republicans will underperform compared to earlier expectations I had of them sweeping the elections this year. Republicans seem to be making the mistake of assuming that a second midterm is automatically good for their party and bad for the Democrats since Obama is in office while disregarding that we've had a lot of exceptionally bad second midterm election years. And frankly, while 2014 is not a great year, the Democrats at least have a better economy and Obama is perceived as being strong on foreign policy and social issues just are not working for the GOP this time around. <br /> <br /> While turnout will hurt the Democrats, as will the map which naturally favors the GOP, the momentum just does not seem to be on the GOP's side. In any other year the +4 would be something to brag about, this year though not so much.<br /> <br /> The only four seats this season I see switching hands are Democratic seats in hostile dark blue territory (in Atlas terms). Competitive races in Georgia and Kansas which could pull lots of time and money out of otherwise competitive races to defend the home turf. Both Georgia and Kansas should be pretty close but ultimately stay in the GOP column. Perdue will be Nunn by maybe .5-2% which will cause a lot of restless nights for Republicans in the near future given the state's trends towards the Democrats. Dems will start winning in Georgia pretty soon, but my gut tells me that 2014 is not that year. Pat Roberts is polling badly now and losing to an Independent candidate, however in the coming months his campaign should be able to distance itself from Sam Brownback's failed governorship enough to win over lukewarm Republicans but by an embarrassingly low margin.<br /> <br /> On other races:<br /> <br /> Kentucky: A traditionally Democratic state, however I don't believe Grimes can overcome the inevitable trend towards Republicans in the state due to the bad perception of Democrats as anti-coal. She will lose by maybe 5-7 points.<br /> <br /> Arkansas: This race is polling closer than I thought it would be, but I don't see Pryor really motivating turnout in this race. The era of the Blue Dog died in 2010. Democrats can win in states like Arkansas, but they won't win by nominating people like Pryor who are two feet to the left of Lamar Alexander. A new base needs to be built, especially with the younger more progressive generation. Cotton victory by 1-3 points.<br /> <br /> Colorado: Udall has incumbency and the political trend of the state's climate going for him. Victory is quite likely, but not guaranteed. Colorado is still somewhat a purple state (though still leans Dem) and a possible game changer could result in a Republican victory (which again I consider highly unlikely). Prediction: Likely Udall victory of 5-8 points.<br /> <br /> Iowa: I predict a very tight race between two gaffe prone candidates. In the end I predict a Braley victory, particularly given that Ernst's "cutting pork" ad has become something of an election year joke that will overshadow any of the Braley's accidental elitist remarks about farmers. Prediction: Tossup election in favor of Braley. Braley victory of .2-2 points.<br /> <br /> South Dakota: Larry Pressler is a pretty popular figure in South Dakota, having served as a US Senator for several terms. However, in this race he'll probably end up being more of a spoiler for the Democratic candidate, that is assuming he doesn't end up becoming a "defacto Democrat" candidate a la Charlie Crist in 2010. Nevertheless, I see a vote split among Democratic/liberal voters that likely guarantees Mike Rounds the victory. Very likely Rounds victory of 7-15 points.<br /> <br /> Montana: An explanation is probably not at all necessary, but I figure I should give a few words here on the current Democratic strategy in Montana. Look, I would love it if an actual left winger was elected US Senator of Montana after the idiot Lieutenant Governor sunk his chances of winning by plagiarizing his Master's degree paper or whatever. I must commend Democrats on their experiment here, but anyone who knows Montana should know how futile this is. Yes there is kind of a left wing populist element in the state, but it will be very easy for Daines and crew to portray Curtis as an out of control gun grabber (which is actually pretty false, she said she supported background checks and not really anything about a blanket gun ban, but perceptions matter) to sink her chances of winning. I could be wrong about all of this and she could motivate enough people to get mad at the GOP to go out to the polls to vote but I highly doubt it at the moment. It's kind of freaking hard to recover from something like the Walsh Scandal. If the MT Dems somehow do with a Curtis victory it will be the political comeback story of the ages. Prediction: Daines victory of 10-18 points.<br /> <br /> Michigan: Gary Peters has taken the lead in recent months in most polling. However, Land could possibly score an upset if she plays her cards right or if the environment changes drastically in the next 60 or so days. At the moment though it's lean Peters. Prediction: Peters victory of 5-10 points.<br /> <br /> North Carolina: The state's drift towards purple state status should help Hagan retain office with a more Dem friendly electorate if she's able to connect some of the unpopular austerity measures of statewide Republicans with the national agenda. She's a gifted politician who should not be underestimated. Prediction: 1-3 point victory for Hagan.<br /> <br /> West Virginia: Capito is practically a state establishment already with her long tenure in Congress. Combine that with the continuing anti-Democratic trend in West Virginia and a "who the hell is Natalie Tennant?" factor I see this as likely GOP. Capito victory of 15-20 points.<br /> <br /> New Hampshire: Yeah, Scott Brown is not going to win. In this day and age nobody falls for blatant carpetbagging. Especially against an experienced incumbent like Shaheen. If he does win I'll eat my shorts. Prediction: Likely Shaheen victory of 2-8 points.<br /> <br /> That's it for now.