Post by Ramen on Sept 24, 2014 1:17:09 GMT
Here's a roundup of campaign forecasts from the best sources. Skip down to the bottom for the "Summary" if you don't want all the specifics.
The Monkey Cage - The Washington Post
www.washingtonpost.com/wp-dre/politics/election-lab-2014
Senate - most likely is 51 Republican seats,
House - most likely is 244 Republican seats, which is a 10 seat gain
TheUpshot - The New York Times
www.nytimes.com/newsgraphics/2014/senate-model/
Senate - Republicans have a 60% chance of winning the majority. This is up slightly from a few weeks ago, but is still down from the high in late August.
Nate Silver's 538 forecast:
fivethirtyeight.com/interactives/senate-forecast/
Senate - Republicans have a 58.5% chance of taking the majority.
Political Scientist Predictions:
www.huffingtonpost.com/2014/08/29/political-scientists-midt_n_5739064.html
Nice summary of their models, although these forecasts vary widely.
Summary
In sum, it's looking like a very good year for Republicans in the Senate and an okay year for them in the House. They should pick up between 4-8 seats in the Senate, with 9-10 possible but unlikely. They should pick up between 5-15 seats in the House, although some forecasts (like Jim King's) predict even larger gains.
Here's a list of the closest Senate races, with current polls and thoughts about the race.
Alaska
Sullivan (R) - 45%
Begich (D) - 45%
Arkansas
Cotton (R) - 47%
Pryor (D) - 44%
Colorado
Udall (Inc - D) - 47%
Gardner (R) - 47%
This race probably leans towards Udall. In a state like Colorado, even during a bad election year for Democrats, it's going to be hard to unseat an incumbent who has not done anything extremely controversial. Udall is not locked in any kind of scandal PLUS the other CO Senator (Bennett) is in charge of the DSCC PLUS Udall has an unprecedented ground operation. Remember that the polls had Colorado very close for Obama, and that Obama won by 5% in Colorado because of the ground operation. Turnout efforts have big effects in that state. If the polls show a tie going into the election, I have to lean towards Udall.
Georgia - not actually all that close
Purdue (R) - 47%
Nunn (D) - 41%
This one leans heavily towards Purdue. There's not a compelling reason for independents and leaning Republicans to support Nunn, and there's not a real weakness with Purdue.
Kansas
Roberts (R) - 42%
Orman (I) - 41%
This is a crazy race. The Democrat dropped out and the independent candidate surged into a tie with the Republican incumbent. Normally, an incumbent in a solid red state during this election cycle would be an easy choice. However, the governor in Kansas is blamed for a lot of their fiscal problems, and that anger is transferring over to the Senate race. Still a tossup, especially if the Democrats succeed in not having a candidate on the ballot.
North Carolina
Hagan (Inc - D) - 45%
Tillis - 41%
Incumbency will probably pull Hagan through this, and the polls have tilted her way slightly. Unless something dramatic happens, I have to think this leans Democratic.
Iowa
Ernst (R) - 45%
Braley (D) - 44%
Iowa's a funky state when it comes to partisan politics. It went for Obama twice, but the incumbent senator is a Republican, the legislature has split partisan control, the governor is Republican, and the congressional delegation is split. I literally have no idea how this one turns out.
The other races are actually not all that close. Michigan's Democratic candidate has a 5% lead, Mitch McConnell in KY has broken out a 51-44% lead, Landrieu is behind by 3% and probably is not going to win reelection, and the others have pretty clear front-runners.
The most likely pickups for Republicans in the Senate are Louisiana, Arkansas, Montana, West Virginia. They would then only need to gain seats in two of the following to get control of the Senate - Alaska, North Carolina, or Iowa. Obviously, if Roberts loses in Kansas, it makes it harder for them to gain the majority.
So in the final analysis, we know there's a very good chance the Republicans will pick up at least 4 seats, and possibly as many as 8 seats. But the chances of control of the Senate are still a coin flip, based on what I've seen.
The Monkey Cage - The Washington Post
www.washingtonpost.com/wp-dre/politics/election-lab-2014
Senate - most likely is 51 Republican seats,
House - most likely is 244 Republican seats, which is a 10 seat gain
TheUpshot - The New York Times
www.nytimes.com/newsgraphics/2014/senate-model/
Senate - Republicans have a 60% chance of winning the majority. This is up slightly from a few weeks ago, but is still down from the high in late August.
Nate Silver's 538 forecast:
fivethirtyeight.com/interactives/senate-forecast/
Senate - Republicans have a 58.5% chance of taking the majority.
Political Scientist Predictions:
www.huffingtonpost.com/2014/08/29/political-scientists-midt_n_5739064.html
Nice summary of their models, although these forecasts vary widely.
Summary
In sum, it's looking like a very good year for Republicans in the Senate and an okay year for them in the House. They should pick up between 4-8 seats in the Senate, with 9-10 possible but unlikely. They should pick up between 5-15 seats in the House, although some forecasts (like Jim King's) predict even larger gains.
Here's a list of the closest Senate races, with current polls and thoughts about the race.
Alaska
Sullivan (R) - 45%
Begich (D) - 45%
Arkansas
Cotton (R) - 47%
Pryor (D) - 44%
Colorado
Udall (Inc - D) - 47%
Gardner (R) - 47%
This race probably leans towards Udall. In a state like Colorado, even during a bad election year for Democrats, it's going to be hard to unseat an incumbent who has not done anything extremely controversial. Udall is not locked in any kind of scandal PLUS the other CO Senator (Bennett) is in charge of the DSCC PLUS Udall has an unprecedented ground operation. Remember that the polls had Colorado very close for Obama, and that Obama won by 5% in Colorado because of the ground operation. Turnout efforts have big effects in that state. If the polls show a tie going into the election, I have to lean towards Udall.
Georgia - not actually all that close
Purdue (R) - 47%
Nunn (D) - 41%
This one leans heavily towards Purdue. There's not a compelling reason for independents and leaning Republicans to support Nunn, and there's not a real weakness with Purdue.
Kansas
Roberts (R) - 42%
Orman (I) - 41%
This is a crazy race. The Democrat dropped out and the independent candidate surged into a tie with the Republican incumbent. Normally, an incumbent in a solid red state during this election cycle would be an easy choice. However, the governor in Kansas is blamed for a lot of their fiscal problems, and that anger is transferring over to the Senate race. Still a tossup, especially if the Democrats succeed in not having a candidate on the ballot.
North Carolina
Hagan (Inc - D) - 45%
Tillis - 41%
Incumbency will probably pull Hagan through this, and the polls have tilted her way slightly. Unless something dramatic happens, I have to think this leans Democratic.
Iowa
Ernst (R) - 45%
Braley (D) - 44%
Iowa's a funky state when it comes to partisan politics. It went for Obama twice, but the incumbent senator is a Republican, the legislature has split partisan control, the governor is Republican, and the congressional delegation is split. I literally have no idea how this one turns out.
The other races are actually not all that close. Michigan's Democratic candidate has a 5% lead, Mitch McConnell in KY has broken out a 51-44% lead, Landrieu is behind by 3% and probably is not going to win reelection, and the others have pretty clear front-runners.
The most likely pickups for Republicans in the Senate are Louisiana, Arkansas, Montana, West Virginia. They would then only need to gain seats in two of the following to get control of the Senate - Alaska, North Carolina, or Iowa. Obviously, if Roberts loses in Kansas, it makes it harder for them to gain the majority.
So in the final analysis, we know there's a very good chance the Republicans will pick up at least 4 seats, and possibly as many as 8 seats. But the chances of control of the Senate are still a coin flip, based on what I've seen.
