2014 Election Predictions

Quickly, my completely unscientific predictions for the election today:

Senate: +9 for the Republicans (MT, WV, SD, AR, AK, LA, CO, IA, NC and holding all of their open seats)

House: +11 for the Republicans (NC 7, UT 4, NY 21, AZ 1 and  2, IL 10 and 12, FL 26, WV 3, NH 1, MN 8, IA 1 while only losing CA 31)

Governor: -2 for the Republicans (losing PA, KS, ME and AK, gaining AR and MA)  The Dems would only gain a net of 1 seat since Alaska would go to an Independent.

Amateur 2012 Election Predictions

My completely unscientific election predictions:

President: Obama 294-244 (Obama wins Ohio and Virginia; Romney wins Colorado and Florida)
Popular vote: 51-49 (two-party vote)

Senate: Republicans +1  Gain Nebraska, North Dakota and Montana; Lose Maine and Massachusetts

House: No net change.  242-193.

I actually went through and assigned each seat based on my predictions and came out with exactly the same numbers as last election.  This result is pretty unlikely given that there are new seats, lost seats and redistricted seats due to the new census.  I have 10 seats switching from R to D, 7 switching from D to R and both member-on-member fights going to the GOP. These numbers don't necessarily add up because of reapportionment and redistricting new seats.

 I tend to be skeptical of big gains from redistricting since there seems to be some backlash by voters to it.  However, I do see multiple seat gains for the Republicans in North Carolina and for the Democrats in Illinois because many Representatives in those states won by close margins last time as it was.
 
Edit: Right after I posted this, I decided to flip-flop Colorado to Romney.

Way Late 2010 Election Prediction Analysis

I was thinking about doing another House of Representatives prediction.  Since my last prediction hit the Republican gain exactly, and I did not do a full gloating analysis post, I am doing that first.

While my overall prediction last time of a 63-seat gain for the Republicans was exactly right, I did miss about 8-12 seats in both directions, so my actual prediction may not have been that great.

I think there were three causes of my missed predictions.

 1.  I just did not have enough knowledge about individual candidates.  For instance I picked the Republicans to win Arizona 8, but Gabrielle Giffords was probably a much more appealing candidate than was apparent to me as an outsider, and she held it for the Dems.

2.  My main point that I missed for the Republicans was their strong showing in the suburbs.

3.  The main thing I missed for the Democrats was that I expected African-American turnout to drop more than it did.

Those last two things only affected a couple of races each.  The truth is that there are actually very few true suburban districts.  Either they are suburban-urban and therefore solid Democratic or suburban-exurban and solid Republican.  Very few districts are suburban and thus toss up.  The few seats like this are concentrated around New York and Chicago.  With majority-minority districts, there are not very many districts where African-Americans make up a sizable but not determinative proportion of the electorate.

Long Live the Electoral College

Garett Jones has a post at Econlog stating why he is a fan of the electoral college.  His basic argument is that it lowers regional conflict by forcing politicians to aim for the median voter in swing states as opposed to running up the score in individual states, and thus overemphasizing regional differences.

Another advantage I see is that it weights the votes of each state by their population, not by the number of people voting.  Therefore, if we see a scenario where a natural disaster hits one or two states and depresses their voting, it lessens the probability of that affecting the election.

State Poll-National Poll Divide

One thing that has bugged me about the coverage of the election was the idea that there was a high probability of Romney winning the popular vote by a decent amount (>1%) and Obama winning the electoral college.  Joe Scarborough seems to be one of the worst offenders at this.  Much of the polling seemed to suggest that not only is this a possibility but a fairly likely event as well.  However, it never seemed intuitively to me to be a all that probable.  Yes, there have been a few elections where the popular vote loser won the electoral college, but in each of those cases the popular vote was quite close.

Well today a couple of the better analysts have examined the problem and basically said that one of the polls has to be wrong.  Sean Trende says when you aggregate all of the states, the state polling is not consistent with the national polling.    Nate Silver suggests that the state polling is better.  While Jay Cost seems to be skeptical of the state polling, at least in Ohio.  It is good to see some confirmation of my own intuitive thoughts.

One caveat is Hurricane Sandy.  If the hurricane really depresses voting in New York and New Jersey, it probably will not have much of an impact on the electoral college as Obama would likely win both states anyway.  However, it could impact Obama's popular vote total enough that Romney wins the popular vote by a bigger amount (i.e. 1-2%) and Obama still wins the electoral college.  There again this scenario would not be able to explain the current polling discrepancy.

2010 Election Prediction Analysis Part I

Well given my prediction for the 2010 Elections, I'm claiming victory/validation. The Republicans have picked up a net of 63 seats in the House, exactly the same as my prediction. Of course I'm claiming victory now before the final two seats are called and make my prediction not exactly correct. Also, I only got the correct number by missing about ten seats in each direction. Lastly, my Senate prediction was not really all that great, over-predicting the Republican gain by two seats. However, those last three points are just details.

2010 Election Predictions

Since my 2010 Econ Nobel Prize prediction was so on the mark, I have decided to predict the 2010 US elections. Gains for the GOP:

House: +63 (net)

Without listing every seat, here is the list of some marginal seats I predict victory in:

AZ 8, ID 1, IL 14, KY 6, MO 4, NC 2, 7 and 8, NH 2, NY 23, OH 6, OR 5, and WV 1.

Seats lost by the GOP:

DE AL, IL 10, and LA 2. (Barely holding HI 1 and FL 25)

Marginal seats GOP misses barely:

AR 4, AZ 7, CA 20, CT 4 and 5, MI 9, MN 8, NY 1 and 24, OH 18, VA 9 and 11. (They also lose races that they should have done better in IA 3, IN 2, and PA 12).

 

Senate: +8

Gains: AR, CO, IL, IN, ND, NV, PA, and WI (GOP loses close race in WA)

The GOP holds all currently held Senate seats.

For additional perspectives and invaluable resources in making my predictions:

538

Jay Cost