Sunday, November 11, 2012

The Post-Election Analysis

Now that the election is over and the votes have been counted, I would like to compare Battleground270’s projections to the actual results.  As it turns out, Battleground270 called 49 out of 50 states correctly for the Presidential election and all but one state in the Senate elections. 

As you can see in the table below, Florida is the only state I projected incorrectly.  My predicted margin of victory for Mr. Romney was 0.2%, making it the closest of any of the states.  While it seems Florida will indeed be the closest state, it was actually Barack Obama who emerged victorious, by 0.9%.

However, just calling the state correctly is not a sufficient way to analyze an election projection.  Instead, I have compared my projected margin of victory for each candidate to the actual margin of victory. 

Looking at the results individually, it seems the biggest outliers from my projections were Arizona, Colorado, Michigan, and New Hampshire.  Two of these states, Arizona and Michigan, were not truly taken seriously by either of the campaigns, and, at least in Arizona, there were very few public polls for the Presidential race in the weeks leading up to the election.  

Colorado and New Hampshire, on the other hand, were hotly contested states that were both expected to be extremely close.  Mitt Romney even made his last stop of the campaign in New Hampshire, usually a sign of strength for a state.  However, both states went strongly for the President, by a margin of 4.7% and 5.8%, respectively. 

In the Presidential election, Battleground270’s margins were not drastically off.  On average, I missed the actual margin by 1.785 percent.  Additionally, my projections had a slight Republican bias of 1.092 percent.  Because the projections are made with respect to public polling (along with weighting and personal judgment), this bias can be explained mostly by looking at the polling, as we will do in a minute.

The story was a little different, however, in the Senate.  While 49 out of the 50 states I called were correct, Democrats beat my projected margins by an average of 2.607%.  And remember, that’s just an average.  Some candidates like Senators Claire McCaskill and Bill Nelson ended up clobbering their opponents.  Other races were closer to the predicted results, but in all races but three, the Democrat beat the projections. 

The race I called incorrectly, North Dakota, was one of those states.  Democratic candidate Heidi Heitkamp won in a state that voted for Mitt Romney by nearly 20%.  Other Democratic candidates also did much better than expected.  So what caused such a large difference in expectations and reality?

Karl Rove and the right-wing talking heads got at least one thing right this election cycle: the polls were wrong.  However, where Rove and others were wrong is in the direction that the polls were skewed.  During the waning weeks of the campaign, it essentially became GOP orthodoxy to denounce nearly all public polls.  The idea was that pollsters were oversampling Democrats, and instead, they believed, Romney was actually winning or tied in most states and was likely to go on to win a landslide victory.  

Obviously, that didn’t happen.  Instead, as it turns out, most of the polls were actually skewed towards the Republicans (by ~1% in the Presidential election and ~2.5% in the Senate elections).  As noted above, Battleground270 uses a weighted average of polls to help calculate the expected final vote.  Therefore, Battleground270’s error for the Senate elections is probably at least somewhere close to the overall bias for polling this cycle.

There will be those who try to blame Obama’s larger-than-expected victories on Hurricane Sandy or perhaps other factors.  But in reality, it is likely pollsters were too strict when screening for likely voters.  They undersampled young voters, latinos, African Americans, and, overall, those who no longer use landline phones.  What’s the common link between these groups?  They all tend to vote much more Democratic.  Let’s hope they don’t make the same mistakes in the future. 

Sunday, November 4, 2012

Election Projection: The Presidency

It’s here.  It’s finally here.  Ok, technically, it’s still two days away.  But after months of (mostly) negative ads, billions of dollars, dozens of gaffes, and four debates, the Presidential Election will be decided by just a few people in a few states. 

So you are now likely wondering, “who will win”?  Throughout this article I will go through state-by-state and give my projection, as I did yesterday for the Senate.  However, like my projection of the upper chamber, I am only going to focus on states in which the campaigns have focused their time, money, and resources this year.  Therefore, foregone conclusions such as New York and Alabama won’t be discussed.

Arizona:  No, Arizona is not a swing state.  However, in the early part of the summer, there was a sense that it could have been.  Since Mitt Romney became the GOP presumptive nominee, there have been two polls in the state putting Obama ahead and three others with the President only trailing by a couple points.  However, while the state has certainly shifted towards the Democrats since 2008 when it’s own Senator was running against Obama, the shift is not likely to be enough.  As the demographics continue to change in Arizona, expect it to become more of a battleground in future elections. 
            Final Result:   Romney – 53.1
                                    Obama – 46.1

Colorado:  Colorado isn’t a “must-win” for either candidate.  However, it represents more of a “Plan B” if they fail to win Ohio.   Colorado certainly has shifted more towards Democrats in the last few decades, but if President Obama carries the state, it will be by a much smaller margin than he did in 2008.  Republicans so far have a small lead in early voting, but according to polling, it seems as though Obama may have a small amount of momentum going into Election Day.  This state will be a nail biter, but I predict Obama will win.
            Final Result:   Obama – 49.9
                                    Romney – 48.7

Florida:  The Sunshine State, like Ohio, seems to be thrust in the national spotlight every four years.  Yet, in 2012, though the Democrats certainly haven’t conceded Florida, it really isn’t seen as a must-have for Obama.  The opposite is true for Mitt Romney.  Simply put, there is little to no path to victory that doesn’t run through Florida.  Below is the most likely scenario I could conceive for Romney winning the election without winning Florida. 

Either way, it’s going to be a close election, and there may even be another recount, but for now, it’s Romney with the (very) slight advantage.
            Final Result:   Romney – 49.9
                                    Obama – 49.7

Iowa:  This is the state that was ultimately responsible for Barack Obama capturing the Democratic nomination in 2008.  It, like Colorado, is more part of a Plan B coalition.  However, polls in the state have shown a small but consistent lead for the President.  Further, Democrats currently hold an 11% advantage in early voting.
            Final Result:   Obama – 51.5
                                    Romney – 48.0

Michigan:  Polling has been all over the place in Michigan this year, but despite a few polls showing the race nearly tied, the Romney campaign has not invested much into the state.  It is possible that Romney’s “home state” advantage was partly nullified because he is running as Massachusetts being his home state.  But the more-likely cause for President Obama’s consistent lead is four words: “Let Detroit Go Bankrupt”. 
            Final Result:   Obama – 53.1
                                    Romney – 46.6

Minnesota:  Let’s be clear; Minnesota is not a swing state.  It is perhaps the exact inverse of Arizona, in that if the Romney campaign had invested in it from the beginning, the race would likely be very close now.  That isn’t to say that Minnesota hasn’t narrowed since 2008, when Obama won it by double digits.  The Romney campaign is now making what could be seen as a political head-fake by throwing some money into the state, but it is too-little, too-late.  Obama will win Minnesota.
            Final Result:   Obama – 53.2
                                    Romney – 45.8

Nevada:  As I explained in the Senate projection yesterday, Nevada has been trending democratic for some time now, due mostly to the huge influx of minorities, mostly Hispanic.  Further, the Republican party in the state is very disorganized, whereas President Obama’s early vote push and campaign infrastructure seem to give him the edge.
            Final Result:   Obama – 52.2
                                    Romney – 47.3

New Hampshire:  Unlike other states in the Northeast, New Hampshire has always had more of an independent streak.  The state voted for George Bush in 2000 and Barack Obama in 2008.  Now, polls show a dead heat, with perhaps a slight advantage to the President.  New Hampshire is not necessary for Obama, but its four electoral votes are still seen as highly valuable in the case of either campaign losing Ohio.
            Final Result:   Obama – 51.0
                                    Romney – 48.5

New Mexico:  This state has been one of the most frequently polled non-swing states all cycle.  I’m truly not sure why pollsters are so enamored with New Mexico, especially as it has consistently shown solid leads for the President.  One thing that does make New Mexico interesting is that its former Governor, Gary Johnson, is running as a libertarian.  Polls have shown Johnson pulling about as much support from Obama as Romney, so there is no real chance of a spoiler here.
            Final Result:   Obama – 52.7
                                    Romney – 43.5
                                    Johnson – 3.8

North Carolina:  The Tar Heel State has long been thought of as a long-shot prize for President Obama, just as Wisconsin likely is for Republicans.  Yes, most polls there show a very close race, but the President hasn’t led in a poll in North Carolina since the first debate.  While the President’s early vote operation has been good, he is swimming against the stronger Republican tide.  North Carolina will likely continue to be a swing state and will probably continue to shift towards being more friendly towards Democrats, but it is unlikely Obama will be able to re-create his 2008 surprise victory there.
            Final Result:   Romney – 50.8
                                    Obama – 48.9

Ohio:  Finally, we have come to the Buckeye State.  If you’ve watched the news (or if you live in Ohio), you would think the two candidates are running for President of Ohio.  It is true that no Republican has ever won the White House without winning Ohio, and that is unlikely to be changed in 2012.  Mr. Romney’s path to 270 narrows considerably without Ohio.  To win, he would also have to win Colorado, New Hampshire, and Wisconsin, in addition to holding on to Virginia and North Carolina. 

President Obama’s path to 270 isn’t quite as difficult if he loses Ohio, but it certainly isn’t desired.  To do so, he would need to keep Wisconsin and Nevada, as well as winning Iowa, Colorado, and New Hampshire. 

There is another path.  The President could win in Virginia and then lose one (or two) of the above states.  However, the President is unlikely to lose Ohio yet win Virginia.  Luckily for Obama, he holds a consistent lead in Ohio, and despite what each campaign is saying, the President is not likely to lose there.
            Final Result:   Obama – 51.5
                                    Romney – 48.0

Pennsylvania:  After ignoring it all cycle, the Romney campaign in recent days has decided to make a play for Pennsylvania.  There have been arguments as to if this is a show of strength or weakness for the campaign, but let’s be honest; this is nothing but a hail mary.  Mitt Romney hasn’t led in a single Pennsylvania poll since mid-February.  It’s certainly possible the Romney camp sees something promising in their numbers, but it is more likely that they care more about changing the narrative that it’s all about Ohio (where they are likely to lose).
            Final Result:   Obama – 52.0
                                    Romney – 47.4

Virginia:  In my opinion the national popular vote will look most like the popular vote in Virginia.  The two have been trending with each other all cycle, and like the national polls, Virginia is very much a tossup right now.  Polling in the state has been much more sporadic than other states, with one recent poll showing Romney with a five point lead and another showing a four point lead for Obama.  The winner will likely be decided by who has the better GOTV operations, and that currently appears to be President Obama.
            Final Result:   Obama – 50.5
                                    Romney – 49.2

Wisconsin:  No one really expected Wisconsin to be as competitive as it has become this year.  That all changed when Mitt Romney selected Paul Ryan to be his running mate.  Immediately after the selection, most polls showed a dead heat.  A couple even showed Romney with the lead.  However, the state seems to have shifted back into the “lean Obama” category.  Republicans in the state are still fired up from the Walker recall, but Democrats also have a good organization.  President Obama is likely to win Wisconsin this year.
            Final Result:   Obama – 52.2
                                    Romney – 47.4

Assuming my predictions are correct, Barack Obama will be re-elected with 303 electoral votes to Mitt Romney’s 235, as shown below. 

While it certainly isn’t the blowout many were predicting before the first debate, it also can’t be characterized as “too close to call” or a “tossup”.  People who are saying those things are either purposefully misleading the public or they believe the polls are just flat-out wrong. 

Popular Vote:  You may have noticed that I have not talked at all about who will win the popular vote.  That is because, according to the system we have in the United States for electing our Presidents, it just doesn’t matter.  Yes, there is a distinct possibility that we could have a repeat of 2000 in which President Obama wins the Electoral College yet loses the popular vote, but the chance of that happening is still relatively slim. 
            Final Result:   Obama – 50.2
                                    Romney – 49.1

Saturday, November 3, 2012

Election Projection: The Senate

Just a couple months ago, if you had told a politico that the Democrats would likely keep control of the U.S. Senate, they would have laughed you out of the building.  Even Battleground270 predicted a 50/50 split in early July.  At the time, I had written that it was certainly plausible that Democrats could keep or (even less likely) increase their majority, but I also said it was just as likely that the GOP could capture the majority, perhaps by a large margin. 

The election is not over, and like before, there are still some races hanging on a knife’s edge, but it now appears Democrats are very likely to keep their majority in the Senate.  Let’s examine each of the races.

Of the seventeen races I had labeled as “safe” for each respective party, sixteen of those have remained as such.  The exception is in the race in Pennsylvania (which we will examine in a minute).   The only seat that has been moved into the “safe” column is retiring Hawaii Senator Daniel Akaka’s seat.  I’ve classified the rest as either “leans Dem”, “leans GOP”, or “tossup”.

Leans Dem

The reason Democrats are more likely now to keep their Senate majority is because many races that were previously considered tossups are now leaning towards the Democrats. 

Maine:  Though I technically shouldn’t consider this (yet) as a Democratic pickup, Angus King, the independent, is largely thought to caucus with Democrats in the Senate.  He has endorsed President Obama, and he hails from a state that is likely to overwhelmingly vote for Barack Obama. 
            Final Result:   King – 51.2
                                     Summers – 35.8
                                     Dill - 12.7

Florida:  Bill Nelson has remained as a favorite in his race against Republican Connie Mack.  This is partially due to Mack’s inept campaign, but it can also be said that Nelson has ran a near flawless campaign. 
            Final Result:   Nelson – 52.8
                                     Mack – 47.1

Connecticut:  This race is one of the few examples in which the GOP actually gained ground.  Before the primaries, Democratic candidate Chris Murphy was leading Linda McMahon by nearly double digits.  Since then, however, McMahon caught up in September, only to begin to fall back again in the final weeks of the campaign.  President Obama will likely greatly boost Murphy and help carry him over the finish line.
            Final Result:   Murphy – 52.8
                                     McMahon – 46.8
Massachusetts: This race was one in which the polls had been tied all spring and summer, and it appeared to be a genuine tossup.  Not anymore.  Ever since the Democratic National Convention, at which Elizabeth Warren had a prime-time speaking slot, the Democratic candidate has inched her way up to a solid lead against incumbent Senator Scott Brown.  She will likely be further helped on Election Day due to the fact that President Obama has a staggering 30+ pt. lead against Gov. Mitt Romney in the Bay State.
            Final Result:   Warren – 52.6
                                      Brown – 47.2

Pennsylvania:  Everyone seemed to think Senator Bob Casey didn’t need to worry about his GOP challenger this year.  It seems as though even Bob Casey thought so.  Now, he is in a closer-than-expected race with Smith, but still favored to win, given Pennsylvania’s Democratic tilt and the fact that Barack Obama is expected to carry the state. 
            Final Result:   Casey – 51.8
                                      Smith – 46.9

Ohio:  Like Pennsylvania, the polling has been somewhat unstable but has consistently shown a lead for Senator Sherrod Brown, even after upwards of $40M spent by outside GOP groups against the Senator. 
            Final Result:   Brown – 52.4
                                     Mandel – 47.1

Missouri:  When one thinks of he Missouri Senate race, they think of Todd Akin and his rape comments, and that is precisely why Senator Claire McCaskill is favored to hold on to her seat, despite an increasingly Republican tilt in the state.  GOP groups have been flooding the state in recent days, but it will likely be too-little, too-late. 
            Final Result:   McCaskill – 49.8
                                     Akin – 45.6
                                     Dine – 4.5

Leans GOP

There are only three states that are currently in the “leans GOP” column, and only one of those has become more favorable for the Republican candidate since July. 

Nebraska:  In recent days, Democratic candidate (and former Senator) Bob Kerrey has been getting (relatively) good news.  He has received the endorsement of former Republican Senator Chuck Hagel and was down by only 3 points in a recent poll.  However, the poll is probably be an outlier, and the state has such a Republican-lean that it is hard to picture Kerrey pulling off the win.
            Final Result:   Fischer – 53.7
                                      Kerrey – 45.7
North Dakota:  This race could be considered the inverse of Massachusetts; while they were both essentially tied throughout the summer, the candidate aligned most with the state’s partisan tilt is now pulling away.  If it were a non-presidential year, Heidi Heitkamp could actually pull it off.  However, with Romney at the top of the ticket, Berg is looking like the favorite.
            Final Result:   Berg – 52.1
                                     Heitkamp – 47.8

Arizona:  There hasn’t been any public polling in the Grand Canyon State for nearly two weeks now, but the consensus is the race was trending from tossup to leaning Republican.  The Democrat, Richard Carmona, may still be able to pull off an upset, but his independent appeal likely won’t be enough to grant him victory in the Republican state. 
            Final Result:   Flake – 50.5
                                     Carmona – 47.8


The remaining five states are those in which either the polling is inside the margin of error, or there is some outside factor that I believe puts the race in tossup instead of leaning in a certain direction (..cough..cough..Indiana and Nevada…).

Indiana:  Richard Mourdock was on his way to victory until his debate with Joe Donnelly.  At the debate, he seemed to say that God intends for rapes to occur.  Before the comments, he had a small, but consistent lead over Donnelly.  However, since the debate, Mourdock has been trailing or tied with Donnelly, even by as much as eleven points.  The reason I hesitate to put this in the “leans Dem” column is because Indiana has a heavily GOP lean, and the Republican Party has not disarmed as they did in Missouri.  I personally believe Donnelly is the favorite, but not by the margins polls are projecting.
            Final Result:   Donnelly – 48.5
                                     Mourdock – 47.2
                                     Horning – 4.3

Nevada:  The Silver State is another example of myself being hesitant to put it in a “leaning” column.  While nearly all polls have shown Dean Heller with a small lead (~3 points), Nevada is an increasingly Democratic state, and one that has been visited frequently by Barack Obama this year.  He is expected to win the state, and his organizational advantage could be a boost to Berkley.  Finally, there is evidence to suggest that, like 2010, the Hispanic population, a heavily Democratic constituency, is not being accurately represented in the polls.  However, despite all this, I tend to still believe Heller has a very small lead in Nevada.
            Final Result:   Heller – 48.8
                                     Berkley – 47.2

Montana:  There likely has not been a race in which the candidates have been more consistently close in polling.  Jon Tester is arguably a perfect fit for the state, but in a year when Montana is expected to overwhelmingly vote for Mitt Romney, the Democrat may not survive.  Currently, polls are showing a literal tie.
            Final Result:   Tester – 50.1
                                     Rehberg – 49.7

Virginia:  Since the beginning of the year, Republican candidate (and former Senator) George Allen has sought to tie Tim Kaine to Barack Obama, and judging by the polling all year long, it has worked.  As Obama and Romney were roughly tied in VA in the summer, so was Allen and Kaine.  As Obama opened up a lead throughout September, so did Kaine.  Now, the President is back to being roughly tied in Virginia with Mitt Romney, as is Kaine with Allen.  However, it seems as though Kaine is doing slightly better now against Allen than Obama is against Romney, and for that reason, Kaine is favored to win.
            Final Result:   Kaine – 51.0
                                     Allen – 48.6

Wisconsin:  There is probably not another race with two candidates who are further apart from one another ideologically.  Though she has done a great job positioning herself in the center for the general election, Congresswoman Tammy Baldwin is one of the most liberal members of Congress, and not exactly a perfect fit for the state.  However, former Governor Tommy Thompson had to join himself at the hip of the Tea Party in order to win his primary, and with limited funds, he is now struggling to win with the broader electorate.  This race, like Montana is about as close to “tossup” as you can get.
            Final Result:   Baldwin – 50.2
                                     Thompson – 49.1

In all, Battleground270 is projecting Democrats will end up with 54 seats while Republicans will lose a seat (46).  That Democrats would be considered favorites to actually gain seats in 2012 would have seemed preposterous even a couple months ago.