The Sports Report

A Statistical Take on Sports and Politics

Final Election Report

Disclaimer: This final election report is the conclusion of a series of looks at the 2008 Presidential Election. From analyzing the delegates in the primaries, to starting to pick apart the competitive states in the general election back in June, I have been covering all aspects of this year’s election process. This article includes all of the previous information in one final review.

Introduction

John McCain is not running in the 2008 Presidential election simply against Barack Obama. Any casual political observer would tell you that an enormous part of his campaign thus far has been distancing himself from the incredibly unpopular Presidency of George W. Bush. To a certain extent, this entire idea is the framework for the rest of my report: John McCain is at an extreme disadvantage in this election because of conditions well beyond his control. As Alexander Mooney at CNN writes, “Amid a failing economy, a Republican party in tatters, and a rival who has presented himself as an acceptable alternative, McCain’s fate may ultimately be out of his hands.”

In addition to the fact that George W. Bush has been one of the most unpopular Presidents in the modern era, the size and power of the Democratic Party has grown substantially since he was re-elected in 2004. However, the question of whether Democratic nominee Barack Obama has had a tremendous impact on this rise in party identification is not a direct issue in this election, as both candidates will have to make do with the current political nature of the United States. What is important to note is that it is a given in this year’s election that there will be more identified Democrats voting than identified Republicans, and from that information the only thing that can be altered is the rate of return on those Democratic or Republican voters, along with the tendencies of Independent-leaning individuals.

All of those factors would already lean towards a heavy Democratic advantage in this November’s election, but there is one other x-factor to the success of Barack Obama’s campaign: community organization. From the University of Dayton’s massive canvassing operations to the overwhelming flow of personnel in St. Louis, Missouri, it appears that Obama and the Democratic National Committee are running the largest grassroots campaign in history. One can only expect that within the final two weeks of the election, this party will not back down just yet and will march on with the largest “Get Out The Vote” drive ever. Reports are only starting to come in from these operations, and I can report first-hand here in Dayton that the organization is unlikely anything ever from a politician.

National Popular Vote

The size and power of the Democratic Party matters a great deal to the national popular vote, and as Nate Silver of www.fivethirtyeight.com wrote, “McCain is likely to run into something of a wall very soon here, brought about the Republicans’ substantial disadvantage in partisan identification.” I show this idea visually just below here, by using recent national averages that indicate a lead of about 5 percent nationwide for Barack Obama:

Current National Popular Vote Breakdown

Current National Popular Vote Breakdown

(Just for the record: I did not pull these numbers out of nowhere. The party identification percentages are a trend line I created based off weekly updates from Rasmussen Reports. I controlled the turnout numbers to keep the established difference among Democratic and Republican voters, along with limiting the actual voting of Independent voters, and having 65.50% of the voter-eligible population participating in the election. These are somewhat sweeping generalizations, but I believe they represent some of the most established information in the business.)

Currently, my averages derived from Rasmussen’s information show that there will be approximately 5.59% more Democratic voters than Republican voters. What does this 5.59% mean? Technically, this percentage depends upon how many people vote in total, but I am already assuming in my calculations that 65.50% of the voter-eligible population will vote. This then figures out to be about 7.6 million more Democratic voters, and with that advantage in hand, along with a current lead among Independents, Obama has quite a significant lead in the national popular vote.

Before I explore the specifics making this equation equal zero, I would first like to point out some very historic information in the table above. In 2000, Al Gore won about 86% of the Democratic vote, but here in 2008, I have Barack Obama winning nearly 90% of the Democratic vote. The reason for this change in my numbers is that Gallup’s party breakdown has had Obama with leads of 87-89 to 7-9 in the past three weeks. These are drastic improvements for Obama, as his numbers from two months ago were 78-79 to 13-14 in back-to-back weeks. Since Obama has a huge lead over such a large percent of the entire voting population, it will be difficult for McCain to make up that difference.

If we take everything besides Independent voting behavior as a given in this election, what would McCain have to do to tie Barack Obama? Since Obama currently leads by 4.32 percent among Independents, adding to his 5.13 lead nation-wide, McCain would have to win by a margin of 17.88% among Independents to tie Obama. I created a formula in which you can change the percentage or number of million-voter difference between the Democrats and Republicans, and within this spreadsheet, one can see how McCain can possibly win the popular vote.

Even if McCain improves his standing in just above every single metric, it will be difficult for him to offset the current standardized difference. In the table shown below, I subtract from the Democratic identification margin by one million at a time and in addition, I show what would happen if the party defector margins return to their totals from about three weeks ago:

Equalling the National Popular Vote to Zero

Equalling the National Popular Vote to Zero

The only scenario in this entire table where McCain forces Obama to win among Independent voters is when Obama has lost three points off his advantage among Democrats, McCain gained about three-quarters of a point among Republicans, and the Democratic identification margin drops down all the way to 3 million. As well, it is important to note that all of these totals are below the established standard in the current political landscape, and represent more of a floor in terms of his production. The top row of information would be one of Obama’s worst possible outcomes in the national popular vote, but he would still win in the end because of his current lead among Independents.

In conclusion, excuse my language, but screw the polls for the next two weeks. Here is what really matters in this election: there will be more Democrats voting, and given Obama’s established advantage among those voters, it will be tough for McCain to win the national popular vote. Obama leads by nearly identical margins among Democrats, as does McCain among Republicans, and even leads by a significant margin among Independents. All of these factors combine to make McCain’s odds of winning the national popular vote very slim.

Electoral College Breakdown

While the above tables may look really cool and meaningful in the way I presented them above, they ultimately carry no meaning on November 4. The process of becoming the next President of the United States is not contingent upon building a national majority; the winner is the one who subsequently wins 270 electoral votes from the Electoral College. The two may seem somewhat correlated, but in the end, the only thing that matters is the number 270 (or more) flashing up on television screens and newspaper headlines nation-wide.

Since my first-ever foray into general election numbers in July, I have been tracking major movement in the Electoral College. The following list is a brief recap of swing state movement since I last touched on the subject:

Old Weak McCain States (used to be 52 electoral votes, now 26):

Alaska, 3 votes – from 5.5 to 12.8 (very safe now)
North Dakota, 3 votes – from 3.4 to 6.4 (still weak)
North Carolina, 15 votes – from 2.9 to 2.1 (toss up now)
Missouri, 11 votes – from 2.6 to 2.1 (toss up now)
Georgia, 15 votes – now at 7.6 (weak now)
West Virginia, 5 votes – now at 5.3 (weak now)

Old Weak Obama States (used to be 85 electoral votes, now 107):

Minnesota, 10 votes – from 10.9 to 7.6 (still weak)
Oregon, 7 votes – from 8.8 to 10.4 (still weak)
Wisconsin, 10 votes – from 8.5 to 8.4 (still weak)
Pennsylvania, 21 votes – from 6.6 to 7.2 (still weak)
New Hampshire, 4 votes – from 6.4 to 4.6 (now barely weak)
Iowa, 7 votes – from 6.1 to 9.7 (still weak)
Washington, 11 votes – now at 10.1 (weak now)
New Jersey, 15 votes – now at 9.7 (weak now)

Old Toss Up States (used to be 110 electoral votes, now 111):

Florida, 27 votes – from R 3.1 to D 0.1 (still toss up)
Nevada, 5 votes – from R 2.5 to D 1.0 (still toss up)
Indiana, 11 votes – from R 1.1 to R 2.7 (still toss up)
Montana, 3 votes – from R 1.0 to R 4.9 (now R weak)
Virginia, 13 votes – from D 0.8 to D 3.1 (still toss up)
Colorado, 9 votes – from D 3.2 to D 4.0 (still toss up)
Ohio, 20 votes – from D 3.2 to D 1.1 (still toss up)
Michigan, 17 votes – from D 3.9 to D 5.9 (now D barely weak)
New Mexico, 5 votes – from D 4.9 to D 5.9 (now D barely weak)

I am not sure what I was thinking not having Georgia, West Virginia, New Jersey and Washington on the list before, but they are on now after I changed around my standards during the course of the last three months. The new standards for the Weak and Toss Up states are that they have to be within six points at least once all year (one exception made for Arkansas recently, more on this later). This means that all the other 29 states (remember, the District of Columbia makes it 51 states in the Electoral College) have been outside of a six percentage point margin since I began tracking information in July. The complete information above is in an odd format, but it helps to outline the trend that has occurred during this time.

From the middle of July to about September 12, there was a steep national trend towards John McCain and the Republican Party. Among these 22 states (not including Alaska, as the inclusion of Sarah Palin on the Republican ticket made this state safe), every single one leaned slightly towards McCain, with the average movement being by about 3.73 percentage points. Following that up, from McCain’s high-water days in mid-September to Obama’s peak performance around October 11, Obama improved his numbers in every single one of these states by an average of about 7.24 percentage points. Now since mid-October, McCain has had some mildly positive results, averaging an improvement of about 0.78 points in these states, but it has not been even close to the rapid pace of Obama’s huge rise last month. This equals out to an average Obama gain in these 22 states of about 2.73 since July:

Toss Up State Breakdown

Toss Up State Breakdown

The above table once again shows what has happened to these 22 states, but instead of comparing their new averages to their old information from July, I show them side-by-side with my most recently inputted results from www.fivethirtyeight.com. Now, I will sidetrack a little to analyze where we are in the election going into the 22 most competitive states. Obama has a lead of 157 to 137 among those other 29 states, and this total is one that I also reported in July. These are established totals for very easy victories for the two candidates, but other websites and analysts may have different numbers from that point onward. Including all of the moderately, and lightly shaded blue states, along with the red states in the table above, Obama’s lead has now expanded to 264 to 163 with 111 electoral votes remaining.

This shows that in order to win the election as it stands right now, John McCain would have to win all of the states in the purple region up above. Even if Obama only won Nevada out of that group of states, he would hit his magic number of 269, which should be enough to help him win anyway. This is such a long shot not only because is it improbable that McCain wins all eight toss up states, but because of the fact Obama is currently leading in seven of the eight. He has a solid lead in both Colorado and Virginia, and many pollsters are actually starting to move those states into the solid Democratic category.

Question & Answer

Which Kerry/Bush state is most likely to switch sides here in 2008?

Iowa, without a doubt, is the most likely state to switch sides, as the writers of both www.270towin.com and evstrength.blogspot.com will argue. The authors of these two phenomenal websites both pointed out Iowa first, and an Obama win in Iowa would trim the Republican lead to 279 to 259 in the Electoral College. From there, analysts point to New Mexico and Colorado as two very likely states, and both are leaning towards Obama by significant margins right now. If Obama only wins New Mexico, Colorado, Iowa and Kerry States, then he would win the election with 273 electoral votes.

In addition, as Sean Quinn pointed out on my favorite blog, New Hampshire is the currently most competitive state that Kerry won back in 2004. With this means in terms of my analysis of the Electoral College, is that all eight states that are toss-ups right now are all states that George W. Bush won in the 2004 election. This further proves how McCain is playing defense in typically Republican states, while Obama is on the move in just about every state nation-wide.

What are the chances of Obama winning Ohio? Will it matter?

As of 3:00 PM on Thursday, October 23, Nate Silver has Obama’s winning percentage in Ohio at 71%. While I think that may be a little bit of a stretch considering the entire span of polling during the past three months, he presents an intriguing argument as to how easy this election could be for the Democrats. The first question here is a very valid one, and certainly one that is on people’s minds all across the state, but I want to emphasize the second question: Will this even matter in terms of who wins the Electoral College? My answer to that question is a resounding no way.

If Obama only wins Iowa, New Mexico, Colorado and Virginia, four states that Bush won in 2004, but are leaning Democratic by significant margins, then he will amass 286 electoral votes and coast to victory. It is highly unlikely that Obama would win Ohio before Colorado or Virginia, and thus in the greater picture does not mean that much for the Democrats. The most important states in my mind are Colorado and Virginia, as they are the extra electoral votes that will seemingly easily push Barack Obama into the White House. Certainly, however, the Republicans will continue to compete in the state until the very last day, as without the Buckeye state it is incredibly unlikely that they would have any chance at winning overall.

Toss Up State Polling Since July

Toss Up State Polling Since July

How can McCain win the Electoral College? What is he planning on doing?

The previous responses lead me into this very important question. The American Research Group had a very interesting piece recently on their website that said, “Because of McCain’s electoral inefficiency, even if McCain were to proportionately increase his national ballot share to 49% for a tie, Obama would still lead with 316 electoral votes. Obama is running a race in the Electoral College while McCain is running a national race.” I am somewhat skeptical of the precise details of that statement as it stands right now, and will probably stand on election day but it is incredibly important to note that there are very few possible paths for John McCain right now to win 270 electoral votes.

One path, as I mentioned earlier, encompasses McCain winning all eight of the purple states on my graph above. The chances of winning every single one of those Bush states from 2004 is quite remote, but there is one other alternative option that it looks like the Republicans are starting to pursue: Pennsylvania. The Keystone State has not been competitive at all in the most recent polling, but it looks like some Democrats are starting to become concerned because of increased Republican attention to the state. If McCain miraculously pulls of Pennsylvania, then he could let Obama win Colorado and still win the election, but the combination of these two results seems very odd to me.

What is the most likely result as the election stands right now?

First, I just want to say that the smear blame game will continue to go on. Both sides, directly or indirectly, are responsible for spreading partial truths about the opposing candidates, and here is one such example from the conservative end of the aisle, and another from the liberal perspective. It is probably “above my pay grade” to conclude which side is completely responsible for initiating this kind of campaigning, but it will continue to be a part of this year’s election and every subsequent Presidential election.

Secondly, As Scott Elliott from www.electionprojection.com said recently, “Honestly, though, an Obama landslide is becoming more and more likely by the day.” In common sense terms, it is seemingly unfeasible for John McCain to win the Electoral College. As long as Obama continues to lead in only Colorado, and not even Virginia, Ohio, Nevada, Florida, etc., then he will win this election. As www.270towin.com recently stated, “McCain has to hope for something significant to occur in the last couple weeks that causes enough people across the country to reconsider their vote that the race becomes closer (some sort of October surprise).   Absent that, he’s going to have to pull out all 8 of those undecided states or somehow pick off Pennsylvania and win most of those undecided states.

My Artistic Representation of the Polls

My Artistic Representation of the Polls

Advertisements

October 23, 2008 - Posted by | Politics | , ,

1 Comment »

  1. This is very impressive research and analysis, very little that anyone could probably disagree with that much. There are a great many unprecedented factors in this election and it will be interesting afterwards to see how they all affected the outcomes in the different states.

    Very well done.

    –Rick

    Comment by Rick Morris | October 27, 2008 | Reply


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: