State of the Race – November 4 (POTUS Prediction)

We are now two days away from Election Day and I have been tracking this race for nearly 3 months. Combined the campaigns spent nearly $1 billion trying to convince you to elect them to the presidency. After all that time and money, I am ready to predict – a VERY long night. I have been saying for some time and laid out in my post on September 29 that this election had many similarities to 2004. If you recall, that election night went into the early morning hours on Wednesday and it wasn't until John Kerry conceded the election around noon on Wednesday that we knew it was over. In my post on September 29, I provided a chart with similarities between the 2004 election and this year's election. I think after Tuesday, we will be able to add another similarity – Swing State That Decides the Race: Ohio. It was that state that put George W. Bush over the top in 2004 and I expect that it will be the key to either of these candidates winning on Tuesday.

As reminder, here's a chart of where I saw the states in my analysis yesterday. Of course, if the states go exactly according to this prediction, then the “Tossup” states do not matter as President Obama will have 281 electoral votes, 11 more than he needs to win reelection. However, I do not think the analysis is quite that easy.

	Safe	Likely	Lean
Obama	CA, CT, DE, DC, HI, IL, ME, MD, MA, NJ, NY, RI, VT, WA	MI, MN, NM, OR, PA	IA, NV, NH, OH, WI
Romney	AL, AK, AR, ID, KS, KY, LA, MS, MT, NE, ND, OK, SC, SD, TN, TX, UT, WV, WY	AZ, GA, IN, MO	NC
Tossup	CO, FL, VA

A note on my analysis: since the first debate at the beginning of October, I began tracking all the polls on a spreadsheet that were done in the 11 battleground states that I previously identified (CO, FL, IA, MI, NV, NH, NC, 0H, PA, VA & WI). When I talk about average poll numbers below, it is an average of all those poll numbers done in a specific state over the last month. I'll also be discussing the seven-day average which looks at a state’s poll numbers from last Sunday to today.

For starters, let's look at the best case scenario for both candidates. I've seen people on both sides predict well over 300 for "their" candidate, but I think both are pipe dreams. Let's start with President Obama's best case scenario. I have it represented in this map:

In this scenario, Obama wins all the states that I have on his side, plus two out of the three “Tossup” states. I know that he has spent resources trying to win Florida, but of all the battleground states, Florida and North Carolina are the only states that have an average poll number that favors Romney. Most political commentators will tell you that undecided voters usually will break against the incumbent, so if Obama is already behind in those two states, I find it hard to believe that he'll be able to catch up. As such, I think Florida and North Carolina are both out of his reach. If on election night Obama ends up winning Florida, then I think it will be a very long night for the Romney campaign as there is really no way for Romney to win without Florida.

However, Florida is not the only state that has major significance. On election night, you can expect a lot of discussion about FLOHVA. Every election, there are usually big battleground states that commentators will point to as being the ones that are most critical. This year, the commentators have taken to calling them FLOHVA (Florida, Ohio & Virginia). This is a slight preview of what I was going to say in tomorrow's post regarding what to watch for, but these three states are critical if Romney is to win. If Obama wins any of these three states, he will most likely win the election. For Romney to win, he probably needs to sweep these three states.

As for Romney, his best case scenario is depicted below:

This map has Romney winning all of the states on his side of the ledger, plus all the tossups and winning all of the “Lean” Obama states except for Wisconsin. In fact, this map shows Romney winning every state that George W. Bush won at least once except for New Mexico. Of the “Lean” Obama states that would be least likely to go to Romney, I would point to Nevada due to their demographics, specifically their higher percentage of Latino voters. In looking at the average poll numbers in the battleground states, there have been four where no poll has shown Romney in the lead. Some polls have shown the candidates tied in these states, but Romney has never led in Michigan, Nevada, Pennsylvania or Wisconsin. As such, I find it hard to believe that he will be able to pick one of those states off, but his best chance is in Nevada if the Latino turnout is down. In other words, a more realistic best case scenario for Romney would be 285 electoral votes. A final note on Nevada and the Latino vote directly relates to New Mexico. As I said above, the one state that George W. Bush was able to win once that Romney has not been competitive in is New Mexico and this is largely due to the higher percentage of Latinos in New Mexico. This is a post for another day, but the Republican Party needs to do something about the Latino vote as today it is just states like New Mexico and Nevada that are harder to win, but in the future if the Latino vote continues to grow, it could be Arizona and Texas and there's no way a Republican can win without those two states.

Now that we've looked at the best case scenarios for both candidates, where do I think things will fall? As I said above, I think it will all come down to Ohio. In looking at the 30 day averages as I said above, Obama has never trailed in any polls in Michigan, Nevada, Pennsylvania and Wisconsin. If we give him those four states, he is at 253 electoral votes while Governor Romney remains at 191. However, Romney has had a pretty consistently in North Carolina and has only trailed in one poll that came out within the last month. This would put Governor Romney at 206 electoral votes. Like I said above, I also find it hard to believe that Obama will be able to pull out Florida, so this puts Romney at 235 electoral votes.

If you look at this scenario, Ohio would already have enough electoral votes to put Obama over the magic number of 270. In other words, even if we gave Romney all the other battleground states (CO, IA, NH & VA), it would not be enough. As far as where these other four states will fall, I think we can break them into two different categories. First is Colorado and Virginia which have been the two states where the average polling has been the closest. As I said above, conventional wisdom is that undecideds will break against the incumbent and these are two states where I could see this happening. As such, I predict that those two states will go to Romney. As for Iowa and New Hampshire, these are two states that are known politically for their early involvement in the primary process and polling in both of these states has been unreliable over the years. That said, the polls in both of these states seem to be moving towards the president over the last week or two. If we give these two states to Obama, then he would lead Romney in the Electoral College 263-257. Once again, it would come down to Ohio.

It's funny how a week changes things. Last weekend I was prepared to write a prediction indicating that Romney would win Ohio and the presidency 275-263, but this week I am inclined to believe that Obama will make history in becoming only the second president to win reelection with fewer electoral votes than he received in his first election. (For more about this historical fact, see my post on October 20). What changed my analysis? Two different things that may very well be interrelated. First of all, Hurricane Sandy hit the East Coast and dominated the news coverage for most of this past week. After the final debate, the polls seem to indicate that any momentum Romney had gained from the first debate had slowed down and things were remaining fairly static. This past week was Romney's last chance to really drive home an argument on the economy and where he felt President Obama had failed. Unfortunately for him, Sandy allowed Obama to do the one thing an incumbent can do to help his reelection and that is act like the president. With Obama touring the areas of destruction and providing comfort to those who were affected by the tragedy, Romney was relegated to expressing sympathy for the victims from afar and not appearing that he was doing anything that would be deemed too political. Ultimately, I think Romney struck the right balance, but this was really a losing proposition for him. In fact, I think the only way that Romney could have been helped by this tragedy was if Obama had made a misstep, which he did not.

The second thing that made me change my mind was what started happening with the polling. One of the main reasons why I had thought about predicting Romney win was that the national polling was favoring Romney despite the fact that Obama seemed to be leading the state-by-state polling. While it was certainly within the realm of possibility that we would have another election where the popular vote winner lost the Electoral College, I found it hard to believe it would happen in a reelection scenario – something that has NEVER happened. In fact, I even toyed with the idea of predicting a split decision where Romney would win the popular vote, but Obama would get reelected based on the Electoral College. However, over the last week the national polling has moved towards the president ever so slightly and most national polls now have the race either tied or have Obama with a slight lead. Additionally, the same thing is happening in the state-by-state polling. Of the 11 battleground states, the average poll number over the last seven days is better for the president compared to the 30 day average in every state except for Pennsylvania and that is a bit misleading as there have only been three polls done in Pennsylvania over the last week. In fact, seeing the uphill climb that Romney has in Ohio might very well be the reason why he is making a last-ditch effort to win Pennsylvania even though he made no campaign stops and did little advertising in the state prior to these last few days. The question is whether it is too little, too late.

Many people believe that the polling done and released the last weekend before the election give a good indication for who will win the race if you compare it to the polling trend. In other words, you will be able to see if the race is breaking in one direction or the other and that should tell you who will win on Election Day. If this is true, then that is good news for President Obama. It is this exact scenario that had me consider putting Virginia in Obama's category as it is the state where I have seen this trend the most pronounced. The average polling in Virginia over the last week is a full point higher for Obama then the 30 day average is. As such, I would not be surprised at all if Obama did win Virginia as well. As I said above, if Obama wins any of the FLOHVA states, he wins. The scenario is clearly shown here because if Obama does win Virginia, that would give him 294 electoral votes meaning that he could even afford to lose the 18 electoral votes in Ohio. However, I will stick with the scenario where it all comes down to Ohio and predict the president will get reelected by a margin of 281-257 and the map will look like this:

Interestingly, if my predictions on the White House, Senate and House are correct, it will mean that everything will remain the same and a country that is sick of business as usual in Washington will have done the unthinkable and returned things to exactly where they were over the last two years. Well, whatever happens, and whatever your persuasion, I encourage you to make sure that you vote on Tuesday if you have not done so already. Tomorrow I'll post my election night preview and give you a few things to look for on election night that may tell us where things are headed. Happy Voting!!