Skip to content
October 29, 2012 / Political Fluency

Obama’s Firewall

[Heavily armed soldiers reach their enemy target]

Rabinowitz: [Reaching over a picket fence] It’s locked. From the inside.
Harbinger: Blow it.
Rabinowitz: It’s not our property.

–          Hot Shots Part Deux

President Obama has overperformed in many swing states compared to national polling against Mitt Romney. A number of these states were being called Obama’s Firewall because even though Romney was surging in the popular vote, he wouldn’t win the electoral college according to polls conducted in the swing states where Obama was still holding a lead. However, states poll less frequently than national polls so those polls lagged the national trends by several days. Once Romney began surging over Obama in those swing state polls, pundits began to doubt the strength of the Firewall.

Nate Silver called the Firewall brittle. Mike Murphy on Meet the Press called it a picket fence and inspired the quote above.

While the Firewall has been broken through in most swing states, I ultimately disagree with both of them. I think Obama’s firewall is substantial in Ohio. Take a look at Obama’s loss of support in swing states:

Let me explain the one column that’s seemingly out of place. Gubernatorial elections usually have the least correlation to the choices the voters make for their federal representatives in the House, Senate, and the Presidency. For example, Republicans won 4 consecutive gubernatorial elections (1990, 1994, 1998, and 2002) with 3 different candidates in Massachusetts!

But we can combine data analysis with intuition and experience. With President Obama not “let[ting] a crisis go to waste,” he signed the controversial $787 billion stimulus with almost no input or support from Republicans in March 2009 and plans were quickly hatched for a massive health care bill that has been a Democrat dream legislation for 70 years before finally passing it without any Republican support and through controversial means in March 2010. With the rise of unprecedented activism of conservatives through the Tea Party movement, the 2010 gubernatorial races were much more nationalized than in typical off-cycle years and just the fact that 2010 was an excellent Republican year affects races at all levels in that year.

This sets up the most salient point of the data above: Obama has lost massive support in the Midwest and in Pennsylvania, Ohio’s neighbor to the east that shares its concern for the coal mining industry that Obama has shown lukewarm support to. All of the states highlighted followed the trend of voting to change their governor from Democrat to Republican during Obama’s Presidency. Ohio is in the middle of these states, but Obama has significantly mitigated his loss off support there compared to surrounding states. How?

Most pundits’ explanation for the Ohio Firewall is the auto bailout supported by Obama that saved the largest number of auto industry jobs in Ohio compared to any other state except… Michigan! Even though Michigan retained the most auto jobs, the state still reduced its support for Obama by triple the amount that he has lost in Ohio. So how is Obama retaining enough support to still win the state if the election were held today?

Let me separate myself from the other pundits by saying, “I don’t know.” It certainly isn’t that Ohio polls are lagging the national polls as there has been a plethora of polling in that state. And it’s very surprising to see the state that voted 1.5% lower than Obama’s 2008 national vote percentage to suddenly be supporting him at 1.6% above his current national polling number. Note that this 3.1% swing in Obama’s favor is well outside the margin of error because RCP’s average might have a .5% margin of error or less since the sample is so high when they combine so many polls.

II.

A clearer way to show the Firewall is in the race to the 270 electoral votes needed to win the Presidency. These are the most updated polls that show what would happen if the election were held today.

Romney’s electoral college gains are read from top to bottom. In other words, he’ll win Indiana to attain 181 electoral votes “before” he wins Missouri when he’ll then have 191 electoral votes up until he likely wins Virginia to attain 257 electoral votes.

Obama’s electoral college gains are read in the opposite direction from the bottom to the top. He’ll win New Mexico to attain 201 electoral votes “before” he wins Nevada for his 207th electoral vote up until he likely wins Michigan for his 263rd electoral vote.

Let me explain why Ohio is the demarcation at 48.6%. The 2012 election is like the previous two presidential elections in 2004 and 2008 where there isn’t a credible 3rd party challenger this year that would take a significant percentage of the vote as there was in 1992 and 1996 (Ross Perot) and in 2000 (Ralph Nader). In the close election of 2004, only 1% of the vote went to a candidate outside of the two major political parties and even in the landslide election of 2008, which afforded voters an opportunity to voice a protest to the Two Party System, only 1.5% of the vote went to 3rd party candidates.

The narrowest victories by state in the 2004 and 2008 Presidential elections actually went to the loser. John Kerry managed to win Wisconsin by .38% with 49.70% of the vote. And the closest state in 2008 was John McCain’s victory in Missouri by .13% as he garnered 49.36% of the vote.

The expectation for 2012 should be the same as the prior two elections. That means a candidate must at least acquire about 49.3% of the vote to win the state. Ohio at 48.6% for the incumbent is close enough to give Obama the edge while the other states might be out of reach.

We only have one point of reference of a final RCP average of all polls for an incumbent President before the election and that is in 2004 when George W. Bush finished with a 48.9% average and received 50.7% of the actual vote. That’s a 1.8% increase, if that same amount is given to Obama’s RCP average in each state that would put Colorado and Virginia above the 49.3% needed to win both of them along with Ohio. I’m not giving him that much and that will be explained in the next post.

One last thing to note is this chart is indicating Romney to win the popular vote and just barely lose the electoral college vote as Obama wins Ohio by the slightest of margins.

Advertisements

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: