Sunday, November 11, 2012

Sorry Ohio ... Colorado Was the Key State in 2012

It turns out Colorado was the tipping point state, not Ohio, in 2012. What do I mean by this?

I took all the states and listed them in the chart below in descending order by the size of Obama's margin of victory (or loss).

Of course, most of the states on the edges are irrelevant to this story so let's drill down to look at the states that Obama won or lost by 10 point or less.

It may not look like it but there is so much worthy of discussion here. First, notice that Obama did better nationally than he did in Ohio. This was not true in much of the pre-election polling and represented a moderate surprise to me. The final pre-election polling average on Pollster had Obama up by 1.5% nationally but up by 3.4% in Ohio. PPP's final national poll had Obama up by 2 but up by 5 in Ohio. There was much discussion in the final days of the campaign that the Romney campaign did not see an easy path in Ohio and that explained the last-minute (and futile) effort in Pennsylvania. But Ohio really was reasonably close.

Don't get me wrong here. That 1.9% victory is not tiny. Remember that Ohio was so saturated with campaign visits and campaign ads that it was probably going to be very, very difficult for Romney to move enough voters to win there. But it was a little bit closer than we thought it would be.

Second, notice that the three closest states Obama won were, as Chuck Todd likes to call them, "FLOHVA." And they were in that order - Florida, then Ohio, then Virginia. Chuck Todd talked a lot about how the election was going to come down to these three important states. I certainly agreed but figured that Florida was kind of irrelevant because, if Obama won Florida, he would already have won Ohio and would already have the 270 he needed. I was right about that latter part but was wrong that ... Obama didn't turn out to need Ohio either. Or Virginia. Wow.

And this brings me to, what is to me, the most interesting point. Colorado was the key "tipping point" state (Nate Silver's terminology), not Ohio. In arguing that Romney was wasting his time in Pennsylvania, I pointed out that the leads Obama had in Ohio and Virginia were so critical that
he can lose Florida, Pennsylvania, and hell, let's give Romney Colorado too. It is still an Obama victory.
But, of course, Obama was never going to lose Pennsylvania. So, when you give Pennsylvania to Obama, it turned out that Obama didn't need Ohio or Florida or Virginia.

Obama finished with 332 electoral votes. But some of these states were close. If Obama lost Florida (29 electoral votes) and Ohio (18 electoral votes) and Virginia (13 electoral votes) - a complete sweep of FLOHVA - he would still have had 272 electoral votes ... because of Colorado. Colorado is the state that won it for Obama.

And here's the interesting thing about that. Obama won Colorado by 4.7 points. Imagine for a moment that the states all shift equally as the national margin moves. This is not quite true but is not a crazy approximation of reality. If this were true, you could move the national vote 4.6 points in Romney's direction and Romney would still lose. You would be moving from a +2.7 Obama margin to a 1.9-point margin for Romney ... and Romney would still lose.

For most of the 2012 campaign, Colorado was thought of as a sort of safety-valve swing state. It was not given as much attention because the Obama campaign knew they win if they win Ohio. And, while that was true, it was also true the Obama campaign was winning without Ohio. Colorado was the true tipping point state.

When you start to meditate on that reality and then think about what drove Obama to victory in Colorado, you start to realize the trouble the Republican Party is in at the national level.

No comments: