Today's post is the first entry in a four-part series analyzing the 2010 midterm election. My plan is to break it down by region, and I begin today with the South – or more specifically, the 11 states that made up the old Confederacy.

To start, let’s sketch out a broad overview of the Southern states, based on the available exit polling data. Exit polls were conducted in 5 of the 11 states we are evaluating, and here are some notable results.

Using Obama’s 2008 statewide vote as a baseline, we can see that there have been relatively few shifts in opinion on the president in many of the Southern states. Only two stand out – Florida and Texas – and much of these shifts might be due to the fact that the percentages of white voters in the 2010 electorate were higher than the percentages in 2008.

Let's juxtapose this overall stasis with the change we saw last week in Southern House districts, which was quite dramatic. This is the map of Southern congressional districts after the 2008 election.

Now, here is the map after last week.

This is quite a change, a historic one in fact. The Republicans control 94 of the 131 congressional districts in the South, or about 71 percent of all districts. The GOP has not held such a large share of Southern districts since 1868, during Reconstruction.

How do we square only modest movements in public opinion of the president with the big shifts in the congressional map? It has something to do with the point I made in the latest issue of THE WEEKLY STANDARD: The 2010 midterm represents the re-emergence of the Bush coalition, at least on a congressional level. Throughout most of the South, the Bush coalition was largely in place to vote for McCain in 2008. The trick was that many districts that backed Bush and McCain for president split their ballots to elect Democrats to Congress. They stopped doing that this time. Sean Trende made a similar point in an insightful piece last week, writing:

Much as 2008 was the year that Democratic-leaning districts determined that they could no longer afford to send Republicans to Congress, no matter how moderate, 2010 is the year that the Republican-leaning districts made the inverse decision. It didn't matter whether the challenger had experience, or if he raised a lot of money. Neither of those variables is statistically significant.

To frame it another way, recall retiring Democrat Marion Berry’s account of a conversation with President Obama in early 2010:

Berry recounted meetings with White House officials, reminiscent of some during the Clinton days, where he and others urged them not to force Blue Dogs “off into that swamp” of supporting bills that would be unpopular with voters back home.

“I’ve been doing that with this White House, and they just don’t seem to give it any credibility at all,” Berry said. “They just kept telling us how good it was going to be. The president himself, when that was brought up in one group, said, ‘Well, the big difference here and in ’94 was you’ve got me.’ We’re going to see how much difference that makes now.”

Well, I guess now we know: It made one hell of a difference!

A final note on the congressional districts is that this does not represent the theoretical ceiling for the GOP. In particular, the gerrymandered congressional districts in North Carolina continues to favor the Democrats heavily. The party managed to win 7 of 13 House districts even as it lost the statewide House popular vote, but this will soon change, as the GOP now controls the North Carolina legislature and will redraw the district lines next year.

That’s the broad, region-wide takeaway I would derive from the midterm. Now, let’s look more closely at Florida, a state that had a tight, highly competitive gubernatorial contest last week and that will surely be a key presidential battleground in 2012. Republican Rick Scott won a very narrow, 1.2 percent victory over Democrat Alex Sink last Tuesday. The exit polls suggest that his victory was due not to a shift in the white vote, but the minority vote.

As you can see, the white vote did not move statewide. Instead, Hispanics and African Americans inched ever so slightly to Scott, which contributed to the victory. What also helped was the fact that the Florida electorate (as noted in the chart above) was whiter in 2010 than in 2008. Without that shift, Scott would have lost.

This is good news for Democrats, as it suggests that Florida will be very competitive in 2012. The lesson for Republicans echoes a point I made last week: The Republican Party needs to run a strong candidate in 2012. In fact, the exit polls suggest that Scott was a relatively weak candidate. He had a lot of baggage, and the data indicates that it had a negative effect. Consider the following.

What I’m doing here is using the national House vote as a baseline to evaluate Scott's appeal. Nationwide, 86 percent of all voters who disapproved of Obama voted for a Republican House candidate, while 13 percent of Obama approvers voted for a Republican. Scott underperformed both those numbers. He pulled in just 81 percent of Obama disapprovers and 9 percent of Obama approvers.

Where might those losses have come from? My hunch tells me it was among white voters who fall on the higher end of the socioeconomic scale. Consider this chart.

What we see is that while the white vote between 2008 and 2010 was constant for Republicans, there was movement within white subgroups. Scott did a little better than McCain among whites without college degrees and whites who make less than $50,000. Yet he did a little worse among whites with college degrees and whites who make more than $50,000.

Generally, as we run through these evaluations, we are going to see this time and time again. Good Republican candidates got better results on Election Day than bad ones. That’s the big lesson from 2010 for the GOP: Candidate quality matters a great deal. Scott was a relatively weak candidate, and it showed in the results. A stronger Republican – say, Republican Senate candidate Marco Rubio in a two-person race – probably would have done much better than Rick Scott. Indeed, Rubio actually got more votes than Scott even though he had to deal with Kendrick Meek and Charlie Crist. Rubio's success indicates that ideology is not a critical factor in evaluating candidate quality. He was a very conservative candidate, and still did very well, considering the three-way nature of the race. (And as we move forward, it's important not to mistake my emphasis on candidate quality as a subtle plea for Dewey-style moderation. It most certainly is not. In the last 114 years, the most successful Republican candidates have often been conservative.)

The next entry will be on the Northeast, and it will come out on Friday morning. Check back then!

Note: The above map of the 2010 midterm election results incorrectly states that Louisiana's Fifth Congressional District is held by the Democrats. In fact, Republican Rodney Alexander has been the representative from this district since 2002.

Next Page