Today we continue our post-election overview by looking at the West. Historically, the West has been a fairly volatile region. In the 1880s the Republican Party figured that the West would be a GOP bastion, and accordingly the 51st Congress (1889-90) added four western states to the Union (plus North and South Dakota). However, the GOP was in for a surprise, as the West tilted Populist in 1892, then went Democratic in 1896. From that point until about 1960, a victorious Democratic coalition always depended upon an alliance of the South and the West.

Today, the West is an integral part of both party coalitions, but it is now an almost inverse of what it was 50 years ago, when the Republicans dominated the Pacific Coast and the Democrats ran strongly in the sparsely populated interior. In the intervening years, the coast shifted toward the left, and thus to the Democrats. Meanwhile, air conditioning helped grow the suburbs, and thus GOP strongholds, in the Mountain West -- and the urbanization of the Democratic coalition pushed rural, Bryanesque progressives into the Republican camp.

Unsurprisingly, the West tipped Democratic in 2006 and 2008, so that by the time Barack Obama was inaugurated as the 44th President, this was the breakdown in Western House districts.

But two weeks ago, the GOP rebounded.

What's interesting here is how similar these results are to the Northeast. In all of the Pacific West, the GOP netted precisely zero congressional districts (if you count the loss of Hawaii's 1st Congressional District against the gain of Washington's 3rd). This is just like New England. And also like New England, there was indeed a shift in the congressional vote in the Pacific West. It just wasn't big enough.

The bottom line is that the three major states of the Pacific West are just so strongly Democratic that even when a Democratic president is pulling in less than 50 percent support nationwide, the party can still hold the line.

But on the other hand, the GOP did have success in the Mountain West, at least on the House level. In 2004, the last time the Republicans controlled a House majority, 20 Republicans came from the Mountain West. In the upcoming Congress, 18 will hail from this region, and a shift of just 5,000 votes in southern Arizona would have brought the GOP fully back to where it was six years ago. In this way, it appears similar to the Mid-Atlantic, where the GOP victories earlier in this month resulted in a return to the pre-2006 party strength.

This snap back in the Mountain West is significant, but it has been somewhat overlooked -- in part because of terrible Republican under-performances in Colorado and Nevada, where the GOP lost two Senate races and one gubernatorial race it really had no business losing.

While there is no doubt that Democratic strength has increased in the Mountain West in the last decade (or, more specifically, in Colorado and Nevada, as the Democrats have typically been strong in New Mexico and the GOP still dominates the other states), the GOP losses in Colorado and Nevada amounted to unforced errors. It's like a football game where one team is slightly stronger than the other, but has a net turnover rate of -3. That's a deficit that is just too much to come back from.

In both Colorado and Nevada, GOP House candidates won the statewide popular vote; in both states a majority of voters disapproved of President Obama's performance in office; and in Nevada, the Republican gubernatorial candidate, Brian Sandoval, crushed a candidate whose last name was the same as the Senate candidate -- Reid. That last point is about as close to a "smoking gun" as we ever can get in this kind of analysis, and it demonstrates that Sharron Angle was a very weak candidate, indeed. There's no other way to explain why so many voted to endorse the father but reject the son.

A point I have been making through these write-ups is that the GOP needs to run "broad" candidates. I want to expand on this point by putting together a simple "breadth of appeal" metric. Suppose that a Republican candidate won 85 percent of Obama disapprovers while losing 80 percent of Obama approvers. We can get a pretty clear idea of how broad the candidate's appeal was by subtracting the percentage of Obama approvers who backed the Democrat from the percentage of Obama disapprovers who backed the Republican. In this case, it'd be 85 percent - 80 percent = 5 percent, which we'll label the "Breadth of Appeal." So, the higher the number, the more broad the candidate's appeal. The advantage of this metric is that it enables us to compare many candidates at once without getting bogged down in a sea of statistics. Here's a look at notable Republican candidates in the West.

Check out Angle's number. That's terrible. As a point of comparison, Christine O'Donnell's "breadth of appeal" number came in at -9. I haven't done a systematic search just yet, but I would not be at all surprised if Sharron Angle turns out to be the most narrow of all statewide Republican candidates who had enough money to fund a full campaign. Notice also how narrow Buck's appeal was. That's how two easy wins turned into losses.

While the share of the minority vote is growing in both Colorado and Nevada, Angle and Buck lost because of white voters, specifically white women. Buck, in particular, underperformed dramatically among this group. Nationwide, House Republicans won white women by 19 points, while Buck lost them by 7 points.

Among non-white voters, Republicans held their own. This group is a key element of the "emerging Democratic majority" thesis, but one problem is that advocates of this thesis make a kind of category error: you can't really lump African-American voters, who are monolithically in favor of Democrats, with Latinos or Asian Americans, whose support for Republicans is always greater and is highly dependent upon other factors like region and candidate quality. This chart makes that point pretty well.

As you can see, the Latino vote in Florida and even Texas is very attainable for Republicans, while it is much more strongly Democratic in California. Unfortunately, no exit poll was taken in New Mexico, but Republican Susana Martinez would not have been able to win 54 percent in the gubernatorial race without strong backing from the Latino vote, which amounted to 31 percent of all New Mexico voters in the 2006 midterm.

Politicians have been aware of the monolithic quality of the African-American vote for nearly a century. It was a chief reason why the two parties moved so quickly to adopt liberal civil rights planks after World War II: both sides knew that the African-American vote was up for grabs, and it would probably be all or none. But the Latino vote is different. It remains in play, and Republicans can usually pull at least 25 percent support from this group. If anything, the Latino vote reminds me of the white Catholic vote prior to 1968. It was typically Democratic, but not monolithically so, and the GOP enjoyed significant support from this group in the 1940s and 1950s, in large part thanks to the emergence of Communism as a key political issue. That's the better comparison to Latinos, I think, than the African-American vote, and there is a lesson there for Republicans. The right Republican candidates with the right mix of issue positions/emphases should help the GOP hold its own with this growing bloc of voters. That may not be enough in California, but it should enable the GOP to keep the Mountain West states competitive, if not tilted toward the Republicans.

This, in turn, suggests that Republicans must tread carefully on the issue of immigration. Border security is important, but there is a fine political line on this issue. Democratic strategists would love to tar all Republican candidates who favor securing the border as anti-Latino, and the GOP must be prepared to combat this image. Thus, candidates like Tom Tancredo must be avoided assiduously, and the proper political tone must be struck. To that end, the 2012 Republican nominee for president would be very smart to craft a position on immigration that promotes border security while also taking Latino sensitivities into account. And he or she should seriously consider Martinez, Rubio, and Sandoval as potential vice presidential picks. This bloc of voters remains up for grabs, and there is no good reason why the GOP should cede it to the Democrats without a knock down, drag out fight.

Next Page