Mitt Romney won a decisive victory last night in Florida, carrying 46 percent of the vote. Let’s take a close look at how he did it.

First, let’s start by comparing John McCain’s hauls among key socioeconomic groups to Romney’s hauls this time around. Because Romney did about 10 points better than McCain did, we’ll see that the Massachusetts governor outperformed the Arizona senator among every group. Accordingly, the numbers in parentheses show the relative over/underperformance of each candidate among that particular group. So, for instance, John McCain won 39 percent of voters who made under $50,000 per year in 2008 and he won 36 percent of the total vote, so the value in parentheses is +3.

What we see here are two coalitions that differ in some ways, but not in others. In terms of age groups and ideology, they are very similar – McCain and Romney both sampled disproportionately from older voters and more moderate voters. However, McCain’s voting coalition was slightly more “downscale” in terms of socioeconomic status – less college educated and less wealthy.

What is worth pointing out as well – and it is not in the chart – is that Romney did better in 2008 among “very conservative” voters than he did this time around, despite the fact that he overall won a much smaller vote share in 2008. He won 44 percent of very conservative voters in 2008, compared to just 29 percent last night. This is a phenomenon that we generally see repeated among conservative opinion leaders, who were much more comfortable with Romney in 2008 than now. Much of this shift is presumably due to Romney’s connection to Obama’s health care plan.

But that might not be the entire story. Indeed, there appears to be a growing geographical divide that is being layered on top of this ideological divide. To appreciate this, let’s look at ideological support by state – in particular Romney’s margin of victory or defeat over Gingrich among ideological subgroups, and then among differing views of the Tea Party.

The results are intriguing.

As we can see, Romney and Gingrich basically split the “very conservative” vote in the Midwestern state of Iowa while Romney won a decisive victory among these voters in New Hampshire. The same goes with the Tea Party vote – a virtual tie between Romney and Gingrich in Iowa followed by a substantial Romney victory in the Granite State among Tea Partiers.

But when we turn to the South, we see a vastly different picture. Gingrich rolled over Romney among South Carolina’s very conservative voters and won a strong victory among Tea Partiers in the Palmetto State. In Florida, Gingrich won the very conservative vote and they split the Tea Party vote. What’s more, the vote tended to break down along regional lines in the Sunshine State. The panhandle of the state is the most culturally and historically Southern, with the peninsula being much more Northern in its outlook. Check out the wide divergence in terms of voting:

Taken together, this indicates a geographical split among conservatives and Tea Partiers. Those in the North and Midwest are more sympathetic to Romney, viewing him perhaps as one of their own. But when we turn Southward, the links between Romney and the right seems to be much more tenuous. What is so fascinating about this is that we’re talking about people in different states who answer the ideological question similarly. This is geography, not ideology.

This should not come as a huge surprise. Geographical issues have really been part and parcel of American electoral politics since the beginning of the 1800s –Thomas Jefferson and his Republican party basically continued the policies of Alexander Hamilton and the Federalists, but because the former were from the South, the outrage in Dixie was virtually nonexistent. Something like that might be going on here – with Southerners simply being more comfortable with the Speaker of the House from Georgia rather than the governor from Massachusetts, even though both claim to be conservative and have very little space between them in terms of policy proposals. The opposite looks to be holding true in the North.

This is still a tentative conclusion, but it would seem to be terrible news for Gingrich. If he cannot find a way to appeal to Northern conservatives, he is going to lose the nomination by a very large margin, for two reasons. First, he just got walloped in Florida, a winner-take-all delegate state, and he is going to collect exactly zero delegates from Virginia because his name is not on the ballot. It is an enormous challenge to be the Southern candidate who does not win any delegates from two of the largest Southern states.

Second, the electoral and legislative power of the South has long depended upon an alliance with the West. In fact, the electoral strategy of nearly every major Southern candidate for national office dating back to Thomas Jefferson has been to unite these two regions. Gingrich is going to have a huge problem recreating this because many Western GOP electorates are relatively moderate now – e.g. California and Oregon – or have very strong Mormon blocs. Romney finished a distant third to John McCain in 2008, but check out his strong performance in the Mountain West (and note that the contests in Idaho and New Mexico happened after he dropped out of the race). That's the Mormon vote in action.

A final note: Gingrich looks to have been damaged in Florida after he fell off the strategy that initially brought him to the front of the pack – which was his relentless focus on the issues. His engagement with Romney in terms of negative campaigning seems to have taken a toll on his numbers, at least in Florida.

Bottom line: Nobody wins a party nomination when his unfavorable rating is 40 percent, or when he gets blown out of the water among women. If this is the start of a trend, Gingrich is finished.

Next Page