There is a very dangerous and stupid analysis floating around the Republican blogosphere that goes something like this: John McCain got 2 million more votes that Mitt Romney. How can this be? Either Romney was a terrible candidate – unable to win even all of the McCain voters from 2008, or evangelical Christians refused to vote for a Mormon, or we just experienced voter fraud on a scale never before imagined.
Stop it for two reasons. One, Republicans are in danger of blaming everything but the real reason they lost the election. Two, they’re at the risk of sounding nutty.
Mitt Romney was an excellent candidate. His defeat was due to exactly two things: one, an effective Barack Obama ad campaign that successfully painted Romney as a heartless robber baron who didn’t respect women (hurting him in Ohio and Pennsylvania), and two, the changing social mores of a society that is simply no longer socially conservative in the main (hurting him in Florida, Virginia, and New Hampshire).
In a previous blog post, I took on the McCain beats Romney theory and pointed out that while turnout was down nationwide by about 8 percent, turnout in most of the battleground states was virtually unchanged.
Since the McCain beats Romney theory is still very much alive, and because my previous analysis was based on incomplete election returns, I thought it was time to update my turnout map. Incidentally, the more-complete (but still not complete) election data indicates the turnout drop from 2008 nationwide is about 6.75% (about 9 million votes). Of that total, President Obama got 7.4 million fewer votes than in 2008. It is true that McCain got more votes than Romney – but the difference is now just 1.2 million votes (and falling).
Where was turnout depressed? Here’s the update (with one major correction – I discovered an error on Wikipedia’s 2008 Colorado totals, which were higher than in my initial data set):
The updated figures only reinforce my previous conclusion: while you might blame the Romney campaign for failing to excite voters in Ohio, Pennsylvania, Michigan, and New Hampshire, you cannot say the same thing for the battleground states of Minnesota, Wisconsin, Nevada, Virginia, North Carolina, and Florida, where turnouts were higher than 2008. Something is going on in the Midwest “rust belt” states, clearly, though note there is no perfect relationship between turnout and the winner. Ohio, Michigan, and Pennsylvania were Romney disappointments, but West Virginia and Indiana went for Romney in a big way, with Romney’s lead over President Obama about 10 percentage points greater in both states than the differential in the McCain-Obama contest.
More to the point, though, while Mitt Romney did earn less votes than John McCain, almost all of those votes were in two states: California and New York! By my current accounting – the final numbers will no doubt be a little different – John McCain earned 1,171,230 votes more than Mitt Romney. But check this out: McCain earned 1,027,896 more votes than Mitt Romney in California alone. He earned 526,091 more votes in New York. That’s the entire magnitude of difference, and then some. Why “and then some”? Because Mitt Romney earned more votes than John McCain in 30 of 50 states. Many of those states were election-critical states, and in 7 of the states that he didn’t earn more votes than McCain, he beat President Obama anyway.
Here’s a graphical representation of the vacuousness of the McCain beats Romney thesis:
McCain blasted Romney in McCain’s home state of Arizona (beating Romney by 158,554 votes). Being on the West Coast may have also suppressed turnout generally – President Obama was also well short of his 2008 total there (165,352 fewer votes). That’s right, even where Mitt Romney “lost” 150,000 votes to John McCain, he beat his opponent by 1.8 more percentage points. Ditto California, where Romney’s loss to President Obama was 3.1 percentage points fewer than McCain’s loss to Candidate Obama. In New York, Romney’s loss was also slightly narrower than McCain’s loss there.
In fact, if I were to make the equivalent graphic for President Obama – comparing his 2012 vote totals to his 2008 vote totals – only two states would be colored green. One of those is a battleground state – North Carolina, though it is one he lost because Romney did much better than McCain did there (gaining 147,379 votes). The other is Louisiana. In every other state, the President received fewer votes (pending the final vote totals, of course.)
A similar story would be told for the red and orange states above. Romney has two states in red (at least half a million votes fewer than 2008’s total for McCain) and two in orange (100,000 – 500,000 fewer). The President would have three states in the red (California, New York, and Illinois) and twelve states in orange (Pennsylvania, Michigan, New Jersey, Ohio, Texas, Indiana, Missouri, Washington, Arizona, Tennessee, Oregon, and Maryland). Many of those are safe states for one side or the other, but Pennsylvania, Michigan, and Ohio – where the President lost 358,000 votes or more in each – Indiana, Missouri, and Arizona stand out as key states going forward.
In terms of margin of victory, the President increased his advantage in 4 states: Alaska (where his loss was 8.1 percentage points narrower), New Jersey (increasing his win 1.6 percentage points), Louisiana (1.3 point narrower defeat), and Mississippi (1.1 points narrower). He didn’t lose any ground in New York. But in 45 states, Mitt Romney’s win over President Obama was greater or his loss narrower than the equivalent differential between John McCain and President Obama.
With all of this information, though, Mitt Romney’s losses in Virginia, Florida, Wisconsin, Colorado, Iowa, and Nevada just seem all the more perplexing. In those states, all of the following was true:
- Turnout was high
- Mitt Romney did considerably better than John McCain
- The President did poorer in 2012 than he had done in 2008
The analysis is a little more complex in states like New Hampshire, Ohio, and Pennsylvania, but even in those states, at least some of the above factors were present.
Definitively, though, we can stop this nonsense that Mitt Romney somehow lost the John McCain voters in any way that could have reasonably explained his failure to defeat the incumbent. There’s simply no evidence for this idea. Worse, it will distract Republicans from the true lessons of the election. Indeed, one could even find these data quite encouraging – if there hadn’t been so many forecasts of a Republican win (only among Republicans, by the way, including me) – the election could rightly be seen as an electorate moving back to the right from as far left as it had swung in 2008.
Crucially, Republicans have made gains in the Midwest. If it is true that Republicans are sliding among the Latino population – and I believe this is true, and should be rectified if at all possible – then Republicans would do well to become competitive in states where the Latino population isn’t increasing as rapidly. Texas will be safe for some time, and California will be out of the realm of possibility for some time – but the upper Midwest and the rust belt, where the Latino population increase is small – may be coming back into play for Republicans. Of course, Republicans should better-appeal to Latinos, and one place to start might be Florida, where this election gave some encouraging signs and where Latinos have been most-friendly to Republicans thanks to the Cuban population and the recent Republican Senators of Latino heritage (Marco Rubio and Mel Martinez, not to mention Governor Bob Martinez).