Tag Archives: lizza

Kvetching About Etch-A-Sketching: What Really Determines Who Will Win This Race

The fallout from Romney spokesman’s Eric Fehrnstrom’s etch-a-sketch comments has dominated media coverage of the Republican nomination fight the last 48 hours, leading pundits to proclaim that Romney’s team has once again stepped on his own campaign message. Rather than building on any momentum generated by his solid victory in the Illinois primary last Tuesday, Team Romney instead has spent the last two days putting out the brush fires ignited by Fehrnstrom’s remarks. For those of you who missed it, here’s what Fehrnstrom actually said:

For many pundits, Fehrnstrom’s comments are damaging because they refocus media attention on the long-standing skepticism among many Republican voters that Romney is a true-blue conservative. Rather than any fixed set of conservative principles, he’s willing to say or do anything depending on his audience – a theme that found its way into the seemingly endless parade of etch-sketch parodies that dominated the “internets” in the wake of Fehrnstrom’s comments.  Many of those parodies – like this one – were created by groups working for Romney’s opponents.

And both Santorum and Gingrich lost no time in holding press conferences mocking Fehrnstrom’s comments while carrying their own etch-a-sketch devices.  The punditocracy, meanwhile,  used the etch-a-sketch comment as a reason to replay the long line of celebrated Romney’s “gaffes” that have figured so prominently in media coverage of his campaign.

But while the “massive campaign blunder” certainly cost Team Romney a couple of media cycles, it will likely have no long-term impact on the race for the Republican nomination. This is because at this stage the race is all about demographic-based voting blocs rather than media-generated “momentum” – or lack thereof. As many analysts, including myself, have noted before, Romney’s support is centered on higher-income, better educated, more moderate Republican voters.  It drops off, however, as one goes down the income and educational ladder.  In this respect, the Republican race bears a resemblance to the latter stages of the 2008 Democratic nomination fight, when Hillary Clinton and Barack Obama slogged it out through dozens of primaries and caucuses, with each drawing on an increasingly  identifiable set of voting blocs.

As I noted in my post last Tuesday, and as several other analysts have pointed out, a key indicator of how Romney will do in any primary is the proportion of evangelicals who vote. The Washington Post’s Jon Cohen used exit poll data to create a series of tables showing how the Republican candidates did in each state among different classes of voters.  Not surprising, the table based on the evangelical vote proves most revealing; Romney loses every state where evangelical turnout breaks the 50% mark.

Beating me to the punch, the New Yorker’s Ryan Lizza has extrapolated from that insight to make a rough forecast of the 23 remaining nomination contests. Of course, Lizza can’t know what the evangelical turnout will be in the remaining states, but by comparing the 2008 exit poll data with state-level surveys of the current evangelical population, he can make an educated guess.  This table lists the remaining states in descending order of evangelical population.

The key point here is that actual turnout among white evangelicals in Republican primaries is typically higher than the proportion of evangelicals in the overall state population, something Lizza checks by looking at the 2008 exit polls. Thus, although evangelicals are only 34% of the adult population in Texas, they constituted 60% of the Republican primary vote in 2008.  Using a 30% threshold of evangelicals as the dividing line, then, and giving Santorum the two remaining caucus states (Nebraska and Montana), Lizza estimates that Romney and Santorum will divide the remaining 22 state contests evenly, at 11 each.  In the tiebreaker, Romney is forecast to win the District of Columbia. Here are the states as allocated by Lizza:

Romney: Rhode Island, Washington, D.C., Utah, Connecticut, New York, New Jersey, Maryland, Delaware, California, Wisconsin, New Mexico, South Dakota.

Santorum: Arkansas, Kentucky, North Carolina, West Virginia, Indiana, Texas, Louisiana, Pennsylvania, Nebraska, Montana, Oregon.

Note, of course, that this is only Lizza’s forecast of state winners – not of total delegates, which of course is what is really at stake in the remaining contests.  However, even if we accept Lizza’s forecast, estimating how many delegates each candidate will win is more complicated due to the different state delegate allocation rules and because, as I noted in my previous post, that it matters whether Gingrich stays in the race.  If I get some time free from grading, I’ll try to produce a very rough estimate of how the delegate race may play out.   In the meantime, pay no attention to media kvetching about etch-a-sketching or, for that matter, how Jeb Bush’s endorsement may finally indicate – finally, I tell ya! – that the Republican establishment is ready to fall in line behind Romney.  Beyond the immediate media entertainment value, these aren’t going to influence how this race plays out.  Instead, a better indicator are the increasingly identifiable demographic-based voting blocs.  It’s not the etch-a-sketch Romney should worry about – it’s the evangelicals.