If a caucus falls in the forest, and nobody hears it, does it make a sound?
Yesterday there was a Republican caucus in Wyoming. It was won my Mitt Romney and his boy band.
You probably didn't hear anything about it, though.
News of the Wyoming caucus was relegated to a single, below-the-fold link at the New York Times politics homepage. Mark Halperin, who's website is usually a pretty good proxy for media buzz, no longer has a mention of it at The Page. Jonathan Martin blogged about it just once at Politico, with the appropriate headline "Don't Forget Wyoming!".
No public pollster took a poll in Wyoming. Not one. No candidate had a field office in Wyoming.
The Republicans made a grand total of 8 appearances in Wyoming. Three of those appearances were by Sam Brownback, who has dropped out of the race. Two were by Duncan Hunter. The other three were by Romney, the only first-tier candidate to visit the state. Even Ron Paul, who gets a lot of campaign dollars from the Mountain West and could have picked off some cheap delegates, didn't bother to visit.
And there sure aren't very many delegates up for grabs in Wyoming.
All in all, the Wyoming caucus had about as much impact as a favorable Newsweek cover or an endorsement from a junior senator. It's a favorable story that lasted for a about half a media cycle. Of course, maybe we'll be surprised, Mitt Romney will win New Hampshire by 10 points on Tuesday, and we'll all be talking about the "Wyoming bounce". But I suspect that we won't.
This ought to have some implications for the next four Democratic electoral events after New Hampshire, those in Michigan, Nevada, South Carolina and Florida, respectively. The value of primaries and caucuses is largely symbolic at this stage. None of these states, including Iowa and New Hampshire, are worth very many delegates (Michigan and Florida would be, of course, if the DNC hadn't stripped those delegates to punish the states for jumping ahead in line).
At this stage, rather, the impact of a state is proportional to how seriously the candidates and the media take it. With that in mind, let's take a look at a couple of metrics that will help us to rate the importance of the next four states on the Democratic docket. The first two metrics concern how seriously the candidates are taking the states; the latter two concern how seriously the media is taking the states.
Campaign Appearances
The New York Times has a very cool applet on campaign events and appearances. I have translated this information into graph form, based on the number of appearances made by the three major Democratic candidates since Labor Day.
Somewhat to my surprise, Nevada has actually gotten more traction than South Carolina. It has received a total of 28 appearances to South Carolina's 21. There has been almost on-the-ground activity in Michigan or Florida at all. Hillary Clinton and Barack Obama had two events in Florida each; the lone post-Labor Day appearance in Michigan also belongs to Clinton. Obama is the only major candidate to visit either state since October 1st (he made a two-day swing through Florida in early November).
Field Offices
This information is gathered from the respective candidates' websites and hopefully is as accurate as it can be.
Three fairly straightforward things to point out here. First, no candidate has a field office in Florida or Michigan. Secondly, Nevada once again outranks South Carolina. Third, we see a little bit of evidence of the resource advantages that the Clinton and Obama campaigns have relative to Edwards.
It's also interesting that Obama has more offices open in Nevada than South Carolina, since the latter should be a much better state for him, while just the opposite is true for Clinton. Frankly, this is now looking like a pretty good gambit for Obama. If he wins New Hampshire, it is hard to imagine him losing South Carolina, which inherently has to be one of the best states on the map for him because of its heavy African-American population (which was trending heavily toward Obama even before Iowa). On the other hand, he has never been especially strong in Nevada, which has polled more like a February 5th state. If he can beat Clinton in Nevada, that would really speak to his momentum and could mean the beginning of the end for her campaign. But, if Obama loses New Hampshire, he would probably have to write off Nevada, and hope to make a comeback in South Carolina. This suggests that the Obama campaign was quite confident that it would win New Hampshire -- if it won Iowa.
Media Hits
Now we'll move to the two media-based measures. The first is a survey of current "hits" on Google News, which reflect any articles published in the last month. The syntax I used for these searches is:
"florida" democratic primary -republican -court
Note that I exclude any articles where the word "Republican" is mentioned, since I want to avoid instances where the Democratic race is mentioned in passing (e.g. "There is a surge of support for John McCain in the Michigan Republican primary; there is also a Democratic primary that day, but nobody cares about it). I also exclude any article where the word "court" is mentioned, since a fair number of the Florida and Michigan articles discuss court challenges to the primary date. Finally, for Nevada only, I permitted the word "caucus" as well as "primary", since Nevada's election is a caucus, but still often referred to in the context of the primaries.
We see that South Carolina dwarfs the other states in terms of media buzz. It's gotten about 3x as much media attention as Nevada, 5x as much as Florida, and 6x as much as Michigan. The gap in media attention between South Carolina and Nevada is interesting, given how the campaigns are treating them. On the other hand, the lack of attention to Michigan and Florida is not surprising, and in fact this metric is probably overstating their importance because it is not possible to remove all out-of-context mentions.
Public Polls
The other metric for media attention is the number of polls taken since Labor Day, as according to Real Clear Politics. It costs a fair amount of money to do a poll, and so the reward for a pollster is the free advertising it gets as a result. Pollsters, therefore, have to anticipate how much media attention their polls will get, and may be a good leading indicator of sorts.
There has been almost no recent polling activity in Michigan, which makes sense since Obama, Edwards and Richardson have all pulled their names from the ballot (nobody has yet had the sense of humor to poll Clinton against Kucinich and uncommitted). On the other hand, there have been as many Florida polls as South Carolina polls. This may be because the pollsters are already in the field in Florida, which is a very important contest for the Republicans, and it's comparatively easy to do a poll for the Democrats at the same time you're doing one for Republicans. Still, the pollsters are anticipating at least some demand for their Florida polls, which is not true for Michigan. Meanwhile, the number of polls in Nevada may be deflated because it is a caucus state.
Putting it Together
We can build an index of the four metrics by assigning each state a share out of 100 points. For example, 59% of the media hits have been in South Carolina, so South Carolina gets a 59 in this category. We can add these scores up to create what I'll call the "Total Importance Index". The totals are:
South Carolina 184
Nevada 141
Florida 60
Michigan 15
So, we see that South Carolina is a little bit more important than Nevada, but the difference is perhaps less than you'd gather from the media's treatment of it; the campaigns are certainly investing in Nevada. Michigan is completely, and I mean, completely irrelevant; it's a bullet point in a Mark Penn memo.
Florida will be the interesting one. It is certainly not on the same scale as South Carolina or Nevada. In fact, only the pollsters are treating it with any importance at all so far. But the way that the contest is playing out, I suspect that Florida may begin to generate more buzz, in large part because South Carolina looks like it will be an anti-climax. But there are interesting tactical decisions to be made by both the Clinton and Obama campaigns.
Florida should be a pretty good state for Clinton. It is a closed primary state, has a large elderly population, and a lot of emigrants from New York. About the only advantage for Obama is that the state has a fairly large African American population (16%). (It also has a large Hispanic population, of course, but I've seen very little discussion of where Hispanic votes are going). On the other hand, there are some perceptual risks for Clinton. All the candidates have pledged not to campaign in Florida (as well as Michigan), and so if she has explicit campaign events in that state (instead of surreptitious ones like fundraisers), she may look like she is putting her own best interests before those of the party -- a theme that is liable to grow in importance if Obama wins big in New Hampshire, etc. Also, it may simply make her look a little desperate. And naturally, any resources spent in Florida are resources that cannot be spent elsewhere.
Obama has an easy decision to make if Clinton sits out Florida. The lesson of Wyoming is that if the candidates do not build a campaign, the media will not come, and Obama has no incentive to build up the importance of a state where he could stumble. On the other hand, if Clinton does begin to campaign in Florida, Obama's decision is more difficult. If he matches Clinton's campaigning in the state, he certainly improves his chances of winning it; on the other hand, he also increases the cost of not winning it, since the media will treat it with more importance. He is going to need to do some internal polling of that state and think it over carefully with his media team.
n.b. Implicit hat tip to JedReport