Larry Sabato has an article out today emphasizing uncertainty in the early states, noting that “From 1976 to 2008, there has been a major surprise every time either in Iowa or New Hampshire.” That’s true, as far as it goes: he’s right that no one expected Pat Robertson to surge to second place in Iowa in 1988 ahead of a sitting Vice President, or John McCain to upset George W. Bush in New Hampshire in 2000.
The truth is, however, that what those examples reveal isn’t that Iowa and New Hampshire can be volatile (although they are), but that surprises there often don’t matter very much, at least not since the modern system was set firmly in place in the 1980s. After all, George H.W. Bush in 1988 and George W. Bush in 2000 wound up cruising to their nominations, despite the hiccups along the way. In both cases, party leaders had pretty much settled on a winner, and looking back the day-to-day excitement of the primaries just wasn’t where the real action in those cycles was.
On the other hand, in contests in which party actors are unable to reach consensus on a nominee – most notably, in the last two Democratic nomination cycles – then the primaries and caucuses do have an important part to play. In 2004, Howard Dean’s miserable showing in the Iowa caucuses appeared (in my view) to have convinced party actors that Howard Dean’s popularity was hype, not reality, something that many of them were not at all certain of in fall 2003. And in 2008, Iowa had the opposite effect, confirming that Barack Obama would in fact appeal to rank-and-file voters – and setting up, as we all remember, a long contest with party actors and voters split, leaving the long sequence of primaries and caucuses to determine the nominee.
The problem for observers trying to figure out what will happen this time is that the “invisible primary” is, in fact, not very visible; it’s not always easy to figure out whether there is a winner or not. After all, we’re not just talking about a handful of party leaders in Washington; there are hundreds of party actors all around the nation, and it’s not always easy at all to know which are the most important ones or what they really think. The most tangible indications are high-profile endorsements and fundraising. Based on those it appears so far that Mitt Romney has a lead, but not nearly as solid as, say, George W. Bush in 2000. So we’ll see: Romney could add to that lead in the next few weeks, making Iowa and New Hampshire much less important, or party actors could stay on the fence, still hoping that they can support someone else. If that’s the case, the early states could indeed by critical.
But most Iowa and New Hampshire surprises have nothing to do with who eventually wins the nomination. They're great for political junkies (myself definitely included), and they're certain to be pounced on by the press, which has a strong interest in portraying nomination battles as close fights for as long as it can. They can change the structure of the losers, keeping a Huckabee alive longer than otherwise would have been the case, or killing off a Glenn or Graham or Connally earlier than might have been. One Iowa surprise, George H.W. Bush's upset of Ronald Reagan in 1980, even eventually produced a president, by way of a VP selection. But while it's nice to be reminded that we could get something wildly unexpected out of an early state in January (where's that Roy Moore campaign, anyway?), it's more important to remember that those surprises usually don't mean very much.