Measuring the McCain Effect, Part 1

U.S. Sen. John McCain: Getty Images PhotoThe minute U.S. Sen. John McCain clinched the Republican nomination, conventional wisdom immediately congealed around the assumption that the nominee would have the inside track in New Hampshire for the general election because of his long-term relationship with Granite State voters and activists.

This special relationship was held up as especially important, given the recent decline of Republican fortunes in the state.  It even inspired GOP chair Fergus Cullen to unleash his inner Yoda and describe McCain as his "only hope." Washington Post columnist E. J. Dionne was impressed enough with the McCain mystique that he labeled New Hampshire a "jump ball." 

All of this talk begs the question: How much is McCain's "special relationship" worth in the Granite State?

One way to measure this over the next four months is to compare McCain's poll numbers in New Hampshire with his numbers nationwide.  If McCain's relationship is truly special, he should outperform his national numbers in New Hampshire, all other things being equal.

One caveat: things are not exactly equal in the Granite State during presidential elections. Using Charlie Cook's Partisan Voting Index, let's compare New Hampshire to the nation.   

2000

• George W. Bush's share of the national two-party vote: 49.7 percent
• Bush's share of the New Hampshire two-party vote: 50.5 percent

2004

• Bush's share of the national two-party vote: 51.2 percent.
• Bush's share of the New Hampshire two-party vote: 49.5 percent

From 2000 to 2004, Bush's share of the New Hampshire vote dropped a full percentage point. (Let's keep in mind that in 2000 Ralph Nader won 4 percent of the vote in New Hampshire, 2.7 percent nationwide.)

This drop occurred in the midst of a 1.5 percent rise in his vote share nationwide from 2000 to 2004. 

Using Cook's formula and taking the average of the partisan differences over the last two elections:

(50.5 - 49.7) + (49.5 - 51.2) / 2 = a PVI of R - 0.5 for New Hampshire. 

Using the PVI, we would expect New Hampshire to vote slightly less Republican than the entire country this year (again, as a percentage of the two-party vote).  If McCain won 50 percent of the vote nationally, we would expect him to win 49.5 percent in the Granite State. 

Of course, the PVI is just a lagging indicator. It does not factor in the Democratic landslide of 2006, nor the Democrats' gains in voter registration since 2004. 

And because it's just an average, it does not factor in the size of the swing in New Hampshire's partisanship from 2000 to 2004. 

If a similar pro-Democratic swing is currently underway from 2004 to 2008, then New Hampshire may have a PVI of R - 3 when all is said and done in November.  In other words, a generic Republican nominee who won 50 percent of the two-party vote nationally would carry just 47 percent of the vote in New Hampshire.

But McCain, of course, is no generic Republican. More on that Wednesday.

Dante Scala teaches American politics at the University of New Hampshire and blogs at Graniteprof.