RALEIGH – For months, I’ve been urging folks not to watch the election match-up polls too much. They discourage supporters of lesser-known candidates, give those of better-known candidates a false sense of security, and obscure important matters of strategy, issue contrasts, and the news environment that can erase – or open up – a double-digit lead faster than you can say Jack Robinson.
Or, if you like, change the first name to Vernon.
But now, with only three weeks left in the race, the poll numbers start to take on real significance. By now, virtually all partisan voters have already decided whether to back their horse. Most undecided voters who are actually going to show up on Election Day have begun to pay attention to races. Their views may not be fully formed, but at least they exist. A month ago, many such voters still weren’t sure who the candidates were.
None of this means that the dynamics of a given political race can’t change in the last three weeks. As I noted a few days ago, 1984 Democratic gubernatorial nominee Rufus Edmisten had a double-digit lead over Republican Congressman Jim Martin just two weeks out. He ended up being washed away by the Reagan tsunami, a wave that did not really form until after Reagan’s debates with Walter Mondale (the president was judged to have lost his first one, by the way). More recently, former Georgia State Sen. Sonny Perdue was badly trailing Democratic Gov. Roy Barnes just days before the 2002 election. An early October poll had his 17 points behind Barnes, who was widely expected to win reelection easily and had, not coincidentally, vastly outraised Perdue in campaign cash. But a last-minute break of undecided voters for the challenger gave Perdue 52 percent of the vote to Barnes’ 46 percent on Election Day (the same margin that Mike Easley beat Richard Vinroot in 2000, as it happens).
So it’s time to start paying attention to the polling numbers. You’ll have to keep watching them, though, if you want to keep up with the twists and turns of a mercurial electorate.
On the presidential race, this helpful summary from the folks at RealClearPolitics.com is a good place to start. They track all of the national face-to-face polls and create averages of results from roughly the same time frame. This is not really kosher, statistically speaking, as the various polls have differing methodologies. But I won’t tell if you won’t, and it is a handy summary, as is their electoral-map projection. Another good version of the latter is available at the ElectionProjection.com site of Scott Elliott, who has North Carolina ties.
On the North Carolina races for U.S. Senate and governor, the next round of polls will give us a better sense of what’s going on. You can see a summary of the Bowles-Burr numbers here. As for the Easley-Ballantine match-up, the key number to watch is the governor’s. The late-September and early-October polls had him several points above 50 percent, which far more important than his double-digit margin over the lesser-known Ballantine (who will almost certainly get a minimum “GOP floor” vote of about 45 percent). By attacking the governor’s record on taxes, the GOP hopes to drive that number below 50 percent. If the next round of polling shows no such downward movement, that will be telling.
Courage.
Hood is president of the John Locke Foundation and publisher of Carolina Journal.