One method of forecasting election outcomes is to focus on contests individually -- assessing voting trends in a particular district or state, the candidates' strengths and weaknesses, and any issues or circumstances that loom as potentially decisive factors. While the year's general tone may be taken into account, this method largely agrees with the late House Speaker Tip O'Neill: "All politics is local."
The macro alternative is to assess the national issue agenda, the tilt of the national political terrain, and the parallels to previous election years. This is essentially a sophisticated way of licking one's finger to ascertain which way and how fast the wind is blowing, and making a guess based on historical patterns, national polls, and subjective judgment.
For House elections, the race-by-race micro method usually works best. But about every four elections, it simply doesn't. It just didn't in 1974, 1978, 1982, and 1994, and (to a lesser extent) didn't in 1998. (In big macro elections, when a tsunami develops, even the most sophisticated micro analyses prove useless, because first-rate candidates often lose to inferior ones simply because of party affiliation.) Micro analysis once worked reasonably well as a predictor of Senate outcomes, but that hasn't been true in the past four elections. Close Senate races have tended in recent years to break in one direction.
At this stage in the 2005-06 election cycle, just about any micro analysis would predict minimal changes in the House and Senate: Too few seats appear to be truly in play for either side to roll up a big win. Yet, history reminds us that the midterm election during a president's second term tends to deal a setback to the president's party -- generally because that party's members felt discouraged and failed to show up at the polls while the opposition felt energized and turned out in large numbers. The presidential party can get into electoral trouble because it has run out of intellectual gas, grown lethargic, overreached, suffered a scandal, or hit economic hard times. Any of those factors could, at least theoretically, turn 2006 into a macro election.
The scandals surrounding Republican lobbyist Jack Abramoff and his ties to House Majority Leader Tom DeLay are making a lot of Republican strategists very nervous. And Rove has a far higher public profile than DeLay. So, GOP strategists have a new reason to fret over whether Rove knowingly disclosed the identity of an undercover CIA operative. We don't yet know which reporters Rove talked to, what he said, or whether he realized that Plame was a secret agent.
Rove's involvement in the Plame case is already creating a political problem for the White House and the Republican Party. We just don't know whether the problem will escalate into a GOP nightmare capable of damaging the party in the 2006 midterm elections.
The irony, of course, is that the Rove story is developing just as President Bush has tipped back "right-side up," as pollsters say, with his job-approval rating higher, although just barely, than his job-disapproval rating. His uptick is most likely the result of the London bombings. The two Gallup/CNN/USA Today polls conducted in June had pegged the president's overall job-approval rating at 45 percent and 46 percent. Now it's 49 percent, with his disapproval score, which had been 53 percent and 51 percent in June, 48 percent. Pew Research Center polling showed a bit more of an upward jump -- from 42 percent to 47 percent in approval, with disapproval dropping from 49 percent to 46 percent.
Obviously, hundreds of things will happen between now and the 2006 midterm elections. And the long-term political impact of the Rove/Plame affair and of heightened concerns about terrorism is impossible to know. But suffice it to say that Republicans certainly don't need another scandal right now. The current spotlight on Rove's behind-the-scenes maneuvering isn't something that heartens anyone in the GOP. The same can't be said, of course, for Democrats.