(Had Senate elections been held in December, Republicans would have had an 80 to 90 percent chance of retaining control. Now that GOP Sen. Ben Nighthorse Campbell of Colorado has decided not to seek re-election, the Republicans' chances have dropped to 60 or 70 percent. Such is the nature of this incredibly volatile environment.)
Earlier this year, attacks by the Democratic presidential candidates campaigning in Iowa and New Hampshire clearly took a toll on Bush. His approval rating in the CNN/USA Today/Gallup poll dropped from 53 percent on January 12-15 down to 49 percent, the lowest of his presidency, on January 29-February 1. His rating then hovered between 50 and 52 percent until the latest sampling, March 26-28, when it ticked up to 53 percent again. Bush pulled narrowly ahead of presumptive Democratic nominee Sen. John Kerry of Massachusetts in the latest survey, both in two-way trial heats and in one that included independent candidate Ralph Nader.
While these aren't exactly earth-shatteringly large swings of public opinion, they are important, because Bush's approval ratings are teetering in a pivotal range. Past presidents with higher job-approval ratings at this point in their tenure went on to win; those with lower ratings lost.
Bush's strategists are right to point out that no past president with a 53 percent job-approval rating in March has gone on to lose, but it is also true that no elected president has trailed the other party's nominee-to-be after the end of January and gone on to win.
Although President Reagan briefly trailed Sen. Gary Hart, D-Colo., in early 1984, he never trailed former Vice President Walter Mondale that year.
This year, there is volatility even on the state level. Colorado wasn't on the early lists of potential battleground states, yet two recent polls -- one for a Republican Senate candidate, the other for a Democratic one -- indicate that Bush has less than a 6 point lead over Kerry. Those results caused strategists on both sides to scratch their heads and wonder whether Colorado's growing Hispanic vote and the movement of new voters into the state has dramatically changed the state.
In early 2000, Florida was not expected to be the big problem that it turned out to be for Bush. But Florida had changed faster than experts realized. Now, some observers in Florida think that because the Kerry campaign and pro-Kerry independent groups have not weighed in with as much Sunshine State advertising as Bush has, the Democrats might be writing off the state.
Insiders suggest, however, that Democrats are merely avoiding wasting money in the conservative, increasingly Republican panhandle and are keeping a lid on their spending in Miami because it is such an expensive market and because their dollars go much further elsewhere. In every other battleground state, the two sides are spending roughly the same amounts on television.
Ohio might be the most important battleground of this election. Bush seems to have real problems in northern Ohio, where the loss of manufacturing jobs has been particularly acute. In the state's southern half, which is more conservative socially and culturally, he is doing much better. Bush campaign strategists argue that southern Ohio has much more in common with neighboring Kentucky than with, say, industrial Michigan.
What is different about this election is that never before have we looked at a presidential election on a state-by-state and even media market-by-media market basis this early. Perhaps that is because, until four years ago, no one thought it conceivable that in a country of 290 million people and 106 million participating voters, a presidential election could be decided by 537 votes in a single state.
No one expected the last race to be that close. This time, virtually everyone expects a very, very close contest. And, as a result, neither side is likely to leave a single arrow in its quiver.