The race for president of the United States is now just about where it was last week — and the week before that, and the month before that. [IMGCAP(1)]

Two weeks from now, things may change a little, since the Democrats should get a small bounce from their convention. But you need a microscope to find movement in this race.

The general contours of this race have been in place for a while. While many observers act as if this vote or that campaign speech is of earth-shaking importance, the reality is very different. Most voters have made up their minds about Sen. John Kerry (D-Mass.) and President Bush, and those “early deciders” have split almost down the middle.

Kerry probably has a narrow edge among those already decided, and he has reason to be optimistic about getting the few undecided registered voters, since they haven’t signed up yet for another four years with Bush, even after seeing the president’s performance.

The map also looks a bit better for Kerry than for the president. He appears to be slightly better positioned to win key states that went Republican in 2000 — including Ohio, Florida, New Hampshire, Nevada, West Virginia and even Arizona — than does Bush to carry Gore states such as Iowa, Wisconsin, New Mexico, Oregon and Pennsylvania.

No, Kerry isn’t dramatically better positioned in his target states than is Bush in his targets, but in a tight contest, any advantage can’t be overlooked.

And as I’ve noted in earlier columns, I encounter anecdotal evidence all the time that more voters who are normally Republican than voters who are normally Democratic are going to sit this race out, creating another potential headache for Bush.

But Kerry, of course, has some reason for concern.

First, his selection of Sen. John Edwards (N.C.) as his running mate was greeted with enthusiasm by Democrats but seems to have done little to change the basic contours of the contest. Recent polls show that Kerry got anything from no bounce to a small single-digit bounce from his vice-presidential selection.

This shouldn’t surprise anyone, since VP picks rarely matter. But you knew that already.

Second, in spite of that fact that the Massachusetts Senator seems to be leading some great national crusade, the reality is that he isn’t winning the contest. Bush is simply losing it.

I’m sure there are some people (most likely Democrats) who really, really like Kerry and think he’s the answer to many of the nation’s ills. I just haven’t met any of them. (This probably guarantees that I’ll receive a flurry of calls and e-mails from voters who swear they just love Kerry.)

Third, while Kerry and his allies in anti-Bush interest groups are winning the war on television by outspending the Republicans in advertising, most television ads aren’t very valuable at this point.

So far, most of the “attacks” have been predictable and have only reinforced stereotypes. Voters can see the candidates on TV for themselves, which undercuts the effectiveness of the ads. So does the fact that voters are skeptical about the claims made in those ads. New, credible information in a TV spot could move some voters, but not many.

Reporters, producers, columnists and commentators who analyze these TV ads don’t seem to want to acknowledge the fact that the spots are ineffective, and I can understand why. After all, we already have too many reporters chasing too few real stories, so nobody wants to lose another potentially juicy storyline.

Fourth, future events could still change the way voters view the general election, and that’s probably good news for the president.

Sure, more bad news from Iraq, Afghanistan, the economy or someplace else could sink Bush’s prospects, but he’s already in serious re-election trouble. More bad news would only make a difficult situation worse.

For Kerry, however, good news from Iraq or progress in the war on terror, or some sort of terrorist attack in this country or elsewhere, could turn the current presidential race on its head.

With fewer than 16 weeks to go until Election Day, it’s time for election watchers and polling organizations to move from national polls to state polls in key states.

Yes, national numbers tell us something about the direction of the race, and they allow us to see how things are different (or the same) now as they were two months ago or six months ago.

But this race has boiled down to about a baker’s dozen of states: New Hampshire, West Virginia, Ohio, Iowa, Wisconsin, Pennsylvania, Minnesota, Missouri, Florida, New Mexico, Arizona, Nevada and Oregon. That’s where the polls matter, because that is where the race is really taking place. In other states, there is no contest of any importance.

I suppose you could add a couple of other states (North Carolina, for example), where a “cherry-picking strategy” might work, but all that talk about Virginia, Louisiana, Colorado or even Michigan being in play looks pretty half-baked now, barring a late tidal wave for one presidential ticket.

Kerry has the edge now because he is not Bush, while Bush still has a fighting chance because he is not Kerry. And in the end, the winner will claim a mandate that he almost certainly doesn’t have.

Stuart Rothenberg is editor of the Rothenberg Political Report.