On June 11, the day President Obama made his infamous "the private sector is doing fine" remark, the Real Clear Politics polling average had him leading Mitt Romney by 1.4 percentage points.
The gaffe capped a series of bad weeks for the Obama campaign. There were the leading Democrats -- including former President Bill Clinton and Newark Mayor Cory Booker -- who had criticized the attacks on Romney's time at Bain Capital. There was the May jobs report, which showed the economy had added a meager 69,000 jobs that month. There was the Romney campaign's fundraising, which, when added to that of the pro-Romney super PACs, was easily outpacing the Obama campaign's efforts
On June 12, this paper ran an A1 story that asked, in the very first line, "Is it time for Democrats to panic?"
On June 21 -- so, 10 days after "the private sector is doing fine" -- Obama's lead in the polls had increased to 2.3 percent. Ten days after that it was 3.6 percentage points.
All this raises the question: What actually matters in this campaign?
It's not gaffes and surrogates and assorted other political ephemera. While the conventional wisdom about which campaign is doing a better job tends to swing wildly, the polls have been far more stable. Case in point: The Romney campaign has been panned in the press over the last week, with the Wall Street Journal editorial page making headlines for writing that Romney is "slowly squandering an historic opportunity." Yet today is the Romney campaign's best day in the polls since June 18.
But nor are jobs reports and other economic data exhibiting much evident power over the election. A theory that put payrolls first would have expected substantial deterioration in the polls in recent months, as job growth has fallen sharply. Yet Obama's lead in April looked much like his lead this month -- despite the fact that the economy looked much better in April than it does today.
Another possibility is that the Obama campaign's ads are having a major effect. But the Obama campaign's ad spending is concentrated in a few battleground states, while the numbers we're looking at are national.
Perhaps it's candidate quality. Today's Washington Post/ABC News poll, which found the race tied 47-47, reports that Obama leads big on all the questions testing personal qualities. By large margins, voters think Obama understands them better, is a stronger leader, is more likely to stand up for what he believes in, and is more friendly and likable. So should this poll make us think the personal qualities matter a lot -- perhaps they're keeping Obama competitive where he'd otherwise be trailing? Or should it make us think they mean very little, as Obama's lead on these question is huge but he's tied in the poll? And, at any rate, anyone who is going to vote for Obama because they just like the guy has probably made up their mind already.
As a columnist and occasional talking head, I worry this next line will get me kicked out of the guild, but here goes: I have no idea what's driving this campaign. It's not what we're covering in the media. It's not what we're seeing in the economy. It's not what the campaigns are doing. It's not the personal qualities of the candidates. But here are a few observations:
1) The race is tight and stable. Obama's job approval -- which is in the 46-48 percent range -- basically hasn't budged since January. His lead against Romney has held between 0.5 percentage points and 3.5 percentage points since Romney secured the Republican nomination. Obama's strategists tell me they believe they have a floor of around 47 or 48 percent in the popular vote, and these numbers give me no reason to doubt them.
2) The corollary to that point is that the number of persuadable voters is very low. Romney and Obama are realistically fighting over three or four percent of the electorate, and of that three or four percent, they're fighting over the members who live in swing states. That is to say, they're fighting for a tiny fraction of a tiny fraction of voters.
3) The small ranks of persuadable voters are not going to be easy to persuade. The people open to being convinced by monthly jobs numbers and round-the-clock coverage of gaffes have already made up their minds. The potential voters who haven't made up their minds yet are, almost by definition, the kinds of voters who aren't easily moved by the kinds of political information that the rest of us are obsessing over.
4) The campaign is unlikely to break hard in a particular direction. Romney strategists occasionally compare 2012 to 1980, when undecided voters broke toward Ronald Reagan en masse. But according to Gallup, at this point in the campaign, Reagan was at 37 percent among registered voters, while Jimmy Carter was at 34 percent. Much of the remainder was going to third-party candidate John Anderson, who was at 21 percent. That was just a much more unsettled race than this one.
5) The election forecasting model I built with the help of political scientists John Sides, Lynn Vavreck and Seth Hill basically showed that it's very unusual for an incumbent president to lose in an economy exhibiting any amount of growth. Right now, that result looks pretty good. And I'm starting to think that the key variable there is as much "incumbent" as "growth."
For all that, we've still got four months until the election. Four months in which we're going to be covering the campaign also nonstop. So help me out here: What's going to matter between now and then?