Gov. Martin O'Malley's re-election prospects seem markedly less certain than they did just weeks ago. At the very least, Maryland's governor may have a harder time winning his second term than his first.

According to this week's Gonzales Research poll of 807 likely voters, conducted July 13-21, the rematch contest between Mr. O'Malley and former Gov. Robert L. Ehrlich Jr. has tightened to a margin of just 3 points. The Democratic incumbent who occupies the second floor of the State House leads his Republican predecessor 45 percent to 42 percent, a difference within the poll's margin of error.

Let's unpack this result.

The 3-point lead represents a net 6-point gain by Mr. Ehrlich since the last Gonzales statewide poll, released in January before the former governor had even officially declared his candidacy. Mr. O'Malley led then by 9 points, 48 percent to 39 percent, in what was at that moment merely a potential rematch of the 2006 contest in which he beat Mr. Ehrlich by 6 points.

What's interesting about the latest numbers is that the tighter spread does not appear to be driven by voters' assessments of the incumbent. Indeed, Mr. O'Malley's net approval in the past four Gonzales statewide polls — dating back almost two years now, to September 2008 — has hovered between a high of 12 points and the present low of 9 points, 48 percent to 39 percent. That's not a typo, folks: Mr. O'Malley's current approval rating split is identical to what his lead was over Mr. Ehrlich just prior to his January announcement that he would seek a rematch.

So how it is Mr. Ehrlich has drawn closer despite very consistent approval numbers for Mr. O'Malley during a two-year period characterized by domestic economic problems for both personal and public budgets?

One might suspect that voters' worries about the economy are helping Mr. Ehrlich. After all, the poll's executive summary states that the share of people who cite the economy as their greatest concern "has stayed dramatically higher" than results over a number of previous years of polling, which "is not good for incumbents."

But this explanation falls flat. In fact, the percentage of Marylanders who rate the economy as the most important issue, after clearly spiking in a series of polls from just 10 percent in January 2008 to 61 percent in January 2009, actually leveled off slightly in the two most recent polls, in January 2009 (54 percent) and last week (52 percent).

If neither rising disapproval of Mr. O'Malley's performance nor growing angst about the economy explains why the race has tightened since January, what's going on here?

Putting aside the poll's sample (yes, there are some slight differences between last January's and the new survey's sample shares of Democrats and white voters that could explain a small bit of the variance), the shortest and most obvious answer is that a race featuring Mr. Ehrlich as an actual and active candidate is, as might be expected, simply more competitive that it was seven months ago, when he was still deciding whether to run.

Whatever the case, the final margin of the Ehrlich-O'Malley rematch this November could be closer than four years ago. And while a lot can happen in the 100 or so days between now and election day, what's most interesting about this contest is that most of the key variables are known and in some cases similar if not identical to four years ago.

What I mean is that, although Mr. O'Malley now has to run on a record as governor he did not have in 2006, this contest is the closest thing to what social scientists call a "natural experiment" because the candidates, their records, their political views and the state demographics are all pretty similar to what they were four years ago.

The primary differences are three-fold, only one of which favors Mr. O'Malley. He's had four years to travel the state, build relationships, deliver government services and in general prepare the groundwork for his re-election bid. Meanwhile, the two key differences advantage Mr. Ehrlich, and I suspect largely account for the tightening of this race: the national political-electoral winds have shifted away from the Democrats nationally, and the burdens of incumbency now fall on Mr. O'Malley's shoulders.

Short story? This should be an interesting race to follow right to the end because, yet again, the outcome is partially dependent upon, and could be indicative of, national factors beyond either candidate's control.

Thomas F. Schaller teaches political science at UMBC. His column appears regularly. His e-mail is schaller67@gmail.com.