Last week I argued that if the “political environment” determines election outcomes, ’06 will be a great year for Democrats.

Rarely have we had a president so unpopular going into a midterm; when we have, it has led to massive congressional turnover. Moreover, while Democrats have expanded our advantage on key domestic issues, Republicans’ once vast lead on national security has been largely neutralized.

However, as you have guessed, there is more to the story. Two other factors complicate predictions. First, things change. Second, this election cycle is setting up as a powerful test of whether the political environment on the one hand or structural dynamics on the other is the more important determinant of election results.

Before addressing the structural questions, one must recognize the real possibility of change. Prognosticators make no greater error than simply projecting the current political situation into the future, assuming present trends will continue.

This spring, the administration had been planning to announce withdrawal of 30,000-50,000 U.S. troops from Iraq. Though less likely than it appeared a month ago, such a withdrawal, combined with a strengthening economy, could lift the president’s ratings. While Bush will never see 60 percent approval again (absent another 9/11), the difference between going into an election with 34 percent and 48 percent is real.

Even more important than the potential for change may be the limitations of structure. Let me borrow an analogy from sand-pile physics. Imagine a pile of sand. Pour more on top, one grain at a time. At some point, an avalanche will result. However, no one can predict exactly which new grain will cause the avalanche.

In searching for causality, should we focus on the qualities of the added grain or the structure of the sand pile? When we analyze our debacle in 1994, should we fixate only on the environment or rather on the underlying political structure?

Open seats constitute one component of that structural dynamic. Because some 98 percent of incumbents are reelected, we often look to open seats as the source of turnover. They do play a role. However, in big waves it can be marginal. In 1994, 37 percent of the seats Democrats lost had been open. Of the Republican seats lost in 1982, just 19 percent were open.

System equilibrium is another critical structural element. A fundamental insight of modern physics is that the magnitude of cause and effect are often unrelated because system dynamics are decisive. Often big changes do not reflect big causes, but rather systems that are in disequilibrium. Just as one grain collapses an unstable sand pile but has no effect on a stable one, so too the state of political situations can render them prone to, or resistant to, significant turnover.

One measure of political instability: the number of Republicans holding seats that vote Democratic for president and vice versa. When big political waves hit, that is precisely where much of the action is. In the two prior presidential elections, Bush (the father) or Reagan had won 30 of the 34 seats Democratic incumbents lost in ’94. Similarly, two-thirds of the Republican incumbents who lost in ’82 were running in districts presidential Democrats had won just previously.

Today, though, there are fewer mismatched seats than at any point in recent history. Going into 1994, 53 Democrats held seats won by Bush in 1992. Today just 18 Republicans hold seats won by Kerry. So, while forces in the political environment push strongly in a Democratic direction, they are acting on a relatively stable structure: Hence the test.

Which will prove more important? That is the big analytic question of ’06.

Mellman is president of The Mellman Group and has worked for Democratic candidates and causes since 1982, including Sen. John Kerry (D-Mass.) in 2004.