Daniel Marsula, Post-Gazette Click photo for larger image. Thomas E. Mann is a senior fellow at the Brookings Institution and co-author with Norman J. Ornstein of "The Broken Branch: How Congress Is Failing America and How to Get It Back on Track" (Oxford University Press). This article appeared in The Washington Post.

WASHINGTON -- There's probably no way congressional Republicans can lose this fall, no matter how unpopular President Bush is or how unhappy the voters are with the war in Iraq. That's the prevailing view in Washington today.

But it's wrong.

If history is any guide, we're heading into a major political storm. And that means we could see a national tide in November that will sweep the Democrats back into the majority.

Virtually every public opinion measure points to a Category 4 or 5 hurricane gathering. Mr. Bush's job-approval rating is below 40 percent, and congressional job approval is more than 10 percentage points lower. Only a quarter of the electorate thinks the country is moving in the right direction, and voters are unhappy with the economy under Mr. Bush. Finally, Democrats hold a double-digit lead as the party the public trusts to do a better job of tackling the nation's problems and the party it would like to see controlling Congress.

What's causing the skepticism about Democrats' chances for victory in November are changing election patterns. Until recently, one of the few iron laws of American politics was that the president's party loses House seats in midterm elections, with the size of the loss depending on how many seats are at risk and how the public evaluates the president's performance. But all that seemed to change in 1998.

That year, a healthy economy plus public distaste for the Republicans' impeachment drive against President Bill Clinton allowed the Democrats to gain five seats. Four years later, with Mr. Bush in the White House, Republicans picked up eight additional House seats in the midterm election. GOP fortunes were boosted by the fact that fewer seats were at risk, the president's post-9/11 approval ratings exceeded 60 percent throughout 2002, and the Republican campaign elevated terrorism over the economy as the central public concern.

In the past five House elections, the number of seats changing party hands and the number of defeated incumbents have been historically low. Because of gerrymandering, stronger party-line voting and Americans' fondness for moving, the partisan makeup of House districts has become more lopsided, with many safe Republican and Democratic districts and very few competitive ones.

Since 1994, when the Republicans won back the House, there has been a dramatic decline in the number of majority-party members out of sync with the partisan makeup of their districts and of those who won the previous election by a narrow margin.

All this leads to the consensus that the Republicans can't lose this fall. But I think they can.

When there's no strong national issue at stake, local forces (a district's partisan makeup, the incumbent's reputation, the challenger's resources, etc.) dominate congressional elections. But a sharply negative nationwide referendum on the party in power -- causing a national vote swing of 5 percentage points or more -- buffeted local factors in the 1946, 1958, 1966, 1974, 1982 and 1994 midterm elections, producing losses of 26 to 56 seats.

In each of those elections, changes in the national vote were not distributed evenly across districts. The party losing ground found itself besieged in districts previously thought to be safe, where the average swing was double or more the national swing.

The new pattern of uncompetitiveness that developed after the 1994 Republican landslide has not been tested by a surly electorate. The Democrats' hopes rest on intense public unhappiness with Mr. Bush and the GOP -- and enough districts in play -- to allow them to pick up the 15 seats they need to become the majority party.

What might keep a national tidal wave from developing this year?

First, party divisions may have hardened so much that few voters will be open to conversion. Party-line voting is at its highest level in decades. While many GOP voters are critical of Mr. Bush and the Republicans in Congress, many may return to the fold by November. On the other hand, there are enough pure independents and weak partisans to make a significant shift in the national vote possible.

Second, polls reveal a Democratic advantage in the level of interest in the midterm elections comparable to what the Republicans enjoyed in 1994. But it's still uncertain whether Republicans' traditionally higher turnout rates, combined with the GOP's vaunted get-out-the-vote operation, will significantly reduce or eliminate that advantage.

The final factor is the Republican decision to pursue a risky strategy: turn their greatest liability -- the war in Iraq -- into an asset by linking it to the broader fight against terrorism. The risk is that this will reinforce the public's association of the GOP with an unpopular war. The potential gain is that it will allow the Republicans to highlight alleged Democratic division and irresponsibility.

Will it work? No one doubts the Republicans' resourcefulness and discipline in castigating the "cut-and-run" Democrats while beginning to withdraw some U.S. forces before November. Less certain are the Democrats' skills in framing the election as a referendum on the Bush administration and Republican Congress.

Public unhappiness with the Bush administration and Congress might diminish over the next several months if the economy and the situation in Iraq improve. Republicans may succeed to some extent in shifting public focus from their past performance to a choice about future directions and policy. Efforts to rally and turn out the Republican base may compensate for the Democrats' advantage in the intensity of public discontent. Extraordinary efforts to protect potentially vulnerable Republican incumbents may pay off. And the limited number of GOP seats at risk may prove an insurmountable obstacle for Democrats.

But my own reading is that the odds favor a Democratic takeover of the House. The 15 seats that the party needs for a bare majority is well below the range of minority-party gains in past tidal-wave elections. The national winds blowing against the GOP are strong and have not diminished over the past nine months. Credible progress on the ground in Iraq before November is implausible.

The public's harsh evaluation of the president's performance on the economy is unlikely to be reversed by Election Day. Prospects for significant legislative achievements in the remaining months of this Congress are remote. Enough seats will be in play (including some that Republicans carried in 2004 with more than 60 percent of the vote) to allow Democrats to gain majority status in the House.

Prospects for a Democratic majority in the Senate are less bright, given the limited number of Republican seats in play. But even here, a national tide could tip all of the close races in the same direction, allowing the Democrats to hold all their threatened seats and to win the six Republican seats they need to take control.

Energized voters can hold their government accountable and throw the rascals out. Chances are good that, this fall, they will avail themselves of the opportunity.