A struggling economy has historically meant trouble for the president’s party in midterm elections. So it comes as no surprise that Democrats are girding for a tough November.

They should be. The economy is slowly recovering but remains on its sickbed, and most signs still point to a rough cycle for the party. Political analysts expect Republicans to make gains — possibly significant ones — in Congress in November, threatening to retake the House and maybe even the Senate.

But digging deeper, beyond the national numbers, reveals at least a few glimmers of hope for Democrats — still fairly distant and faint, but bright enough to get campaign strategists scanning the horizon and weighing the odds.

That is because different parts of the country are recovering at different rates — and, in a bit of electoral good luck for the Democrats, some of the areas that are beginning to edge upward more quickly, like parts of Ohio, Pennsylvania and New York, happen to be in important battlegrounds for the House and the Senate.

"A lot of the trend lines are turning positive in many of these contested areas," said Mark Zandi, a chief economist for Moody’s Analytics. "It really boils down to: Is there enough time for the trend lines to trump the still pretty difficult conditions in the minds of the voters?"

Certainly, the economy will not be the only factor voters weigh at the ballot box. There are areas that weathered the recession relatively well where Democratic incumbents are still in danger. There are also places trailing the national recovery, like Nevada and Florida, which could hurt the party.

But officials from both parties agree that the economy remains — at least, for the moment — the paramount concern for most voters. And studies have shown that local conditions influence the degree to which economic worries figure into voters’ decisions — a kind of Tip O’Neill rule of pocketbook voting.

A detailed examination of House and Senate seats in play, alongside state and local economic data compiled by Moody’s Analytics for The New York Times, yields some surprising bits of encouragement for Democrats but also adds color to the overall daunting picture confronting the party. At the very least, any such signs of hope are certain to affect the strategies being hammered out now in campaigns.

"It’s too early to know how the more rapid economic improvement in parts of the country will influence voters’ decisions," said Rep. Chris Van Hollen, chairman of the Democratic Congressional Campaign Committee. "But there’s no doubt that voters’ perception of the future of the economy will be a main factor in determining their vote."

The Times has identified 114 House seats and 17 Senate seats that are expected to be the most competitive in November. The largest numbers of closely contested elections are expected to be in Pennsylvania, New York and Ohio, where The Times projects a combined 26 competitive House races. All three states, coincidentally, are considered to be on the leading edge of the nation’s recovery. Since December, they have added jobs at a faster rate than the country as a whole and even led the country in the total number of jobs added in April. One reason is that manufacturing, a traditional backbone, has been on the rebound; another is that these states generally did not suffer as acutely as other regions from the housing boom and bust.

While much attention has been paid to the nation’s stubbornly high unemployment rate, political scientists have found little correlation between that measure and midterm elections results. Instead, they have found more broad-based indicators, particularly real personal disposable per capita income, which measures the amount of money a household has after taxes and inflation, to be better gauges.

About a quarter of the House races The Times expects to be competitive in metropolitan areas that are projected by Moody’s to have had positive, albeit mostly modest, after-tax personal income growth between 2009 and 2010. They include several swing districts in upstate New York that have long struggled economically but largely escaped the real estate bubble.

In New York’s 24th congressional district, for example, Rep. Mike Arcuri, a Democrat, said he senses the improvement but has to be careful about how he couches it with voters.

"They just want to know there’s light at the end of the tunnel," he said. "I think they’re just starting to see that."

The economic forecasting company also predicts housing prices will rise in metropolitan areas connected to a fifth of the competitive House seats identified by The Times. The forecasters project three-quarters of the competitive districts will experience employment growth in their corresponding metropolitan areas between the fourth quarter of 2009 and the fourth quarter of 2010.

The picture is grimmer for Democrats when it comes to the 17 Senate races that are expected to be in play. Only one state on the list, Connecticut, is projected to have experienced disposable income growth between 2009 and 2010. On the other hand, all but one state, Nevada, is predicted by Moody’s to experience job gains between the fourth quarter of 2009 and the fourth quarter of 2010.

Unfortunately for Democrats, important Senate battlegrounds like Nevada, where the Senate majority leader, Harry Reid, is fighting for his political life, and Florida are predicted to be among the last to begin experiencing steady employment growth, with economists not forecasting consistent increases until late 2010 or early 2011.

Both states suffered the worst of the housing speculation. Making matters worse for Democrats, both lived through a long, happy Sun Belt boom, so economic turmoil is a relatively new experience for residents — and studies indicate it is such locales where economic anxiety is most likely to drive voters’ decisions.

Ohio promises to be an interesting laboratory. At least eight competitive House races are expected there, as well as a hard-fought Senate race. A struggling Rust Belt state, it gained more than 37,000 jobs in April, the largest increase for the state in years. But Ohio remains a long way from replacing the roughly 400,000 jobs it has lost since 2007.

"The question is: how big are people’s memories?" said James Diffley, a managing director at IHS Global Insight.

Historically, political scientists have found that voters’ memories tend to be short. Larry M. Bartels, a political scientist at Princeton University, has studied the impact of economic conditions on presidential elections and found that it is the second and third quarters of the election year that matter most.

In the end, however, the ultimate deciding factor will be voters’ perceptions — not how well the economy is actually doing but how well voters believe it is doing.

With communities climbing from such dire straits, it remains to be seen whether modest economic gains will be enough.

© 2010 The New York Times Company