The fundamentals of electoral change in the 2010 midterm elections are unmistakable. Electoral change depends on two general sets of political conditions: those of the previous election or elections and those of the current election. Both favor the Republicans this year.

Republicans are coming off two consecutive elections in which the prevailing political winds ran strongly against them. 2006 was a big year for the Democrats. The controversial Iraq and Afghanistan wars were dragging on, and the political fallout from the mishandling of the aftermath of Hurricane Katrina dropped President Bush's approval ratings to 41%. Compounding Republican problems were several congressional scandals, particularly the Mark Foley scandal that broke in September, that put formerly safe Republicans on the defensive. Together, these conditions cost Republicans 30 House seats, six Senate seats, and control of both the House and the Senate.

The political climate in 2008 was scarcely any better for Republicans. Having disappointed his base with his immigration reform proposal and government spending levels, President Bush's approval rating descended to 31%. An unpopular presidency, along with the Wall Street meltdown in September, and the base-defying presidential candidacy of John McCain led to the election of Barack Obama and the addition of 24 more Democrats to the House and eight more to the Senate. Having lost a total of 54 House and 14 Senate seats in two elections, 2010 looks good for Republicans, if for no other reason than that it is not 2006 or 2008.

Although a return to normalcy in itself should be a welcome relief to Republicans, the political climate of 2010 favors them as well. President Obama's approval ratings are generally running in the low to mid 40s, well below what is needed to significantly blunt midterm seat losses for the in-party. Approval of Congress—with both chambers controlled by Democrats—is in the low 20% range. The generic congressional preference poll numbers generally give an edge to the Republicans. The recovery from the recession is slow and sputtering. A CBS poll in mid-July indicated that 75% of Americans expect the effects of the recession to last another two years or more (CBS News 2010). In mid-June, only 33% of Americans reported in a Pew Research survey that they thought that the Obama administration had done an excellent or good job in dealing with the BP oil leak catastrophe in the Gulf of Mexico (Polling Report 2010). Quite apart from any comparison to 2006 or 2008, the political climate of 2010 looks to be hospitable for Republicans.

There are some factors, however, that may blunt Republican gains. Party polarization may keep the Democratic base intact. This could act as a firewall to Republicans reestablishing their post-1994 majorities. The advantages of incumbency and the general decline in recent decades of the number of competitive seats in play may also limit Republican gains. And while President Obama's approval ratings are not very strong, they are not especially weak either. With these restraints and changes in the general political climate in mind, the question for election forecasters this year is not whether Republicans will make inroads in 2010, but how large will those gains be?

In the articles that follow, five forecasters or teams of forecasters present their models and predictions for the 2010 midterm elections to the U.S. House. Alan Abramowitz also offers a Senate forecast and Carl Klarner forecasts this year's state legislative elections.

The House forecast equations use a wide array of predictors, though several indicators appear in multiple models. Common predictors include some form of a presidential approval measure, the number of seats a party held before the election, the generic preference poll (i.e., if the elections for Congress were being held today, for which party's candidate would you vote in your congressional district?), and a midterm dummy variable in equations estimated using both on-year and midterm elections. My forecast equations consist of an indicator derived from the Cook Political Report, the number of seats previously held by the party, and an index of presidential approval. Alan Abramowitz uses the number of seats previously held, presidential approval, the generic preference poll, and a midterm election variable. Mike Lewis-Beck and Charles Tien use a measure of change in real disposable income, presidential approval, and a midterm election variable. Joe Bafumi, Bob Erikson, and Chris Wlezien take a two-step approach, first forecasting the national House vote using the generic preference poll and the party of the president, and then forecasting district-level results based on the predicted national House vote and the presidential vote in 2008 in open-seat elections and the predicted national House vote, the 2008 district Democratic House vote, and whether the incumbent was in his first term in incumbent elections. Finally, Alfred Cuzán relies on the number of seats won by a party in the previous election, a variable for the 1932 and 1948 elections, a presidential incumbency variable, a midterm variable, and economic variables measuring growth and inflation.

To summarize the forecasts (bearing in mind that a good deal of error can be expected to surround each point forecast), in keeping with the prior review of the previous and current political climates, there is a broad consensus among the models that the Republicans will make substantial gains in the House in the 2010 midterms. There is not a consensus, however, over how large those gains will be. There is a 30-seat spread between the low and high end of the seat change forecast range, with two forecasts giving an edge to Democrats in controlling the House and three placing the odds in the Republicans' favor. Lewis-Beck and Tien forecast a 22-seat gain for the Republicans. Their 200 seats would leave Republicans 18 seats short of a majority. Cuzán forecasts Republican gains of 27 to 30 seats, leaving Republicans with 205 to 208 seats and Democrats with continued control of the House. Abramowitz predicts a 43-seat gain for the Republicans. Since he uses a 179 pre-election seat base, this outcome would install a new Republican majority in place by five seats. Bafumi, Erikson, and Wlezien predict that Republicans are likely to gain 51 seats, which would give Republicans 229 seats and a 12-seat majority. Finally, my forecast is for Republicans to gain 51 or 52 seats, giving them a 12 or 13 seat majority. Whether Democrats or Republicans control the House in 2011, their majority is likely to be much narrower than the current Democratic majority. This may well present a roadblock to the Obama administration's legislative agenda and will quite probably make control of the House a real question again in 2012.