Lynn Vavreck is a professor of political science and communication studies at UCLA. She is the co-author of “The Gamble: Choice and Chance in the 2012 Election.” This column appeared Nov. 2 as a post on The New York Times blog “The Upshot.”

In 1992, Bill Clinton’s strategist, James Carville, made famous a presidential campaign mantra: It’s “the economy, stupid.” He meant that presidential election outcomes are tied to the nation’s economic conditions, something that should not surprise many people. The same, however, is not true for midterm elections. Those are much more tied to the president himself.

In presidential elections, if the economy grows during the election year, the incumbent president (or the incumbent’s party) is more likely to be re-elected. The correlation is so impressive that political scientists often refer to national economic conditions as part of the structural or fundamental conditions that drive election outcomes.

But while the nation’s economy is a strong shaper of presidential election outcomes, the president himself is the shaper of congressional outcomes. In years when we are electing a president along with a Congress, the newly elected president’s party typically picks up seats in both the House and the Senate. You can think of this as hopping on the party bandwagon — some candidates in the president’s party get swept into office on the winning presidential candidate’s coattails just because they are in the right party at the right time.

But in the nonpresidential election years, that same president’s party typically loses seats in both chambers of Congress. On average, over the last 80 years, the president’s party has lost 27 seats in the House and four seats in the Senate in midterm election years.

What bandwagons beget, time takes away.

Political scientists call this pattern of pulses in on- and off-year election cycles the surge and decline of congressional elections, a phrase first coined in 1960. The pattern was originally attributed to two things: changes in the short-term forces that pull voters to one or the other party between elections and changes in the set of voters turning up at the polls — fewer voters in midterms than in presidential years.

These patterns, along with the stunning advantage bestowed upon incumbent members of Congress — more than 90 percent of incumbents who run are re-elected in both chambers — are the fundamental conditions that shape midterm elections. The economy is not part of this structure.

Instead of rewarding or punishing the incumbent president for his handling of the nation’s economy, in midterm years voters address the president more directly — by penalizing his party members, on average, but also by calibrating that punishment based on how the president is doing his job. Average approval ratings of the way the president is “handling the job” explain more of the variation in seat loss than the economic indicators.

And it appears as if the approval rating is made up of more things than just the economy. Where voters jump on the winning bandwagon in the presidential election years, they put a finger in the wind to measure the political atmosphere in the midterms. The happier people are with how things are going generally, the less likely they are to punish members of the president’s party.

Midterm vote choices are by and large driven by different fundamental or structural conditions than presidential elections, namely incumbency and the surge and decline of presidential-party fortunes between on- and off-presidential election years.

Given the last several decades, we’d expect, on average, a loss of four seats in the Senate this year. If what happens is close to that number, the upshot should not be a doomsday story for the president and his party, but instead, a tale of doing about as well as average.

Recently, UCLA’s Marschak Colloquium welcomed Vavreck as a speaker. She shared her research on the effect a president’s popularity has on the success of his party in mid-term elections.

Lynn Vavreck - Marschak Speaker from Brad Rowe Video on Vimeo.