Pollsters Are Looking at the Wrong Race in Louisiana

Don't get me wrong; the Senate race is the only statewide one this year. (Louisiana holds its statewide elections every fourth year; the most recent one was in 2011.)

What I'm referring to is the Pelican State's unique (well, within the United States) system of elections, a holdover from when New Orleans was the largest city in the French North American colonies. A holdover from French civil law, Louisiana doesn't hold primaries; instead it pits every candidate that files from all parties into a great one-versus-the-world battle, known as a "jungle primary". If no majority is reached by a single candidate, the two highest-performing candidates advance to a runoff held on December 6. This allows for some great political theater, but it's hard on pollsters who have to decide which candidates to include in their polling, especially since it's possible for multiple candidates to run from each party.

It can also lead to some lopsided margins of victory. Despite receiving only 51% of the vote in 2004, Republican Rep. David Vitter won the open Senate seat by 22 points because 51% was just enough to avoid a runoff, while the Democratic vote was split between four different candidates. Sen. Landrieu, however, is likely the biggest beneficiary of the jungle primary system in Louisiana history: two of her three Senate victories were in runoffs, and her first election in 1996 saw her come in second place with only 22% of the vote--enough, however, to put her in the runoff, which saw her margin improve by 28 points.

The reason I bring this up is that in the past two months jungle-primary polls have become more common than matchup polls between Sen. Landrieu and her strongest Republican opponent, Rep. Bill Cassidy. In reality these polls tell us little about the ultimate outcome of the election: they all have Sen. Landrieu leading her three Republican opponents, but polling well below 50%. All the polls are telling us is that most likely there will be a runoff between Sen. Landrieu and Rep. Cassidy. But we could figure that out intuitively: from Sen. Landrieu's poor approval ratings to opposition to the Affordable Care Act; from President Obama's poor approval ratings to the fact that Sen. Landrieu has not broken 50% in any poll against Rep. Cassidy in the past year, we can tell that Sen. Landrieu is not avoiding a runoff unless she allows one of her opponents to reach 50%, in which case she would lose outright. At the same time, in any other system other than Louisiana's we might consider the possibility that Rep. Cassidy could break 50%, winning the Senate seat before Thanksgiving. However, although he leads Sen. Landrieu in most matchup polls, he's splitting Republican votes with two other candidates, which together make a not-insignificant 10% (more or less) in recent polling. While those two have no chance of winning the election, they do essentially force a runoff.

In other words, the Election Day election is pretty much a foregone conclusion--Sen. Landrieu will probably come in first place, followed by Rep. Cassidy, who will have a significant number of votes diverted by state representative Paul Hollis and retired USAF Col. Rob Maness. The jungle primary polls don't really tell us anything new at this point, unless the campaign of Hollis or Col. Maness suddenly start showing signs of life. It's the eventual head-to-head between Sen. Landrieu and Rep. Cassidy that will matter for the Senate, and it's most informative if pollsters focus on that matchup rather than on the jungle primary.