The Incumbent Rule

We will know next Tuesday evening who the next President will be.  It has been a fascinating election season.  We have been inundated with information like no other election, but there has been some analysis missing from the mainstream media.  There has been very little talk about something called the “incumbent rule”.  This is a theory that suggests that if the incumbent does not poll above 50%, particularly in the last few weeks of the campaign, they do not win.  For those who think that the mainstream media is in the tank for Obama, the lack of stories on this rule is additional confirmation of that belief.

James Carville, long time Louisiana political advisor, has said, “Incumbents get the last poll”.  In non-Cajun language, he is saying that the incumbent will receive the same percentage in the real election that he gets in the final poll.  This means that the challenger would get his poll number plus the remaining undecided voters.  There is a wonderful historical breakdown of this rule in a 2004 story by Mark Blumenthal that puts its accuracy at 86% for presidential contests.  The media is so hypocritical on this issue, and only refer to the incumbent rule when it suits their agenda.  Republican Senator Scott Brown’s (MA) is not receiving 50% in his race, and the incumbent rule is mentioned frequently in reporting on this contest.  For some reason the national media has shied away from this analysis of the presidential race.

The national polls for the last month have shown Romney consistently ahead.  The media coverage has focused on the state polls, which may be an attempt to construct a case for an Obama victory.  Even with this state focus, the incumbent rule can be used to help determine the outcome.  There has been debate about the validity of sampling for some polls, so hopefully using the last 4 polls mitigates that issue.  Based on polls released through Friday, gathered at, there are 11 states that have the incumbent below 50% in at least half of the polls…

  • Electoral starting point for the other 39 states: Obama 200 – Romney 191 (this includes NV going to the incumbent, since he has reached 50% in all of the last 4 polls)
  • The incumbent has not reached 50% in any of the last 4 polls in VA, NC, or FL.
  • If these states are added to the challenger: Obama 200 – Romney 248.
  • The incumbent has not reached 50% in 3 of the last 4 polls in NH, PA, OR, MI, or OH.
  • If these states are added to the challenger: Obama 200 – Romney 313.
  • The states that show half of their final 4 polls with the incumbent above 50% are WI, IA & CO.

Even with these 25 electoral votes going to the incumbent the total count would show Obama 225 – Romney 323.  This is not a prediction of the outcome, but an analysis of the polls run through the incumbent rule.  This type of examination is almost non-existent in the main stream media countdown to the election.  Why isn’t this part of the story in the final days of this campaign?

This incumbent rule, like any other theory could be wrong.  With its track record, however, it is at least as important as some of the stories that are making headlines across the media.  No matter who wins this election, come November 7th we will still have the same news sources that we have now.  This is just another example of the media deciding, with a good dose of bias, which stories are worth reporting.  It is a shame that NBC, CBS, ABC and the New York Times are not on the ballot Tuesday, running to keep their jobs.  Their poll would definitely show them below 50%, and with the incumbent rule in effect they would likely lose their offices.