I am not a political scientist. In fact, the last and only time I even took a political science course was in the summer of 1991 at Le Moyne College as part of a summer program for "advanced" high school students. However, I believe I have compiled enough evidence to reverse a long-standing thesis within the political community. Contrary to both popular belief and previous findings, undecideds do not break overwhelmingly for the challenger.
In a famous 1989 article Incumbent Races: Closer than They Appear, through an analysis of 155 polls, Nick Panagakis articulated a set of findings, known as the "Incumbent Rule," that has almost become gospel among both the talking head and the blogosphere:
Our analysis of 155 polls reveals that, in races that include an incumbent, the traditional answers are wrong. Over 80% of the time, most or all of the undecideds voted for the challenger. (...)
In 127 cases out of 155, most or all of the undecideds went for the challenger:
Disposition of Undecided Voters %
Most to Challenger 127 82
Split Equally 9 6
Most to Incumbent 19 12
The fact that challengers received a majority of the undecided vote in 82% of the cases studied proves that undecideds do not split proportionally. If there were a tendency for them to split proportionally we would see most undecided voters moving to incumbents, since incumbents win most elections. Similarly, even accounting for sample error, it's clear from the chart above that undecideds do not split equally.
For poll users and reporters this phenomenon, which we call the Incumbent Rule, means: (...)
An incumbent leading with less than 50% (against one challenger) is frequently in trouble; how much depends on how much less than 50%. A common pattern has been for incumbents ahead with 50% or less to end up losing. Final polls showing losing incumbents ahead are accurate. The important question is whether results are reported with an understanding of how undecideds decide.(...)
Why do undecided voters decide in favor of challengers?
It seems that undecided voters are not literally undecided, not straddling the fence unable to make a choice - the traditional interpretation. An early decision to vote for the incumbent is easier because voters know incumbents best. It helps to think of undecided voters as undecided about the incumbent, as voters who question the incumbent's performance in office. Most or all voters having trouble with this decision appear to end up deciding against the incumbent.
Panagakis goes on to show that in roughly half of the few circumstances when the majority of undecideds broke in favor of the incumbent, the challenger was a well-known figure moving from an equivalent position (Senator to Governor or Governor to Senator), thus adding even more credence to the notion that people have already made up their minds about the incumbent.
However, I spent the last five hours going through the archives of Polling Report. Since 1998, over 220 polls that were taken in Governor, Senate and House races that involved an incumbent, that were taken within two weeks of Election Day, and that were the final poll of a race taken by a given polling company, when compared to the final election results of the race in question, the majority of undecideds broke as follows:
Disposition of Undecided Voters %
Most to Challenger 126 57
Split Equally 8 4
Most to Incumbent 86 39
These numbers are strikingly different from those found in Panagakis's study, which was conducted primarily using polls from 1988 and 1986. While the majority of undecideds broke in favor of the challenger a majority of the time, 43% of the time they did not (compared with 80%). If my research is accurate, and I believe it is, then it would be necessary to revise the Incumbent Rule. While incumbents under 50% but leading are in trouble, their predicament is not as bad as is often assumed. In fact, 43% of the time, such incumbents will be able to hold their lead or increase upon it.
Here is the complete breakdown of how the incumbents fared among undecideds in the 220 polls I surveyed:
% of Undecideds to Incumbent % of Cases
100 or more 11.4
0.0 or less 19.5
In a recent article
, Charlie Cook stated that "if a well-known and established incumbent picks up one-quarter to one-third of the undecided vote, he is lucky indeed." However, as we can see here, incumbents pick up at least 1/4 of undecideds in 65% of all elections, and 57.7% of incumbents pick up more than 1/3 of all undecideds.
Like I said at the top of this article, I am not a political scientist. However, unless someone can show me otherwise, I think we should all grow more skeptical about the chances of challengers who are behind in polls to defeat incumbents who are under 50%.