Russ Feingold and the 42 percent problem
Late last week, the Democratic Senatorial Campaign Committee released an internal poll that showed embattled Sen. Russ Feingold (D) down by just two points to wealthy businessman Ron Johnson (R).
The data showed, Democrats argued, that Feingold was still very much in the race -- despite other independent polling that showed Johnson with a high single digit edge.
But, the poll may instead have reinforced just how much trouble Feingold is in thanks to one number in it: 42.
That was the percentage of the vote that Feingold received in the initial ballot test against Johnson. And, that's a major problem for a politician who has spent the last 18 years in the Senate and has been in elected office in Wisconsin for the better part of the last three decades.
Because, traditionally in political campaigns, a lesser-known challenger will win a majority of undecided voters in the final weeks of the campaign since if voters aren't already on the side of the incumbent already why would they decide to get on board in the final days of a race?
That's especially true in an election cycle like this one where voters are willing to believe the worst about politicians and ready to make drastic change in the status quo in hopes of bringing some genuine change to the political process.
(Make sure to read Nate Silver who makes the counter argument about incumbents under 50 percent and their prospects of winning.)
The 2009 New Jersey governor's race provides a case in point of the dangers of Feingold at 42.
For the entirety of the race, Gov. Jon Corzine (D) wallowed in the low to mid 40s in polling matchups against Chris Christie (R).
Here's pollster.com's handy-dandy chart that illustrates Corzine's stagnation:
As election day 2009 neared, Democratic strategists insisted that the race was changing in their favor -- pointing to a narrowing in public polling.
And, indeed, there was a narrowing. But, the problem for Corzine was that while Christie was dropping, he was not rising in any appreciable way -- still stuck in the low 40s in vote share.
On election day, Christie won 49 percent to 45 percent as undecided voters moved heavily against Corzine.
The Corzine example suggests that the more important number to look at in a general election poll between a Democratic incumbent and a Republican challenger is not the spread between the two but rather the percentage of the vote that the incumbent receives.
In the 1994 tidal wave election that saw Republican re-take control of the House and Senate, most Democratic incumbents who were anywhere between 40 percent and 45 percent in hypothetical ballot tests lost as late-deciding voters went against them in droves.
It's not clear whether that same phenomenon will play out in three weeks time since, as we have written many times before, the comparison between this election and 1994 is not entirely exact.
But for incumbents like Feingold -- and he is far from alone on the Senate or the House side -- numbers like these paint a vivid picture of the challenge before them over the last 21 days of the campaign.
| October 12, 2010; 2:00 PM ET
Save & Share: Previous: Fast Fix: Nancy Pelosi as political liability?
Next: Daily Fix Poll: Tonight's high-stakes California gubernatorial debate