Archive for October 28th, 2010

Can the Democrats Hold the House?

(NOTE: click – or scroll down – for my color coded “election night guide” designed to make it easier to follow the results on election night and visualize just how the night is going.)

Can Democrats retain control of the House of Representatives?   Of course.  Are they likely to?   Probably not.   Indeed virtually every political pundit and specialist dismisses talk of a Democratic comeback as wishful thinking.   The depth of that consensus is both impressive and troubling.  It’s impressive because with such clear expert consensus, it’s hard not to consider a GOP House a done deal.   It’s troubling in that whenever there is too firm and consistent a consensus, it feels like a kind of groupthink.   Moreover, elections can bring surprises.   So today I’ll explore the scenarios that see the Democrats defying the odds and remaining in control of the House.

1.  Winning the close ones. Let’s start with the New York Times rankings of races.   The NYT has 152 safe Democratic seats, and 39 leaning Democratic.  There are 174 safe Republicans, and 28 leaning Republican.

If both parties win all their leaners, then it stands at 191 Democrats and 202 Republicans.   That leaves 42 “tossups.”  Assuming that toss ups can go either way, the Democrats would have to win 27 and lose only 15 to hold the House.   That is certainly conceivable.   The Democrats would have a harder time holding on to the races leaning their way, however, which increases the opportunity for the Republicans to expand the size of their gains, and is one reason some are predicting a Republican wave.

Still, winning the close ones is not out of the question for the Democrats.  Most of these are held by Democratic incumbents who, despite the anti-Washington mood, can use the incumbency to their advantage.   There are also some signs that Democrats are “coming home” and efforts to enthuse voters may increase Democratic turn out.    In very close races get out the vote efforts and personal contacts matter; often the incumbent can do better on those fronts.

2.  Systemic bias in the polls.   Polling of House races is often less precise and sophisticated than polling of Senate or Presidential races.   Moreover, it’s spotty.  Yet there could be assumptions about GOP turnout that cause polls to weight their data in favor of one party more than they should.  Rasmussen, for example, has been accused of tending to error on the side of increasing GOP vote count (and there is evidence backing this up).  Finally, there is the recurring concern that people without land lines are not polled, which may under count Democrats.  Much of that, though, can be adjusted in how the data is weighted.

A casual scan of House (and Senate) polls over the past two weeks show that numerous races, including leaners for both parties, have single digit differences, a large number of them under 5%.   Even if the systemic bias is small, that could mean a false read on ten or more races.   Flip ten races and you get a significantly different result.   Instead of, say, the Republicans up 49, the Republicans could be up only 29.  Or they could be up 69.

The chance of systemic bias built into the assumptions of pollsters is real, but it’s also the easiest to grasp if you’re a Democrat wanting to believe things are closer than they appear (or a Republican counting on a wave).     It’s possible, but pollsters have their reputations on the line, and tend to be very careful about their methods and assumptions.   Historically such systemic errors are rare (though it happened to Gallup in 1998), though clearly if that happened this cycle, it could be an historic and dramatic development — and it’s possible.

3.  A Late Democratic Surge.   Some Democrats believe that people on the left have been late to focus on the election, in part because the news has been so bad.   But stories of tea party excesses along with efforts to engage the youth, including a major rally being held by Jon Stewart and Stephen Colbert, will cause an uptick in Democratic enthusiasm and support that will create a mini-wave for the Democrats.

This is counter-intuitive, but conceivable.  If the GOP peaked and Democrats play the last week well, given the closeness of the races a slight improvement to the Democrats might save a significant number of seats.    However, working against them is the fact that most late deciders are independents, and they have been tending Republican in a big way.   Still, so many races are extremely close, and the Republicans have provided fodder for the Democrats to drum up interest — the tea party rhetoric, a Kentucky campaign volunteer smashing the head of an activist from the left, extreme comments by GOP candidates — that the possibility cannot be dismissed.

4.  The Rules of the Game Have Changed. Finally, some Democrats look at the 2008 election and say that although enthusiasm is down, the Obama machine can still turn out a disproportionate number of youth, something pollsters don’t expect in off term elections.  Anyone on facebook has no doubt seen the “commit to vote” messages and certainly the Democrats are working hard to inspire younger voters.   The idea that the dynamics of elections have fundamentally changed due to early voting, social networking, and youth engagement is unpersuasive.  It reminds me of the arguments of late 1999 by dot com traders saying that we had a “new economy” and concern for stock valuations and fears of a bubble were off base.

Perhaps the one way “new rules” could alter the game is if the money being spent on Latino and Black voter outreach significantly improves turnout in those populations.   That is conceivable; some who voted for the first time in 2008 may decide to stick with the habit.    There is a black President, and the immigration issue has evoked often vicious tea party rhetoric against Latinos.   However, there is no way to know if that will translate into anything significant until after the fact.

The biggest problem the Democrats face in these scenarios is that campaign dynamics work both ways.    Each side has similarly plausible scenarios that things will “break their way,” and each is imbued with a dose of wishful thinking.  Still, the Republicans should have some concern.   They are celebrating early, perhaps losing some focus, and thinking about leadership positions and potential internal rivalries after they take control of the House.   That isn’t as harmful in politics as it is in sports, but in very close races losing focus can make a difference.   And that’s probably the greatest hope the Democrats have — the Republicans are assuming victory, and in the last week of the campaign that might make them a bit more mistake prone.

So yes, the Democrats can hold the House.  It would cause great consternation and even anger among many Republicans, and be an embarrassment for most pundits and prognosticators.   Yet it isn’t very likely.  It’s like being behind 35-3 early in the second half, the odds are strongly against a comeback.   Just ask the 1991-92 Houston Oilers and Buffalo Bills.

UPDATE: This article by Nate Silver discusses the fact that “robopolls” tend to favor Republicans, but notes that it’s hard to know precisely what to make of that in terms of figuring out this election cycle.