Wednesday, October 31, 2012

A Partisan Advantage for Obama?

As I pointed out at the end of August, I see political and economic context of the election as favoring Mitt Romney, though not overwhelmingly so.  In fact, I think that Romney's gains after the first debate partly reflected a dissatisfied electorate responding to a campaign event that pushed them in the "right" direction.   But one thing I find interesting is that despite a reversal of fortunes in the last few weeks, there seems to be a clear floor below which Obama's support will not fall, and Mitt Romney has not been able to do any better than draw even or periodically take a very narrow lead.

One thing that may be benefiting Obama could have little to do with the standard indicators of national conditions and may have even less to do with either of the candidates or their campaigns. Simply put, I think a case can be made that the Republican brand name is acting as a drag on Mitt Romney's candidacy.  This idea has gotten a little bit of attention from some journalists, and the data I present below largely provide additional support.

One way of assessing the relative value of party for each of the candidates is by looking at rates of party affiliation in the electorate.  The figure posted below was generated using the pollster.com dashboard and includes the results of hundreds of polls.   These data show that the Democrats have held an affiliation advantage throughout the 2012 campaign, one that has ebbed and flowed a little bit and now stands at approximately six percentage points.  Of course it is possible that slight differences in loyalty (Republicans have been more loyal than Democrats in SOME elections) and turnout might mitigate the the Democratic advantage in affiliation.  Still, I can't imagine this difference has no effect on the current contest.




But party identification isn't the whole story.  Instead, there appears to be a broader problem having to do with the general image of the Republican Party, especially when compared to the image of the Democratic Party.

The figure posted below summarizes the "favorability" of both parties, using results from several polls that have asked respondents to rate their feelings toward the parties as either favorable or unfavorable.  I don't have nearly the same number of data points here as for the figure on party identification, but the pattern is very clear: throughout this campaign period the Democratic Party has been viewed more positively than the Republican Party.  In fact, there is not a single poll in this series in which the Republican party registered a net positive rating, and not a single case in which the net Republican rating was higher than the net Democrat rating.  The average net rating for the Republican Party in this series is -13, whereas the average for the Democratic Party is +.3.  To be sure, the net rating for the Democratic Party is sometimes in the negative, and the gap toward the end of the series is not as great as it was in the wake of the Democratic convention, but it is clear that the Democrats hold an advantage on this front.

Data taken from pollingreport.com and pollster.com.  The dots represent individual polls, and the lines are the smoothed lowess trends in the series.

Of course, there is little doubt that these two things--Democratic advantages in party ID and party image--are connected to each other.  Still the relative stability of the difference, and the approximately 10 point Democratic advantage in the most recent polls, must be working in Obama's favor.  Is this what's keeping a president with tepid approval numbers and a still-sluggish economy afloat?  If so, would there be any payoff for Obama to run not just against Mitt Romney but also against the Republican Party in the closing days of the campaign?

Just to be clear, I'm not arguing that the public is wild about the Democratic party; just that the Democratic Party is viewed much more positively (or less negatively, if you will) than the Republican Party.

Monday, October 29, 2012

Cool Tool

I just saw a link to this really neat tool on The Monkey Cage.  Here's what the Milwaukee area looked like in 2008:


Not a really surprising picture, but it does show the stark contrast between the city and most of the suburbs.  

Saturday, October 27, 2012

Updated Electoral College Forecast


As promised in my last post on the September Model I'm be posting a rolling Electoral College forecast (in the panel to the right), based on past election results and October national and state-level trial-heat polls. As with the September model, these state-level forecasts were developed with Jay DeSart. Clicking on map to the right will take you to Jay's web page, where more details are available. As new polls come in the forecast will change, so please check in on a regular basis.

The results for today are:

Electoral vote: Obama 281, McCain Romney 257
National popular vote: Obama 51.5%, Romney 48.5%

Both the Electoral College and popular vote projections are down a bit from those generated by the September Model, reflecting a trend found in most other poll-based models.  The biggest state changes are Colorado, Florida, and Virginia moving from Obama to Romney.  These are currently the most likely pick-ups for Obama, with all three showing Romney with less than a 56% chance of winning them.  Romney is going to have a harder time picking up any states currently in the Obama column, with Ohio the most likely but currently showing Obama with a 68% chance of winning.


Thursday, October 4, 2012

September Polls and November Outcomes

Jay DeSart and I have done some work over the years on using September state-wide trial-heat polls to predict presidential election outcomes in the states. We pool state-level data from 1992-2008 and use the Democratic candidate's share of the two-party vote in state-level September polls (averaged across publicly available polls), a lagged (four elections) vote variable, and the Democratic candidate's share of the vote in national trial-heat polls in the last half of September to predict the election outcomes. While the other variables are important to our model, the September polling average is the strongest predictor.  Jay has been good enough to update our model and provide forecasts for the 2012 election.  Here's what it looks like:

National Two-Party Vote : Obama 53.1%
  Electoral College: Obama 332, Romney 206
Jay has a bunch more details on his forecasting page.

Even if you don't use a complicated statistical model, September polls are a useful guide to who will win in the states.  A couple of interesting tidbits:
  • In our data set, from 1992-2008, there were 164 states in which one of the candidates had a lead outside the margin of error (based on the average statewide sample size) in the state polling average.  The leader went on to win in 160 of those cases (98%).
  • If you throw caution to the wind (ignore the margin of error) and make predictions just based on which candidate is ahead in the September poll average, the polls predict correctly in 220 of the 250 cases (87.5%).
We also have a rolling state-level forecasting model based on October polls.  I'll post that once we get enough state-level poll results.

Really?

This really stinks.

Monday, October 1, 2012

Debate Expectations

With the first debate scheduled for just a few days from now, media attention has begun to focus on what to expect, or not expect, from this year's head-to-head match ups.  Fortunately, it seems that most commentators are on the same page: don't expect much to change because of the debates.  In the spirit of piling on the "debates don't matter much" argument, I'm reposting and updating some of the analysis I presented four years ago on this blog's predecessor.

So the basic idea is that if debates "matter," levels of candidate support should be appreciably different following the debate than they were prior to the debate, presumably shifting in favor of the candidate seen as having "won" the encounter.  This is fairly easy to measure, and I present data in the table below that gives us a sense the magnitude of debate bumps from the last several elections. Here, I compare the polling average from six days prior to the debate to the day of the debate with the average from the seven days following the debate, for all debates from 1988-2008.

These data suggest that the norm is for very little swing in candidate support following debates. Across all 16 presidential debates the average absolute change in candidate support was just less than 1 percentage point. There are a few notable exceptions, of course. Two that stand out are the second debate in 1992, following which George H.W. Bush lost 2 points, and first debate of 2004, after which George W. bush lost 2.26 points. Other debates with above average (but still small) vote shifts are the first debate in 1996, the second debates in 1988 and 2000, and the first debate of 2008. Each of these debates has its own story, and I'm sure we can all think of anecdotes to explain the bumps and wiggles.   Although the analysis is terribly outdated by now, my debate model from Do Campaigns Matter? came to the profound conclusion that the candidate viewed as having won the debate generally gets a small bump. I told you it was profound.

Focusing on single debate bumps may be obscuring a more general, cumulative effect of debates. The last column in the table shows the change in candidate support from one week prior to the first debate to one week after the final debate. Here we see that the debate period generated a 2.42 point bump for George H. W. Bush in 1988; cost Al Gore 3.52 points in 2000; cost George W. Bush almost 2 points in 2000; and John McCain lost almost two points in 2008. Of these debate period swings, the 2000 debates stand out as the most important, especially in the context of the closeness of the election. Part of the explanation for Gore's swoon during the debate period is perhaps related to his performance but another important factor was the media meme that emerged as a result of the debates, including open discussions of whether or not Al Gore was a "serial exaggerator" (see Jamieson and Waldman).  This brings up an important point: it is probably not a good idea to attribute all of the change in candidate support during the debate period to the debates, since there are many other important events, or perhaps even natural drifts in candidate support occurring during these several week periods.

So what does this mean for the three presidential debates this year? I suspect more of the same, which means that neither candidate will probably benefit much from them.  The key factor will be whether one of the candidates can make small gains from each debate that, together, add up to something like a two or three point gain during the debate period.  I'm doubtful, but it could happen.