Assessing Obama’s convention bounce

Inevitably after a convention the media will try to assess how much “bounce” in approval rating the nominee received due to the extended media coverage. This is an ideal time to assess Obama’s bounce, since his approval ratings have not yet been overshadowed by the Republican coverage.  To make it easy, I’ll  use RealClear Politics average of the polling ratings for both McCain and Obama beginning August 23, the start of the Democratic Convention, through Sept. 4. (Faithful readers will recall that I think there are problems with how RealClearPolitics calculates the polling average, but for now we can set those concerns aside).  I’ll define the “bounce” as the change in the difference in ratings between the two candidates due to the convention coverage.

On August 23, the start of the Democratic convention, Obama led McCain by a scant 1.6 percentage points.   By the end of the convention that had almost doubled to 3.9% (August 29).   On August 31, it dropped down to 3.4%, and on September 1 it was back up to 4.5%  It peaked at 6.4% on Sept. 2 and has since dropped a bit: today it is at 5.8%.

Clearly, then, Obama benefited in the short term, as reflected in polling averages, from the Democratic Convention. However, since RealClearPolitics uses a rolling average to calculate its polling average, it’s hard to define a precise cutoff point to measure the bounce.  But at the most we can say he gained a bit less than 5% in the polls (combination of his bump up and McCain’s drop) as a result of the convention.

Should he be pleased?  Interestingly, (or perhaps not),  political scientist have actually studied the size of the typical convention bump in the modern era.  James Campbell finds that  in the post World War II era (he was writing in 1992), the typical bounce is between 5-7%.  Larry Sabato, looking at the period from 1960-2004, finds a slightly larger average bounce for Democrats – about 7.3%.  But his figures are inflated by including 1992, where Democratic candidate Bill Clinton received a whopping 28% bump. Clinton’s bounce was inflated by the simultaneous announcement by Ross Perot that he was dropping out of the race (he subsequently reentered).  In 2004, in contrast, Kerry received almost no bounce coming out of the convention.  In both 1996 and 2000 the Democratic candidate received about an 8% bounce.

So, historically speaking, Obama’s “bounce” is smaller than average, although not exceedingly so.   Should he be worried?  Not based on the smaller than average bounce.  Convention bumps are almost always transitory, fading over time as the underlying fundamentals driving the electoral race reassert themselves.  I expect that to happen here as well, particularly after the Republicans get their own bounce coming off Palin’s speech followed by McCain’s tonight.

I’ll be on later tonight blogging McCain’s speech. If I get a chance, I’ll also respond ton some your comments re: Palin, Biden and Obama, which continue to come into my inbox.

Finally, I need to apologize to some of you who are getting these messages with a substantial delay built in – Microsoft apparently interprets this email as spam (a judgment that may be shared by you!), and delays sending it on.  I’ll work on correcting that glitch.