My colleague at OTB, Steven Taylor, considers the impact of the debates in the last six presidential elections:
Now, on the one hand, the number of observations is small (6 elections and 16 debates), so no, we cannot draw ironclad inferences from the figures. On the other hand, as Holbrook notes, “Across all 16 presidential debates the average absolute change in candidate support was just less than 1 percentage point.”
I think there are a couple of problems in comparing last night’s debate with those from 1988 to 2008. First, in only three of those, 1992, 1996 and 2004 [ed. corrected], was an incumbent running for re-election. Second, the president’s re-election campaign appears to be unique in that he’s de-emphasizing his record in favor of emphasizing that he’s a better candidate than his opponent. In that respect it more closely resembles his primary campaign against Hillary Clinton or his successful presidential run against John McCain.
The prevailing wisdom is that re-election campaigns are always referenda on the incumbent. Assuming otherwise is going against the evidence of history. The conventional way of stating this going along the lines of “No president since FDR has been re-elected with an unemployment rate above 8%” or something along those lines.
The president is, in effect, running as a challenger rather than as an incumbent. Will he be allowed to do so? If not, performances in the debates by Mitt Romney that suggest that he’s a viable alternative to Barack Obama may provide a lot more than “less than 1 percentage point”. In an election as close as I suspect that this one will be that could be significant.