Romney supporters must be disappointed at the lack of a repeat performance. From tax plans to Romney being called out on his Libya falsehood, including by the live audience (a must-watch), President Obama appeared assertive and in charge. Measures of who won:
CBS undecided voters: Obama 37%, Romney 30%, Tie 33%.
PPP Colorado voters: Obama 48%, Romney 44%, 58%-36% among independents.
CNN debate viewers: Obama 46%, Romney 39%. R+8 sample.
Andrew Sullivan: calmer now that Obama took charge. Sort of a human InTrade.
Ipsos registered voters: Obama 48%, Romney 33%.
For any debate it seems that each candidate is near-guaranteed that 20% will say he won (see Gallup numbers on debate #1). Keep that in mind, i.e. subtract 20 from each candidate to get a better sense of what these numbers mean.
Note that I am leaving out nonrandom samples such as the Frank Luntz hand-selected panel on MSNBC because we do not know whether they were selected to lean in one direction or the other. For the same reason, Web surveys are inadmissible as evidence.
The number of viewers (Nielsen) was down only 2% from the first debate: 65.6 million viewers this time, 67.2 million last time.
It is hard to know how this will affect coming polls. Two scenarios:
- The 5-6 point slide that followed debate #1 could be entirely explained by voter mood swings. Even if Romney supporters increased their probability of voting by just 5%, and vice versa for Obama supporters, that could have accounted for the shift. If that’s the case, it will become apparent in a big way as moods swing back.
- Alternately, Romney (and the media aftermath) really did persuade voters in that first debate. In this case we should expect a smaller change.
In any case, presumably the story line as told by the press will change again. Also – note that state polling slowed down over the last few days. It might be hard to distinguish Debate 2-induced change from pent-up changes that have gone unmeasured.