Who "Wins" Wednesday's Debate?

The pundits and the public will, of course, reach some kind of consensus over whether Obama or Romney wins tomorrow night's prez debate.  But why wait?  The latest Quinnipiac poll has folks predicting an Obama debate victory -- by about a 2-to-1 margin.

Here's the lede:

An 18 point lead among women puts President Barack Obama ahead of Gov. Mitt Romney 49 - 45 percent among likely voters nationwide, and voters expect 54 - 28 percent that the president will win the debates, according to a Quinnipiac University national poll released today. 

That pretty much says it all.

Except, of course, it's probably wrong.

Usually the challenger "wins" the first presidential debate with an incumbent (think John Kerry against George W. Bush not so very long ago).  Why?  Because the incumbent is on the same stage, looks presidential, and assuming no goof comes off ahead merely by not seeming as kooky as some people thought.

That's Romney's mission.  Not to look kooky or scary, to seem, well, as Ronald Reagan did so well in 1980 -- to seem reasonable.

Can people predict the winner?  Are they right?

No.

In part, the expectation comes from people choosing their preferred candidate to win.  It's called wishful thinking.  There's a long body of social science research to back it up.  But that doesn't explain the 2-to-1 margin, much greater than Obama's poll advantage.  No, I think this is a function too of Romney's faltering campaign, of stumbling from gaffe to gaffe (47 percent), from a smooth Democratic convention compared to Clint Eastwood mumbling at an empty chair.

So I suspect the public has it wrong.  Romney will "win" Wednesday night -- for what little that's actually worth -- but unless Obama makes a huge blunder it won't really matter in the numbers.  Yes, they'll tighten a bit in Romney's favor, but the math for him, in a state-by-state comparison, needs to significantly improve for him to win the electoral count.

Prez Debates and Political Learning

Let the pundits argue about this week's presidential debate and whether it persuades voters.  Let's ask a different question -- do people learn from such debates?

As always, there's the short answer and the long answer.

Short Answer:  It depends.
Long Answer:  It depends, with a whole bunch of caveats and academic nuances and all the stuff that gets in the way of a simple answer (unless, of course, you do talk radio or cable news, where everything is simplified to the point of meaninglessness).

Stay with me here.  It's worth it.  

One study, for example, examined five presidential election years (1976 to 2000) using ANES data, which the authors admit "is not optimal for conducting this sort of research."  But this is key: they found learning about incumbents to be minimal, but learning about challengers to be more significant.  That fits this upcoming debate between Barack Obama and his challenger, Mitt Romney. 

The former Massachusetts governor, as pundits have noted for several weeks, has so much to gain from being on the same stage as an incumbent president.  This study suggests people will also learn more about Romney.  What they learn, that's of course of import, but the research suggests Romney gains a great deal from Wednesday night's debate.

Tomorrow (or maybe later today), what else research tells us about this week's debate.