Thursday, October 4, 2012

Judging the Debate

Almost universal claims on the cable channels that Obama lost--but, as I've said from beginning, the instant polls usually refute pundits' claims in these cases.  But not tonight. See my Twitter feed @GregMitch for my many comments during and after debate.  As always: Charles P. Pierce.

Update #6  Very interesting: Daily Rasmussen poll finds NO gain for Romney--and some of polling done post-debate.  See my item here.

Update #5   CNN has now posted its full survey and there are some interesting tidbits.

--Gender gap not as big as expected.  Even 59% of women gave the nod to Romney.

--CNN breaks down some of its findings into the usual categories of age, race, region.  But oddly, it reports "not available" or N/A for all regions outside the South, all ages under 50, and all non-whites.  Just a glitch?  It's hard to believe that this represents its actual sample but we await an explanation.  Very odd because usually N/A means numbers so small for a group that they are "statistically insignificant."  CNN?   One explanation:  When there's less than 100 in a sample they list as N/A.  Doesn't mean they were not included, but still suggests over-sampling older and Southern folks.  Note:  TPM now relates, "CNN provided us with the internals of the poll, and the demographics of the poll respondents are very much line with normal standards for randomized sampling."

--Romney's unfavorable rating actually went UP one point.  And off that his favorable rating in this sample was 54% even before the debate--much higher than other polls have shown.

Update #4, Wednesday  Nate Silver reviews the polls and finds one I'd missed:  a Google experiment that found Obama holding his own during the debate but losing 2-1 after (one wonders if TV pundits giving it to Romney swayed some). 

Straining to find some good results for Obama, Priorities USA points to a focus group it sponsored  in Aurora, Colo.  But its group were all "weak Democrats" and indies, not Repubs.

People like Steve Kornacki of Salon who claim Obama was "lousy" in the 2008 debates and still won are dead wrong. Obama was not terrific but was generally effective in 2008 vs. McCain--and the same polls now giving this year's first debate to Romney gave Obama solid wins in all three debates that year.

Good roundup of views from Greg Sargent  today. 

Update #3  CNN has now clarified its sample in its instant poll after its own John King noted on the air that it held more GOPers than average.   The sample of debate-watchers in the poll was 37% Democratic and 33% Republican.  CNN spokesman:  "That indicates that the sample of debate watchers is about four points more Democratic and about eight points more Republican than an average CNN poll of all Americans, for a small advantage for the Republicans in the sample of debate-watchers."

But here's a very interesting number that may get little play:  The same poll that found easy Romney win also found that Obama's favorable rating in this group remained exactly the same at 49%.   Romney's favorable mark only went up 2%--and his unfavorable rating actually went UP one point.   And only 46% found him more likeable than Obama.  So perhaps few votes will be changed. 

Update #2  Now CBS poll of 500 undecided voters finds Romney with 2-1 edge on winning debate, 46% to 22%.  Opinion of Romney, naturally, rose among many.   Not sure of land line/cell breakdown. 

Update #1  CNN/ORC scientific poll of 430 registered voters--who watched the debate--just out and with staggering edge I have not seen before, 67% for Romney and 25% for Obama.    (Now they will go to their undecided voter "focus group.")   CNN says a number of voters in poll said they're now more likely to vote for Romney--35%,  with 18% saying they'd go for Obama, with 47% no change.

More:  "Who Seemed to Be Stronger Leader? Romney 58% - Obama 37%."  Who better to handle economy and taxes and health care--smaller margins.

John King admits this sample a few points more Republican than average so Romney's edge higher than reality probably.  David Gergen also calls sample "skewed."  Carville jumps on that.  Now I've seen that 5/6 of sample reached by land line, not cell, so there you go.

CNBC poll (not sure of source) had it 51%-38% for Romney.   So, in rare case, TV pundits may have it (mainly) right.  Even gender gap did not save Obama, at least tonight.  Andrew Sullivan charging that Obama "choked" and may have lost election tonight.  Well, we'll see what Gallup and others show in about four days.

No comments: