gohansrage, on 13 August 2012 - 12:52 PM, said:
I didn't call anybody a name. I said the post contained idiocy. It did.
Believing in any media bias other than "trying to get people to consume media more often" is also idiotic.
Really? So you don't think that media outlets like, say, the NY Times, Washington Post, NBC, CBS, ABC, MSNBC, AP and Reuters skew liberal? Or that the NY Post and Fox skew conservative?
TrueBluePhD, on 13 August 2012 - 01:01 PM, said:
I think "much less" credible is underselling the value of the state polls. They're certainly less frequent, but the frequency of national polls is a weakness as well, if viewed the wrong way. Daily/weekly national polls will tend to capture short-term effects which are not useful when trying to evaluate what will happen in November. For example it's likely that the VP selection will give Romney a bump in the national polls, but it's highly unlikely that bump lasts over time--yet inevitably there will be people who read way too much into it (same thing with the upcoming convention bumps). This stuff has to be looked at over time, which is why even though state polls are less frequent they are still useful. Yes they tend to have a larger margin of error than national polls, but they're also more directly applicable to predicting the winner, so there's still a lot of value there--but as with all of this stuff, caution is needed (it's way easier to over-read a poll than to under-read it).
Having said that, anybody who thinks this election is decided now, or is going to be a big win one way or the other is fooling themselves. If this were going to be a landslide election, we'd know it by now. What to really look for will be the revised second-quarter GDP numbers coming out at the end of the month, and polling a little after the conventions (once the temporary bumps have normalized).
I generally agree with most of this, but I think "much less credible" is applicable to most of the state polls (and for that matter to most of the national polls other than Gallup and Rasmussen). Bottom line is that the polling methodologies are fundamentally flawed -- most of them sample registered voters as opposed to likely voters, and most also oversample registered Democrats relative to their percentage of the electorate. In addition, there are other, more technical methodology flaws that Gallup and Rasmussen address reasonably but the others do not.
If you're interested, a commentator named Jay Cost writes about polls fairly often and IMHO takes a pretty analytical and non-partisan approach. His stuff is linked on RCP frequently.
waldo, on 13 August 2012 - 12:49 PM, said:
My opinion is not based on what the conservative commentators think, nor the liberal ones for that matter. I have a friend who runs a major polling operation for the democrats . Once a month we talk and have an honest discussion of what the tabs say in his most recent polls. He is not optomistic in any sense of the word. We will all know more after Labor day/ when things start to gell . Historically,at this pt. the President should be above 48% approval wise and up by 5-6 , registered voters,in the swing states. He is not. We both wonder what happend to the VP. You cannot let him debate Ryan. Right now there are far more scenerios where they lose, than scenerios where they win. All contingent on neither partynot making a fatal error. We both agree that if a fatal error is made in a debate the President will make it . His most recent quote of..... equal opportunity . equal outcome , or something like that is far to close to Marx's famous quote. If i was Ryan I would ask him if he was running as a communist? just for giggles
Interesting stuff -- thanks.
As for Biden, I think what happened is that as VP, people started paying more attention to the nonsense that pours out of his mouth on a regular basis. I continue to believe that he is at least as much of an airhead as Palin was.