Replies

Here is how you test: last two polls by Ras and (I forget other group) had Mandell tied and/or up one. This one has him down eight.

So, that tells me the real number is Romney by about 3.

To answer your question about contrast---and remember, Gravitas had Romney up one in OH 48 hours ago---you need to look at their QUESTIONS. In the previous OH PPP poll, they asked: (now, remember this is OHIO):

"What is your opinion of Elizabeth Warren?" (!!)

"Who is more responsible for killing Osama bin Laden, Barack Obama or Mitt Romney?"

"What is your opinion or Julian Castro?"

"What is your opinion of Martin O'Malley?"

In case you don't recognize these people, they were speakers at the DEMOCRAT convention and only DEMOCRATS would even know who they are. I bet 20% of all Ohioans don't know who Elizabeth Warren is. Yet the PPP Poll did not ask about Mandell that time.

I think these people are as reliable as the old Zobgy internet polls.

Hey LS -

Thanks, but my question isn't about the inaccurate poll, it is specifically about reconciling the below italicized sentence with your early ballot request numbers in Ohio. It's out of mere curiosity, I'm not worried about Ohio.

If both things are true - your numbers, and the numbers referred to below, then we are looking at a wild statistical aberration, which can not be explained by asking leading questions about Liawatha (I'm from MA), nor can it be explained even if they were using as much as a D+10 Model. The question is about early ballots cast D76 to R24 vs your reports of early ballot request partisan breakdown.

"The key finding on this poll may be how the early voters are breaking out. 19% of people say they've already cast their ballots and they report having voted for Obama by a 76-24 margin. "

See what I'm getting at? Nothing to do with the questions asked during the polling, and too great a statistical anomaly vs your numbers re: partisan breakdown of early ballot requests. That's what I'm trying to reconcile.

If you look at internals they have Obama getting 11% of Repubs. And Romney only getting 5% of Dems.

However, in the last 3 presidential elections (even in 2008) the Dem never pulled more from GOP base. It was always GOP winning over Dems. (according to Jay Cost)

Yet after 4 years of Obamanomics we are to believe Obama will beat Romney 2-1 in pulling from base?

IMO, that is the skew.

If base vote were same (90-10), which most every national poll shows, Romney would be tied.