OH-Gravis Marketing: Obama+1

[Tender Branson]

Obama 45, Romney 44

http://www.scribd.com/doc/106848007/Report-Ohio-Sep-21-22-2012-1

[MorningInAmerica]

Party ID of likely voters is 41%Democrat/31% Republican/28% Independent (D+10).

[Okay, maybe Mike Johnson is a competent parliamentarian.]

These are the Smith is leading Casey folks, right?

[sobo]

Johnson is over 10% in the three way race. There is something seriously wrong with this poll.

[Earthling]

Don't blame the poll, blame the pollster.

[Starbucks Union Thug HokeyPuck]

Johnson is over 10% in the three way race. There is something seriously wrong with this poll.

but... but... but.... it's D+10!

[Craigo]

Remember folks, party ID exists to give people who don't know what they're talking about something to say.

[tmthforu94]

The party ID numbers are way off here.

Once you adjust them to what election day turnout will be, Romney's up by around 12.

[Tender Branson]

Nobody knows who this pollster is and who the people behind are.

[J. J.]

Nobody knows who this pollster is and who the people behind are.

Apparently, neither do the people behind it.

This firm sounds like it does marketing surveys; I've always question the transition between marketing and political races.

[dspNY]

This poll is trash...11% undecided in Ohio, where they've been hit by over $100 million in television advertising?

The Ohio Poll/University of Cincinnati is the gold standard, and Obama led by 5 (51-46 with 1% to third parties and 3% undecided)

[Craigo]

Nobody knows who this pollster is and who the people behind are.

Apparently, neither do the people behind it.

This firm sounds like it does marketing surveys; I've always question the transition between marketing and political races.

I get to agree with you for once! High five!

Nobody (and I mean nobody) should hire me to do market research for them. It goes the other way too.

[Landslide Lyndon]

As if one ARG wasn't enough...

[Niemeyerite]

The party ID numbers are way off here.

Once you adjust them to what election day turnout will be, Romney's up by around 12.

OK, bro. Romney's leading. Unskewed polls, now!!

[mondale84]

Nobody knows who this pollster is and who the people behind are.

We know that they are right-wing trolls, and that's all we need to know.

Junk company, junk polls.

[J. J.]

Nobody knows who this pollster is and who the people behind are.

We know that they are right-wing trolls, and that's all we need to know.

Junk company, junk polls.

They are not necessarily a right wing company, but they seem to be a new one.   The president was in mortgage marketing.  http://www.linkedin.com/in/douglaskaplan

[ajb]

Again, they're the people who have Casey losing PA-Sen by 19 points...

[Craigo]

Again, they're the people who have Casey losing PA-Sen by 19 points...

I read that release - not only did they reverse their numbers (234 Casey, 195 Smith, 499 total), they reported the second statistic as 24% instead of the correct 39%. They then displayed a pie chart with Casey's share reported at 28%.

How they managed to make all these mistakes, I have no clue.

Edit: Actually, that's not true. Those mistakes (and the fact that they managed to get a highly improbable 500 likely-voter respondents from 12pm to 4 pm on a weekday afternoon) are red flags for fraud.

Edit 2: 75% of the PA respondents were in their oldest age category - which went undefined in the release.

[J. J.]

Again, they're the people who have Casey losing PA-Sen by 19 points...

I read that release - not only did they reverse their numbers (234 Casey, 195 Smith, 499 total), they reported the second statistic as 24% instead of the correct 39%. They then displayed a pie chart with Casey's share reported at 28%.

How they managed to make all these mistakes, I have no clue.

Edit: Actually, that's not true. Those mistakes (and the fact that they managed to get a highly improbable 500 likely-voter respondents from 12pm to 4 pm on a weekday afternoon) are red flags for fraud.

Edit 2: 75% of the PA respondents were in their oldest age category - which went undefined in the release.

They are not "red flags for fraud" or an indication of bias.  They are indications of being a bad pollster.

[ajb]

Again, they're the people who have Casey losing PA-Sen by 19 points...

I read that release - not only did they reverse their numbers (234 Casey, 195 Smith, 499 total), they reported the second statistic as 24% instead of the correct 39%. They then displayed a pie chart with Casey's share reported at 28%.

How they managed to make all these mistakes, I have no clue.

Edit: Actually, that's not true. Those mistakes (and the fact that they managed to get a highly improbable 500 likely-voter respondents from 12pm to 4 pm on a weekday afternoon) are red flags for fraud.

Edit 2: 75% of the PA respondents were in their oldest age category - which went undefined in the release.

They are not "red flags for fraud" or an indication of bias.  They are indications of being a bad pollster.

Anyone who thinks that 195/499=0.24 is not a "bad pollster." They're not a pollster at all. And anyone that can come up with a press release suggesting a result that's thirty points different from virtually everyone else, without triple-checking everything they've done, is either incompetent to the point of incredulity, or simply making this stuff up.

[J. J.]

Again, they're the people who have Casey losing PA-Sen by 19 points...

I read that release - not only did they reverse their numbers (234 Casey, 195 Smith, 499 total), they reported the second statistic as 24% instead of the correct 39%. They then displayed a pie chart with Casey's share reported at 28%.

How they managed to make all these mistakes, I have no clue.

Edit: Actually, that's not true. Those mistakes (and the fact that they managed to get a highly improbable 500 likely-voter respondents from 12pm to 4 pm on a weekday afternoon) are red flags for fraud.

Edit 2: 75% of the PA respondents were in their oldest age category - which went undefined in the release.

They are not "red flags for fraud" or an indication of bias.  They are indications of being a bad pollster.

Anyone who thinks that 195/499=0.24 is not a "bad pollster." They're not a pollster at all. And anyone that can come up with a press release suggesting a result that's thirty points different from virtually everyone else, without triple-checking everything they've done, is either incompetent to the point of incredulity, or simply making this stuff up.

As I've indicated, the pollsters background was in marketing.  "Bad" and "fraud" are two different things.

[MaxQue]

Again, they're the people who have Casey losing PA-Sen by 19 points...

I read that release - not only did they reverse their numbers (234 Casey, 195 Smith, 499 total), they reported the second statistic as 24% instead of the correct 39%. They then displayed a pie chart with Casey's share reported at 28%.

How they managed to make all these mistakes, I have no clue.

Edit: Actually, that's not true. Those mistakes (and the fact that they managed to get a highly improbable 500 likely-voter respondents from 12pm to 4 pm on a weekday afternoon) are red flags for fraud.

Edit 2: 75% of the PA respondents were in their oldest age category - which went undefined in the release.

Well, the Edit 1 explains the Edit 2.