Mason Dixon Battle Ground Polls/Internals are important....

Mason Dixon ^

[PubliusEXMachina]

Here are presidential race results from battleground states polled by Mason-Dixon Polling and Research for MSNBC, Knight Ridder and in Nevada, the Las Vegas Review Journal.

[PubliusEXMachina]

This may have been posted eariler, but you should look at the internals that are graphed here. As usual the only worriesom one is Ohio. I hope the Pres gets out there and stumps with Rudy, McCain and Ahhnold...by the way what are the track records of Mason Dixon?

[Petronski]

...by the way what are the track records of Mason Dixon?

Excellent. Sterling.

[conservativepoet]

I've heard Mason-Dixon is the best on state polling. I also think Bush is up by 3-5% in Ohio.

[So Cal Rocket]

best in the business

[The G Man]

They were the most accurate polling firm in 2002, missing only 1 race out of 23 and being within something like 1.8% of the actual results. Compare to Zogby who missed about 1/3 of the 2002 races.

[Nonstatist]

Ohio internals are worrisome. The other states look much more assuring.

[finnman69]

Mason Dixon ACED 2002 22 out of 23 races right


http://www.ncpp.org/2002SenGovPoll/2002ElectionPolls.html

[KC_Conspirator]

Are you kidding me? Zogby missed about a third of the races in 2002? Thats horrible. I still have to stick with his last 2 presidential polls as beung the best.

Any info on Zogby's 2002 predictions?

[marblehead17]

ping

[TigerDSL]

Zogby's 2000 final presidential poll was NOT the best. The Fox/OD poll was the best. See link below:

http://members.cox.net/fweil/FinalPolls2000.html

[kcbob]

I used to think that in order for me to be totally comfortable with the polls I need to see W's numbers above 50% and outside of the MOE and all polls have basically the same results. Now I think that we have to overcome the MOF - Margin of Fraud. We need about 3-5% above Kerry to overcome the MOF.

[The G Man]

Here ya go:

Zogby worst 2002 pollster

Polling Organization	Number Of Polls	Average Error On Candidate	Races With Wrong Winner
Mason-Dixon	23	1.8%	1
Zogby International	17	2.5%	5
Research 2000	13	2.1%	2
Gallup	7	1.4%	1
Quinnipiac College	4	2.0%	0
All Others	95	2.7%	12
TOTAL	159	2.4%	21

[Paul8148]

So Gallop was the best in 02. I do think Zogby was the only one to have Sunua up in NH and got that right, but I think he was his posters and kept them privite until after the election so maybe he brag to look good.

[Hermann the Cherusker]

Does Ohio have partisan registration? I can't find any data on it. How realisitic is M-D's 38-39-26 D-R-I breakdown for Ohio? I know neighboring PA is 48-42-10 in D-R-I.

[The G Man]

Yeah, but look at Mason Dixon ... 22/23 correct.

[BlueNgold]

The OHIO numbers are mixed:
The best number - 45% somewhat or very conservative with a TON of moderates.

[KC_Conspirator]

Thank you, that is very explanatory. Zogby got 29% of his predictions wrong in 2002. That is a BAD number. Its twice the amount (and percentage) of the polls that got 14-15% wrong.

[HostileTerritory]

They do, but upwards of 60% are independent. I think M-Ds numbers match exit polling from 2000. The 38/39 may have been reversed but that means nothing.

[HostileTerritory]

How did they do in 2000?

No offense, but the reason pollsters were off in 2002 was because they underestimated Republican turn-out compared to 'RAT numbers. Do we know if their model is flexible to account for varying levels of turnout from year to year, or are they citing 2002 turnout percentages? Because I think that Republicans are an energized as ever, but the 'RATs are also much more energized than 2 years ago.

[kingattax]

people.....please remember...the gay marriage issue is on

the ohio ballot....that will assure a big conservative

turnout......ITS THE TURNOUT,STUPID.....NOT THE POLLS.

GW will win Ohio....take that to the bank