New DailyKos/Research 2000 tracking poll

[Tender Branson]

Friday - 12 September, 2008:

Obama - 47% (nc)
McCain - 46% (+1)

To a Moderator: Sticky it please.

To JFern: Remove "New" from the heading please.

[Tender Branson]

Saturday - 13 September, 2008:

McCain - 47% (+1)
Obama - 47% (nc)

[Tender Branson]

Sunday - 14 September, 2008:

Obama - 47% (nc)
McCain - 45% (-2)

Hmm. Obama gets only 35% of Whites and trails by 5 among Independents, while getting 8% less support from DEMs than McCain is getting from Republicans. And he still leads by 2 ?

Their party and race breakdown is certainly crazy ... 35%D, 30%I, 26% R, 9% Others (?)

13% African-American ? 13% Latino ??

[Tender Branson]

What's with stickying this now ?

[Tender Branson]

Friday - September 26, 2008:

Obama - 48% (-1)
McCain - 43% (nc)

[Tender Branson]

Sunday - September 28, 2008:

Obama - 50% (+1)
McCain - 43% (nc)

Yesterday's sample only:

Obama - 51%
McCain - 42%

[Tender Branson]

Todays numbers adjusted for a more reasonable E-Day composition:

White (76%): Obama 39%, McCain 50%
African-American (11%): Obama 94%, McCain 3%
Latino (10%): Obama 67%, McCain 21%
Asians/Others (3%): Obama 83%, McCain 4%

Obama 49.2%, McCain 40.6%

Dailykos/R2000 is based on 73% White, 13% African-American, 13% Latino and 1% Other.

[Tender Branson]

Todays numbers adjusted for a more reasonable E-Day composition:

White (76%): Obama 39%, McCain 50%
African-American (11%): Obama 94%, McCain 3%
Latino (10%): Obama 67%, McCain 21%
Asians/Others (3%): Obama 83%, McCain 4%

Obama 49.2%, McCain 40.6%

Dailykos/R2000 is based on 73% White, 13% African-American, 13% Latino and 1% Other.

Underestimating both McCain's white and Latino support.

EDIT One more example:
Latino on September 11
Obama 67%-McCain 28%
Latino Today
Obama 67%-McCain 21%

So Obamas support hasn't grown but 7% of McCains has moved to undecided? Yeah right.

http://www.gallup.com/poll/108040/Candidate-Support-Race.aspx

Rasmussen has long not shown a breakdown by race, but when they did they had Obama up by 30 or so among Latinos and that was when the race was close nationally. McCain can be really happy if he gets a third of Latinos right now or on E-Day.

[Tender Branson]

Saturday - October 4, 2008:

Obama - 52% (+1)
McCain - 40% (nc)

[Tender Branson]

Sunday - October 5, 2008:

Obama - 52% (nc)
McCain - 40% (nc)

Saturday sample (363 LV, MoE = 5%):

Obama - 53%
McCain - 40%

[Tender Branson]

Monday - October 6, 2008:

Obama - 52% (nc)
McCain - 40% (nc)

[Tender Branson]

R2K's Obama figures are similar to Gallup's and Rasmussens'. McCain's are much lower, thus resulting in this huge gap. Are they just not pushing leaners as much?

The difference is that they have Barr and Nader included, based on voluntary mentioning.

[Tender Branson]

Tuesday - October 7, 2008:

Obama - 52% (nc)
McCain - 41% (+1)

Today's numbers adjusted for a more reasonable turnout:

Democrats (40%): Obama by 88-8
Republicans (35%): McCain by 85-7
Independents (25%): Obama by 51-39

Obama 50%, McCain 43% - and their numbers start making more sense.

FYI: R2000 uses 35% DEM, 26% REP, 30% IND and 9% OTHER

[Tender Branson]

Thursday - October 9, 2008:

Obama - 51% (nc)
McCain - 41% (nc)

Wednesday Sample (360 LV, MoE = 5%):

Obama - 52% (+2)
McCain - 40% (-2)

[Tender Branson]

Given the 4%-jump for Obama yesterday and Zogby was showing a 2-point jump, I expect the beginning of the Obama-debate-bounce.

But lets wait for Rasmussen and Gallup (well, can Gallup get even any higher ?)

[Tender Branson]

DailyKos is not doing a poll. Research2000, a respected and non-partisan polling firm is doing the poll. DailyKos is giving them the money to do the poll and getting the results in return. There's no conspiracy here; DailyKos is interested in winning elections, and I don't really get how fake polls would help them to that end.

But in their poll you have to subtract about 2% from Obama's numbers and add another 2% to McCain's and the overall numbers are reasonable.

There's no way their party breakdown is accurate.