New DailyKos/Research 2000 tracking poll

[J. J.]

If DailyKos is showing a tie, McCain is winning.

[J. J.]

If DailyKos is showing a tie, McCain is winning.

Just because you wish it to be true doesn't make it so.

Apparently, they over represent Democrats.  So even after that, McCain's tied.

[J. J.]

If DailyKos is showing a tie, McCain is winning.

Just because you wish it to be true doesn't make it so.

Obviously, because Daily Kos does not have an agenda whatsoever.

Daily Kos has an agenda. But Research 2000, whon they contracted to do polling, did not.

Research 2000 over weighted the Democratic sample.

[J. J.]

With any poll, there are tiny methodological differences that might make a difference.  I was referring to this before in comparisons of Gallup to the 'bots.  If we can identify those, compensate for them, we might be able to find the real numbers (with "good" samples).  I'm looking at things to eliminate the "noise."

[J. J.]

This poll might make Zogby look accurate.

(This is the worst thing I've ever said about the DailyKoz.)

[J. J.]

This is about as legit as Zogby.... so it is all irrelevent.

The MOE makes it horrible, in addtion to its biased polling.

lol, Republicans are fun.

It's Daily Kos, dude. That's like JJ, StatesRights, and I deciding to do a poll. Actually, this is still worse. We actually understand that an inaccurate poll is worthless.

The have some problems with their weighting; if they would fix that, I'd be happy with it.

[J. J.]

Sunday - October 12, 2008:

Obama - 53% (+1)
McCain - 40% (nc)