RCP has two Gallup tracking polls today. First one with it tied, and the second one Obama +5

[hrmjustin]

What gives?[link:http://realclearpolitics.com/epolls/latest_polls/|

[Blue Yorker]

1. That is very, very stupid

Because the same 3 days are contained within the 7-day poll. Why aren't they patient and wait until days pass, and simply post the 7-day one?

The point Gallup was trying to make is that in the 3 days after the debate, things changed. That doesn't mean it's a new poll.

[Arkana]

2. The tied one is that 3-day tracker Gallup released.

The one with Obama up 5 is the most recent one, but the Forbes-owned RCP won't tell you that.

[gcomeau]

3. It was not "a 3 day tracker".

It was three of the days of polling from the 7 day tracker. They were just showing what the surveys immediately after the debate looked like but it was not a separate poll.

RCP is being extremely dishonest treating it as one. They have a history of seizing on any opportunity to shift their averages Republican. Delaying dropping Romney-favorable polls from their averages... dropping Obama favorable polls from the average as quickly as they can justify...this is pretty bad even for them though.


Edit: Oh good grief, it's worse than I thought. They are NOT COUNTING the +5 number in their average! THAT is the official tracking result!!!! And it's a day newer than the"tie"!

[TroyD]

6. If that's the case then it is time to take on RCP

They are showing their true colors.

[The_Counsel]

14. Wait ... so "Real Clear Politics" Isn't 'Real Clear'?

Unpossible!

[hrmjustin]

4. Ok so it was a 3 day part of the 7 day tracking poll. ok that makes sense.

[PM Martin]

5. That makes no sense.

[VirginiaTarheel]

7. RCP should be called out for their slant

If they can treat three day post debate sample as a separate poll, why not treat Sunday's polling as its own unique poll too?

[hrmjustin]

8. I agree and welcome to DU

[woolldog]

9. because Sunday is a good number for Obama

Duh!

[TroyD]

10. Gallup messed up too

They should have clarified this to the media when they released this poll this morning by pointing out that it was not a proper Tracking Poll like the one they release each day at 1:00 PM.

They've allowed RCP and others (Eg. CNN) to jump on this and portray it as a Tie.

[flpoljunkie]

11. More on Gallup polling today from Talkingpointsmemo...

Obama Up 5, Approval Rating Above 50 Percent In Latest Gallup Tracking

TOM KLUDT 1:56 PM EDT, MONDAY OCTOBER 8, 2012

President Barack Obama leads Mitt Romney by 5 among registered voters and boasts an approval rating above 50 percent, according to the Monday's daily tracking numbers from Gallup.

In the head-to-head, Obama outpaces Romney, 50 percent to 45 percent.  But that 5-point margin is based on responses gathered during the tracking period of Oct. 1-7, which means that it was partially conducted before the Oct. 3 debate and partially after Friday's jobs report.

Obama's approval rating may serve as a fresher indicator of the state of the race.  Using a 3-day rolling average, the latest approval rating from Gallup is based on the tracking period of Oct. 5-7, which encompasses the aftermath of the first debate and the two days that immeidately followed the jobs report.  Fifty-one percent of American adults surveyed in Monday's release approve of the president's job performance, compared with 44 percent who disapprove.  

http://livewire.talkingpointsmemo.com/entry/obama-up-5-approval-rating-above-50-percent

[titaniumsalute]

12. Posting a partial data release and a full data release is fucking insane

Really...They are posting the post-debate 3 days (the tied stuff) with the full sample of the last seven days. This makes their whole aggregate idea fly out of the window.

They need to stick with similar polls to kee the agreggates the same. If Romney got a bounce it will flesh out in a few days for the seven day poll. If not, then it doesn't. But to stick the same sample in there twice is just dumb and makes Real Clear lose cred.

[madashelltoo]

13. Tamron just reported tied on MSNBC

[VirginiaTarheel]

18. Somebody tell Tamron the new numbers

[mzmolly]

15. It is explained here:

http://www.democraticunderground.com/10021494466

[tarheelsunc]

16. RCP has ALWAYS had a considerable bias

I just use them as a jumping-off point for other polls.

[flpoljunkie]

17. The 47-47 tie is based on three days of polling: October 4-6.

Gallup Tracking* 10/4 - 10/6 1387 RV 3.0 47 47 Tie

http://www.realclearpolitics.com/epolls/2012/president/us/general_election_romney_vs_obama-1171.html