Pollster.com

Articles and Analysis

 

Morning Status Update for Thursday 10/23

Topics: Status Update

Yesterday's 17 new statewide and 12 new national poll releases look like a picture of stability in the race for President. The state of Washington shifts back to strong Obama dark blue, but otherwise, the new surveys show mostly random change in both directions.

At the state level, 11 of the 17 polls show movement of at least a point in Obama's direction, 4 in McCain's direction since the last survey by the same pollster. If we focus only on the 9 that follow up on previous polls conducted earlier in October, they have a hint of more recent gains by Obama: 6 show movement in Obama's direction, 2 in McCain's direction and 1 with no net change.

081023 todays.png

The impact on our trend estimates within the more contested states is again mostly offsetting. The margin shifts slightly in Obama's direction in 6 states and in McCain's direction in 4. Over the past week, Obama's margins have increased in 13 of the more competitive states, McCain's in 7.

081023 trends.png

In the state of Washington, a new Elway survey showing Obama leading by 19 points (55% to 36%). That bumps Obama's lead on the Washington trend estimate to 9.3%, enough to qualify for "strong" Obama status.

Florida is one state that has shown significant movement back to McCain. Although Obama still leads by two points on our trend estimate there (48.1% to 45.9), the margin has narrowed from a 6.1% Obama lead a week ago. Yesterday's new Mason-Dixon survey, showing a dead heat of McCain 46%, Obama 45% helps confirm that tightening.

At the national level, the new surveys again bump up Obama's margin slightly for the fourth day in a row. Of the eight national trackers, yesterday's releases shifted margins slightly in Obama's direction on four (including Gallup's "expanded" likely voter universe), in McCain's direction on two and showed no change on two.

081023 trackers.png

 

Comments
Anstoß:

Any word on the relevance of the AP outlier?

____________________

C.S.Strowbridge:

Every day the numbers don't close, is one less day for McCain to turn his campaign around. He needs to close the gap by 0.6% every single day, or he will lose.

I don't see how that's going to happen, outside of something major. We are practically at the "Live Boy / Dead Girl" level.

____________________

MarylandDad51:

I just saw some poll results from Quinnipiac and the Big10 poll that were linked on DailyKos and are very good for Obama, especially in Ohio. Will they be up here soon?

____________________

Disco Stu:

TWO NEW STATE POLLS HAVE OBAMA UP DOUBLE DIGITS IN OHIO. ALSO ONE HAS HIM UP 10 IN INDIANA! (Big Ten & Quinnipiac)

I think these are out of whack....but still, even if they're off by 4-5-6 points, STILL good news for Obama!

____________________

McSame/Falin:

McCain might get another bounce when the Scooter Libby endorsement comes out later today

____________________

CTPonix4BHObama:

I almost forgot about Scooter Libby, doesn't he have a pardon coming his way soon?

____________________

THX1138:

When do we get the video popping up from Bin Laden, videotaped against a sheet? It's coming soon, right? That's what he did in 2004, days before the election. The question is: what will he say? And whom will it benefit?

____________________

Anstoß:

I'm beginning to think that Bin Laden (as some intelligence reports indicate) is already dead. I would be floored if we had a movie from that sociopath which shows him (with moving lips!) discussing current events. What would be unsurprising: Frozen picture of Bin Laden with a "Bin Laden" voiceover. If alive, he'd probably want McCain to win to keep his cause popular and recruiting smooth.

____________________

ruyen:

If anyone heard Democracy Now's show yesterday, you'll know that voting machines are already flipping Obama votes to McCain and the move to steal the election (like Ohio's hundreds of thousands of lost votes in 2004) is already in motion.

Point being: Don't let up, continue to donate/volunteer, and don't get content, even if Obama goes up 10pts in every single state please keep working toward victory.

____________________

C.S.Strowbridge:

"Point being: Don't let up, continue to donate/volunteer, and don't get content, even if Obama goes up 10pts in every single state please keep working toward victory."

Always
Be
Closing

As I heard on The Young Turks.

____________________

conorx:

You now have OHIO as a lean Obama???? Toss-up surely??

____________________

RaleighNC:

National polls are rather useless at this point. It's all about the state polls.

____________________

SchmaltzyTheGolem:
____________________

I may be wrong, but is this the first time that the pollster.com and RealClearPolitics electoral maps have both shown the same states in the broad categories O, M, toss-up?

____________________

Trivial I know, but is this the first time that the pollster.com and RealClearPolitics electoral maps have shown exactly the same states in the three broad categories O, M, and toss-up? That's stability.

____________________

Jerry B:

I am not familiar with political polling, but I have spent my career conducting survey research for businesses. I heard a plausible explanation for why the presidential polls vary wildly or overstate Obama support.

To estimate the election, each poll is required to gather enough responses from each political party, ethnicity, age group, and gender. There may be other factors, too.

If polls are asking for political party identification at the beginning of the poll, could this cause a position bias which produces erratic results in political polls?

____________________

Mark Lindeman:

It doesn't make sense to attribute the AP result to the high proportion of self-reported evangelical/born-again respondents. I have no explanation for that right now (maybe someone has figured it out), but the proportion is practically identical for total respondents (45%) and likely voters (44% -- actually smaller). see pdf page 24. But Obama leads in the total sample by 10 points, and in the LV sample by only 1. The total sample is 23% 18-29; the LV sample is 15%. (p. 22) That's an even smaller proportion than in the 2004 exit poll (17%). I don't know what specifically is wrong with the AP likely voter screen (or with this sample), but that makes no sense.

____________________



Post a comment




Please be patient while your comment posts - sometimes it takes a minute or two. To check your comment, please wait 60 seconds and click your browser's refresh button. Note that comments with three or more hyperlinks will be held for approval.

MAP - US, AL, AK, AZ, AR, CA, CO, CT, DE, FL, GA, HI, ID, IL, IN, IA, KS, KY, LA, ME, MD, MA, MI, MN, MS, MO, MT, NE, NV, NH, NJ, NM, NY, NC, ND, OH, OK, OR, PA, RI, SC, SD, TN, TX, UT, VT, VA, WA, WV, WI, WY, PR