Pollster.com

Articles and Analysis

 

US: Obama 49, McCain 45 (GWU 10/13-16, 10/19)

Topics: PHome

GWU/ Battleground
Tarrance Group (R)/Lake Research (D)
10/13-16/19, 08; 1,000 LV 3%

National
Obama 49, McCain 45

 

Comments
BOOMFAIL:

Obama 4-8 point lead in the polls so far today still points to a huge win. Two more weeks, my friends!

____________________

mandalorianarmy:

Obama is up by about 5 points right now. All the polls seem to be hovering around that number. Not as high as he was a few weeks back but he is still over 50%.

____________________

BOOMFAIL:

George Bush J. Danforth Quayle Republican 48,886,597 53.37% EV 426

Michael Dukakis Lloyd Bentsen Democratic 41,809,476 45.65% EV 111

Interesting to see what a 7-8 point spread can lead to.

____________________

BarackO'Clinton:

"the race is tightening"

and

"McCain is winning"

are two totally different worlds

____________________

boomshak:

I know no one in the Obama camp likes to look to past behavior as indicative of future events (history repeats itself). However, Obam did have a tendency to poll above perform throughout the primaries.

Many reputable pollsters had Obama winning NH by double-digits. Hillary won by 4. That's at least a 14 point swing.

A 4.8 (RCP Avg) lead for Obama here is truly pathetic. He has the entire MSM as a free advertising arm of his campaign and McCain has run a pretty sh*tty campaign and 4.8 is the best he can do?

Anything can happen, but it is not without reason that one could imagine this being a dead-even tied race on Nov 4.

If that happens, I give the edge to Obama because he leads amongst dead, imaginary and multi-voting voters.

____________________

Indiana4Obama:

Boom,
You're wrong...NH is the only state where pollsters got the Obama advantage wrong. They underestimated him in NC, Indiana, Colorado, Virginia, etc. Why? A few reasons...underestimating the AA and youth vote were major ones.

All the current polls have Obama with a 4-8 pt lead nationally and with leads in key battleground states. These may tighten, of course, but we'll see.

____________________

BOOMFAIL:

Living in Wisconsin, I can tell you that Obama did NOT poll 17 points above Clinton before the primary. He DID however, get 17% over her. FAIL.

____________________

Unbiased08:

Boomshak,

Your statement that Obama polled better than he performed is completely false and has been thoroughly discredited as a premise. Do your homework. In fact, the reality is the opposite. OBAMA OUTPERFORMED POLLS more often than the other way around during the primaries. There were a few notable exceptions, but there were several states where Obama ended up 10+ points ahead of how he polled.

____________________

Adrian B:

This is what I think is going to happen in the next few days:
1/ the polls are going to tighten as all the undecideds come off the fence. No-one really believes Obama will win by +10 in a non-incumbent race.

2/ The pollsters will begin to converge on their findings as they refine their likely voter models and depserately try to make sure they are close to the actual result (no pollster wants to be known as hopelessly inaccurate).

3/ The national polls will settle at around +6 Obama by next weekend (I've been using a model of adding a couple to the Rasmussen tracker and taking a couple off Kos and it tends to be bang of the Pollster.com trend estimate).

4/ The polls will get closer in the final week but on election night it will be a lnadslide.

There - you heard it here first!!

____________________

Unbiased08:

traditional phone polls don't reach a lot of Obama's core voters, and this explains why Obama outperformed polls in the primaries. It also doesn't account for all the newly registered voters that Obama's campaign has added to the rolls, especially in battleground states. So, while there is actually a proven and legitimate reason to expect Obama to do better than his polling numbers, the only (fabrciated) reason why (some) people think Obama will underperform is based on a delusional hope that some magic occurrence will keep Obama from winning.

____________________

BOOMFAIL:

Here's another:

http://www.realclearpolitics.com/epolls/2008/president/nc/north_carolina_democratic_primary-275.html

I'm sure there are many more, but should prove that just repeating talking points does not change the FACTS.

____________________

OneAngryDwarf:

boomdonk said:

If that happens, I give the edge to Obama because he leads amongst dead, imaginary and multi-voting voters.

Boom I love it when you eviscerate us with your wit. It is like being stabbed with a plastic spoon, kind of a symbolic gesture that accomplishes nothing.

____________________

boomshak:

WHY OBAMA IS NOT FURTHER AHEAD:

America is on new ground here.

Never before has America been in a position to elect a Socialist POTUS who hobnobbed with a terrorist, had a black supremicist as his mentor, worked closely with an organization neck-deep in voter fraud and has no significant achievements even though he has been in government for 12 years.

It's a TOUGH PILL to swallow.

____________________

joecooper:

I also think a 4.8 lead is pathetic. The only people at this point who are voting for McCain/Palin are the willfully ignorant. I'm ashamed that my country has so many of them.

____________________

Unbiased08:

And one more thing, rcp's average which has Obama ahead by 4.8 is the most contrived cherry-picked result imaginable. Three polls released 6 days ago had Obama +9 (LA Times), +9 (McClatchy) and +14 (Newsweek). When Obama would get a bad result, those numbers would stay part of the average for two weeks at least. There aren't that many more polls coming out now than there were a few weeks ago, so there is absolutely no reason to discount the LA Times, McClatchey, and Newsweek polls that have Obama with a big advantage.

Plus, rcp's 50% credit to gallup's traditional likely voter model is suspect. If Obama leads by 10 among registered voters, it is total garbage to give any credit to some bizarre interpretation of that result which causes McCain to gain 7 points.

If you average the polls from the last week (which hardly seems like a stretch), and discount absurd likely voter models, Obama's lead is closer to 7 points.

____________________

zen:


In NH primary, Hillary's tears in restaurant just before the election day made a huge turning point for undecides and white women voters.

But the states where AA and youngsters are big population, Obama outperformed by double digits such as VA, NC.
So, I'm actually very certain that in those 2 states, obama will win even if the poll is tied.

____________________

DTM:

Here is a chart detailing how Obama performed in the primaries relative to polling:

http://faculty.washington.edu/agg/Bradley_&_Reverse_Bradley.Dem_Primaries.2008.pdf

The upshot is that using +/-7% as the dividing line for the polls being significantly off (this number comes from the authors of the study associated with that chart), Obama underperformed significantly in three states: New Hampshire, California, and Rhode Island. He overperformed significantly in twelve states: South Dakota, Montana, Wisconsin, Florida, Delaware, Virginia, North Carolina, Alabama, Georgia, Maryland, South Carolina, and Mississippi.

____________________

NHBlue:
____________________

orange24:

It's a TOUGH PILL to swallow

Better get a big glass of water then, because you're going to have to swallow that pill on Nov 4.

____________________

OneAngryDwarf:

@NHBlue

I especially love that you took it off of Fix News. Nothing says you are screwed like having the home team admit you are a crook.

So I wonder if Intrade is going to have a line on this?

____________________

whitetower:

@DTM

Well, the so-called "reverse Bradley effect" noted in the study you cite takes into account several states that very small samples (or no sample) in the week prior to the primary.

Further, the regression model uses the AA population as a variable. In my understanding of the Bradley effect, the important variable is white, blue-collar independents and "Reagan Democrats", not African-Americans.

Sure enough, states showing a (marginal) Bradley effect -- even using the authors' model -- include OH and PA.

____________________

cinnamonape:

"New Hampshire, California, and Rhode Island. He overperformed significantly in twelve states: South Dakota, Montana, Wisconsin, Florida, Delaware, Virginia, North Carolina, Alabama, Georgia, Maryland, South Carolina, and Mississippi."

So using Booms premise...that there are under-reported voters out there not being measured in the polls...one would possibly give McCain New Hampshire (although the gap is pretty wide there)...while giving Obama...Montana, Florida, Wisconsin, Virginia, North Carolina, and possibly even Georgia.

Seems a fair trade-off!

____________________

cinnamonape:

"New Hampshire, California, and Rhode Island. He overperformed significantly in twelve states: South Dakota, Montana, Wisconsin, Florida, Delaware, Virginia, North Carolina, Alabama, Georgia, Maryland, South Carolina, and Mississippi."

So using Booms premise...that there are under-reported voters out there not being measured in the polls...one would possibly give McCain New Hampshire (although the gap is pretty wide there)...while giving Obama...Montana, Florida, Wisconsin, Virginia, North Carolina, and possibly even Georgia.

Seems a fair trade-off!

____________________

DTM:

@whitetower

I was only linking that chart for the purpose of setting forth the factual record on how Obama performed in the primaries relative to his polling. The debate over whether we should expect a Bradley Effect, or a Reverse Bradley Effect, or in fact both, is a much more complex topic. Personally, I am hesistant to overrely on the primaries for that purpose.

But it may be worth just noting, since you brought it up, that the study found that while both effects may have shown up in the primaries, the Reverse Bradley Effect was apparently stronger overall, and as a result dominated in more states (note the authors originally expected the Bradley Effect to dominate). That is in part because in many states not just blacks but also whites seemed to understate their support for Obama, most often whites in non-Southern but traditionally Republican states like Montana or Indiana.

____________________

DTM:

@whitetower

Oh, and the study looked only at primaries where there was at least some polling in the week before the primary, but I can't claim in each state there was as much polling as we would ideally want for these purposes. On the other hand, I don't know of any reason to suspect that was systematically biasing the data.

____________________

political_junki:

I think yuo should all ignore BOOM. Why do you even bother arguing with him??
Let him keep copy pasting messages from rightwing nuts websites and rot in his misery of a liberal america. ignorehim

____________________

tjampel:

If you told Obama, back in early Sept that, on this day two weeks out from the election he'd be almost 5 points up in the RCP avg. do you think he'd be even slightly disappointed? He and his staff would be ecstatic, and for good reason. The only candidate to surmount this kind of deficit in modern history was Reagan, who was up against an unpopular

Add in the fact that Obama has superior GOTV, Ad buying capabilities and greater enthusiasm (as shown in early voting stats in GA and NC, etc.) and you get a portrait of a race which, unless something drastic happens really quickly, is essentially all over but for the extremely perverse ravings of loons who go around shouting that "Obama (the socialist)should be up by 20" AND, at the same time telling us that "America is a Center-right country". How do you square these two statements when made by the same person? Well you don't, you just up the dose...

____________________



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