Tom Holbrook at UW-Milwaukee has an interesting set of data on the impact of Presidential debates since 1988. The typical debate over this period has moved the national polling trend support by an average of 2 points. Tom's whole piece is worth a read, but I'm going to steal his data table and hope he won't mind:
Note that Holbrook's numbers describe the support level for one of the two candidates. Around here, we generally speak in terms of margins ... if the Gallup tracker goes from Obama 48--McCain 44 to Obama 50--McCain 42, we'd call that a 4-point swing, whereas Holbrook would describe it as a 2-point swing. Not that there's anything wrong with that, but it means that you'd need to double these numbers to be consistent with FiveThirtyEight nomenclature.
Still, even when you double these numbers, the swings they produced simply aren't that large. The most lopsided debates I can think of over this period were the town hall debate in 1992, when Bill Clinton shined, and the first debate in 2004, when George W. Bush stunk. Those debates moved the numbers by 4 and 4.5 points, respectively.
If you moved outside of this period, you might come across some bigger swings -- Ronald Reagan had a very large surge in the polls after his debate with Jimmy Carter in 1980, but those were somewhat unique circumstances: the only debate that year involving the two major party candidates (Carter had tried to duck Reagan earlier), and one that came just a handful of days before the election. And of course, Ronald Reagan could run laps around either Barack Obama or John McCain as a debater.
Barack Obama presently has a 2-3 point lead nationally, which means that if McCain wins the debate and gets an average-sized winner's bounce, the race should draw back to a tie, or perhaps a very small Obama lead. The bar for McCain actually pulling ahead after the first debate is fairly high -- Clinton-in-Richmond high.
This is one of those reasons why our model is making Obama nearly a 3:1 favorite in spite of a seemingly small lead in the popular vote. Once you get out of the convention period, voter preferences tend to have become a lot more stubborn, and even terrific or terrible debate performances don't tend to alter them all that much.
9.23.2008
Debates May Not Be Decisive After All
@ 9:10 AM
Subscribe to:
Post Comments (Atom)
253 comments
Kerry was better in the 2004 debates and lost the election.
Well, certainly something to keep in mind. I guess this just means the debates will be unlikely, barring a major happenstance, to sway public opinion dramatically. So, when will we see the economy bump of last week peak? I'm seeing polls still dated from the 14th- Quinnipiac just released some today, showing Obama ahead in MI, WI, MN, and CO dated thusly.
Morning,
Rasmussen has it tied again, 48-48 in national tracker.
geoff how do you know?
It strikes me that debate performances are more about crystallizing existing beliefs than creating new ones in the minds of voters... they reinforce or create a tipping point. ex, Clinton had been saying for months that GHWB didn't care about ordinary people, that he was out of touch.. then he went and proved it -- giving people the visual for a perception that had been primed.
and I take it McCain took the least lastnight in Rass?
I admit it. I'm Nate Silver. I post my theories on here to see how people react under "Inkstain," then the next day or two I come up with statistical evidence that proves I'm right. Win for me.
"So, when will we see the economy bump of last week peak?"
There is no economy bump. There's a return to the natural level of the race after the convention bumps played out.
"Rasmussen has it tied again, 48-48 in national tracker."
Hooray. I was feeling really, really bad for conservatives that they didn't have *something* to cling to in the recent polling.
Rasmussen is so out of line with everyone else I do not what to make of it.
I think thew debates will be much ado about nothing. Both will be competent, neither hit a home run, and the current leader will remain in the lead.
"Kerry was better in the 2004 debates and lost the election."
Yes, but he made up a lot of ground after the first debate. He had been down by a huge amount to Bush.
Anyway, I am not sure about this. The Quinn poll out today does show not that many people saying they are very likely to be influenced by the debates, but add in "somewhat likely" and the figures jump to about 1/3 of independents in four swing states. That's fairly significant.
I actually thought so!
Perot won the debates, Kerry won the debates.
Gore was universally considered to have won the first debate, then SNL parodied him brutally and the tide turned on him, forcing the last two debates into draws.
wait I can't tell if you're joking?
NBC News/Mason-Dixon poll in Florida has Sen. Barack Obama leading Sen. John McCain, 47% to 45%.
"Yes, but he made up a lot of ground after the first debate. He had been down by a huge amount to Bush."
Which goes into what I've been saying. Moving around inside deficits or leads is relatively easy, flipping from one to the other is very hard.
Eve-
Yes, I saw that on MSNBC this morning. If McCain loses FL he must pick off PA or MI to have any shot. Without FL and NC and VA McCain's prospects are pretty bleak.
I also think that a lot more is riding on the debates this year than in most elections.
Not only do we have the economy, but a lot of folks who favor Obama not because they really think he'd be a great president, but just because McCain and Palin are such pieces of shit. They will be looking to the debates for some confirmation that Obama is a real leader and can handle himself well.
I would not underestimate how important these debates are for Obama.
The debate effect may be visual, anyway, rather than a matter of who "wins" the effect.
I'm thinking of the experiment in Blink where, if I recall it correctly, people were able to correctly assess whether a doctor was likely to be sued by viewing a few seconds of video without sound.
(Also, famously, the Kennedy/Nixon debate, which Nixon won on the radio)
good morning guys
From Quinnipiac University/washingtonpost.com/Wall Street Journal polls:
COLORADO: Obama 49 – McCain 45
MICHIGAN: Obama 48 - McCain 44
MINNESOTA: Obama 47 - McCain 45
WISCONSIN: Obama 49 - McCain 42
From NBC News poll- FL:
Obama 47, McCain 45
OMFG..FAKE POLLS
McCain LANDSLIDE !!!
say it with me liberals
McCain L A N D S L I D E !!!
OBAMA will not get even 30% in FL
election is over !
McCain L A N D S L I D E !!!
"Which goes into what I've been saying."
Yes, you keep saying it, but there's really nothing but anecdotal evidence to support it. You're engaging in mathematical myths.
The difference between two candidates has never been so stark so this historical comparison lark is to no purpose-I expect Obama to lead by at least 7 points after the debates.
Mule Rider:
Mark Buse called and wants to hook up with you this weekend.
Be sure to bring the lube.
Charles-
The polls say you are wrong, most folks are voting strongly for Obama and not against McCan't.
The debates can't hurt much if he is competent, the only side he really has is an upsaide, if he does really well he draws independents.
"The debate effect may be visual, anyway, rather than a matter of who "wins" the effect."
As an Obama fan, I'll take that. Half a foot taller and almost three decades younger will do wonders.
Rasmussen is live and yes, it is reporting a tie.
Are expectations really that low for Obama in the first debate?
At the very least, I think that Obama still needs an 'OK, I can picture this guy in the White House' moment that connects with undecideds. There may be as many, 'I'm ready to vote Democrat, but I just wanna be sure with this guy' who are still undecided as there are undecideds who fall in the category of, 'I hate Bush, but I'm reluctant for that minority fellow'.
"The polls say you are wrong, most folks are voting strongly for Obama and not against McCan't."
Which polls would those be?
"Yes, you keep saying it, but there's really nothing but anecdotal evidence to support it. You're engaging in mathematical myths."
15 elections, 13 winners in the post-convention leaders.
I'll start with that.
the first debate is standing.
assmole
keep dreaming
fake polls
McCain is ahead by 20 points in the real polls
McCain L A N D S L I D E !!!
LOL! Yes, the problem for McCan't is that those are real polls and if BO wins FL and CO and MI and WI, how does McCain win?
Yes, bjut the states polls are more pretty for Obama than McCain.
Quinnipiac confirms leads in CO, MI, MN and WI.
In the RCP average Obama is now up by 3.3% in CO and 5% in MI.
I don't think Obama will win the debate per se, but if it's gaffe free and he doesn't have a horrible visual moment, I'll be pretty happy.
Speaking of debate bumps:
http://tinyurl.com/debatebump
"15 elections, 13 winners in the post-convention leaders.
I'll start with that."
That's also meaningless.
You sound like those idiot sports commentators who say things like "the team in the Super Bowl that scores first wins X out of Y games". The reason for the lopsided numbers is because this includes all the elections/games that weren't close.
This one *is* close, so comparing it to a bunch of elections that weren't doesn't mean anything. You are perpetuating myths.
And I'd love to see the actual numbers behind that claim anyway, if you have them. Thanks.
McCain will turn blue states in to red
wait & see
McCain L A N D S L I D E !!!
NBC News/Mason-Dixon poll in Florida has Sen. Barack Obama leading Sen. John McCain, 47% to 45%.
[reposted from previous thread]
Here's an interesting tidbit in Central Florida. Listed below are 'TV Ad Tallies' that I personally witnessed since waking up a few hours ago:
Obama - 11 ads
McCain - 0 ads
Non-Campaign - 1 ad (Obama Against Surge)
I'm not sure why the McCain ads have dropped so sharply in the local market this week. It previously was around a 60%-40% McCain TV Ad advantage right up until about 10 days ago, when Obama began an advertising 'surge' of his own. Of course, this is just one morning's snapshot. But its clear that Obama is at least spending SOME of that '$39mil For Florida' that Mr. Plouffe was laying out last week.
Geoff:
Can you calculate how the Rasmussen poll tracked the last 3 days?
Geoff:
Can you calculate how the Rasmussen poll tracked the last 3 days?
Geoff:
Can you calculate how the Rasmussen poll tracked the last 3 days?
What are the states polls from Rasmussen today?
McCain was up 2 in Rasmussen's polling last night, that's why the tracker is tied now. Gallup will also show a 2 or 3 point gain for McCain today, leaving Obama at +1 or +2 in the tracker.
Obama's Wall St-panic bounce is fading, but he will still have a very slight lead overall going into the debates.
I agree that the debates will not cause a big shift, but in such a close race it could be decisive. Even 1-2% could decide the election.
Please, I know about proper use of statistics.
For one thing, the "team that scores first" statistic is more meaningful than you think. It takes out shutouts, for one thing. For another, football strategies change with relative score.
"This one *is* close, so comparing it to a bunch of elections that weren't doesn't mean anything. You are perpetuating myths."
Okay. What close elections flipped late? 2004 didn't. 2000 didn't. 1992 didn't. 1980 did.
I'll fully admit that, unlike the debunked Bradley Effect, I don't have a pile of statistical evidence for this fact, so if anyone wants to disagree that is their right. But I stand by this phenomenon.
tied
I'm glad Rasmussen is tied. It was getting a little lopsided around here. Boring.
Don't forget this announcement from Ras:
"Results from this past week showed that the number of people considering themselves to Democrats spiked early in the week as the economic problems on Wall Street became visible. Overall, it was the best week for the Democrats since July. It remains to be seen whether this might lead to a lasting adjustment or be more of a bounce like that resulting from a party’s nominating convention."
His party IDs still use an average much less favorable to D's. When he makes announcements like this, he's telling us that the recently favored party is actually polling MUCH better than the ID-weighted numbers show. Presumably BO is +3 or so. We saw exactly the same process in the other direction when Mc's spike didn't show up much at first in Ras, and a similar announcement came out.
Of course if party ID settles back to the D +5.5 Ras is currently using, then his implicit assumption that it changes only slowly will be confirmed. The point is, however, that he's essentially telling us that his current numbers look like everybody else's, but that his ID system damps out some noise at the cost of responding sluggishly to real changes.
/mbw
test your insight at Healthy Wealth, the unlikely blog to grow with?
welcome all comments with democracy for all, by all, in all, to all bloggers and netizens .
http://theinnozablog.blogspot.com
This is the data from Rasmussen.
"The Rasmussen Reports daily Presidential Tracking Poll for Tuesday shows the race for the White House is tied once again—Barack Obama attracts 48% of the vote and so does John McCain. Both men are now viewed favorably by 54% and Rasmussen Markets data currently gives Obama a 51.9% chance of victory.
Republicans typically win a majority of the White vote and this year is no exception. Among White Democrats, 78% say they’ll vote for Obama, 17% for McCain. Obama leads among non-White voters by a 71% to 24% margin".
Pay attention to "Rasmussen Markets data currently gives Obama a 51.9% chance of victory".
"I'll fully admit that, unlike the debunked Bradley Effect, I don't have a pile of statistical evidence for this fact, so if anyone wants to disagree that is their right. But I stand by this phenomenon."
Well, you keep talking about 13 out of 15. What's that based on? You must have some numbers, don't you?
"Pay attention to "Rasmussen Markets data currently gives Obama a 51.9% chance of victory".
About the same as Intrade. Some serious money to be made there.
Just as I have recently come to the conclusion that the Daily Kos poll really is tilted towards Obama, it is becoming increasingly difficult to give Rasmussen the benefit of the doubt on his party ID splits.
"Well, you keep talking about 13 out of 15."
13 of 15 is the fact that leader once the post-convention bounces have faded out have won every election since WWII except Truman and Reagan.
"What's that based on? You must have some numbers, don't you?"
It's mostly my impression, but I do believe Gallup had something on it shortly after the conventions. Having trouble finding it.
"McCain was up 2 in Rasmussen's polling last night, that's why the tracker is tied now. "
The Wall St. wll end up being good for McCain. His new ad eviscerates Obama for a lack of leadership -- it's a very good. McCain has looked decisive. He proposed a plan and demanded people like Cox be held accountable. You may think it's scapegoating, but that doesn't matter. The optics are good -- much better than Obama not proposing anything.
McCain is having a surge in the polls driven by hus handling of the economy. That's bad news for Obama.
That's really interesting, real joe. (wonders what phony joe's analysis is like)
By the way I was right about it being tied going into the debates, real joe. Don't let that worry you.
"Just as I have recently come to the conclusion that the Daily Kos poll really is tilted towards Obama, it is becoming increasingly difficult to give Rasmussen the benefit of the doubt on his party ID splits."
I just look at it as three sets of likely voter screen: Obama-optimistic, Neutral, and McCain-optimistic.
Either of the three might be right, but Neutral is the best bet.
"He proposed a plan".
What plan?
The debates this year may matter. If Obama comes off well he will convince people that it is safe to vote for him and he can open up a lead.
McCain has to show he has his wits and is mentally competent to be president.
My guess is mCcain goes on attack and comes off looking somewhat unbalanced while Obama will keep his cool no matter what happens.
I believe McCain can blow any chance he has with a bad debate performance.
Also the way the debate is set up this year there will be a lot of talk between candidates and McCain may have trouble keeping his temper.
Charles: Here's the Gallup data:
http://www.gallup.com/poll/110074/Shifts-Last-Two-Months-Election-Uncommon.aspx
They go by Labor Day, which puts them inside the conventions and sometimes not, but the basic point remains. Though the median shift was six points, only three campaigns managed to go from losing to tied or leading from Labor Day onward.
On the surface I'm continually surprised that MN is still polling so close, but the more I think about it the Minnesotans I know really do break down 50-50. Its amazing to me though that according to the polls Wisconsin is more "in the bag" than Minnesota is. History would indicate its the other way around.
@Charles- It's likely that Ras' IDs are slightly tilted R systematically compared to other pollsters, but we don't know who's right. What Ras is very explicit about is that his IDs are tilted toward the past. When party ID is changing, one component of his response lags behind the actual change. He can get the same data as everybody else but his weighting system makes the time-dependence lag the actual time dependence. He does it to remove noise, and thus gets what seem to be supernaturally steady results, if you thought the MOE was real. /mbw
"McCain has looked decisive. He proposed a plan and demanded people like Cox be held accountable."
"Decisive"? He flip-flopped several times last week.
And he tried to hold someone accountable who wasn't responsible -- the head of the SEC. Then the next day, he had another senior moment and called for the resignation of the head of the *F*EC.
McCain has been a disaster on this issue.
Charles-
No time to find it, it was a poll earlier this week and showed the enthusiasm gap still exists and that McCain voters are much more likely to be voting against Obama than vice versa - complete reversal of '04. McCain voters are weakly for him, thus OPbama has the upside potential in the debates.
Nate, you don't comment about the 2000 campaigns as a model for debates, or at least the debate period being decisive. It looks like Bush made up a lot of grown in three small increments, though in the report you cited they attribute it more to media coverage of Gore then the debates themeselves. Overall, though, this was very encouraging!
While the debates may not move the national polls all that much they will have an exponentially more important impact in swing states. I think we will need to check MI, PA, FL, OH, VA, and CO after these polls. It seems to me that if these polls move 1-2 percent of the voters in these states that means more than moving peolpe in OK or ID or DC. So older voters who are foreign policy voters who are McCain leaners but not sold could swing and swing voters could swing depending on the two men's performance. I think the data takes a too macrolike approach and needs to be more individualized. Remember, all elections are local etc.
Don't swing states tend to have fewer undecideds going in to the debates? They are the one getting bombarded with info, aren't they?
I know PA has a lot of undecided voters, most democrats. That is according to what I saw on Morning Joe today
And of course independents
I found some actual charts and examined them. Here's what I found:
48 – Dewey led late, lost.
52 – Eisenhower led by huge margin, won.
56 – Eisenhower led by huge margin, won.
60 – Kennedy led by narrow margin, won.
64 – Johnson led by huge margin, won.
68 – Nixon led by narrow margin, won.
72 – Nixon led by huge margin, won.
76 – Ford closed gap late and was ahead in final poll, lost.
80 – Carter led late, lost.
84 – Reagan led by huge margin, won.
88 – Bush led by large margin, won.
92 – Clinton led by large margin, won.
96 – Clinton led by large margin, won.
00 – Bush led by small margin, won.
04 – Bush led by small margin, Kerry closed gap, Bush won.
Of these 15 elections, 8 were laughers. Take those out and you are left with 7 elections that were close late.
Of those, the late leader won 5 of 7. But of *those* 5, 2 of the trailers closed the lead by the time of the election and lost in a squeaker. There were only 2 I believe where the late leader held onto a clear lead throughout.
The numbers say your confidence on the "13 of 15 victories" is overstated.
McCain is vulnerable on foreign policy, as Obama's recent remarks on the Zapatero gaffe show. I know that Republican magical thinking dictates that, since McCain's foreign policy is indistinguishable from Cheney's, McCain is therefore strong on foreign policy. The truth is that he is outside of the mainstream in his desire to re-instate the military draft, start two new conflicts, and in considering our allies as little more than military reserves under American command.
It's interesting that a few days after McCain's campaign visit to Minnesota, Obama's numbers shoot upward here to O+8. That's revulsion you can believe in.
Oh, and of course, a couple of those elections were close enough to be disputed, which would have further dented your numbers quite a bit.
Battleground and Rasmussen trackers are TOTALLY in tank for McCain. Scott Rasmussen made a huge mistake in associating with the FOX Propaganda Network and Sean Hannity. Simple as that.
McCain was not decisive last week.
He changed his talking point on the economy in 3 HOURS. He is a good windsock.
His calling for the firing of Cox was lambasted by the WSJ and in today's column by George Will. Will is saying McCain is unfit to be president.
So can I use 5/7? Because that's pretty close to Obama's odds of winning, imho.
I don't see how a few of the leads turning into squeakers but still winning dents my point any, when my point is that it's harder to close a gap than take a lead.
I agree with Todd
Most of McCains gaffes have been on foreign policy matters, even in recent weeks. The only thing he is comfortable talking about is being a POW and the surge. Beyond that, he seems equally as clueless on foreign policy as he is on the economy.
"So can I use 5/7? Because that's pretty close to Obama's odds of winning, imho."
You can use whatever you like. ;)
Just remember that if the 1960 and 2000 elections had been successfully challenged and reversed, that would be 3/7. And it *does* matter, because an election close enough that it could have been challenged and overturned is one close enough to prove the point that a small lead right now means little.
maybe i am just not a good debate determiner of winners. but how does one technically "win" a debate? i do not see judges that score the candidates as we see in scholastic contests. sure we have focus groups and polls that relate to the debate but is that a good indicator of who is a better debator? as for seeing a bounce from polls after a debate, i think the debate has an influence on those numbers. but again, how much of the current housing, banking & energy crisis (significant events) play into those compared to the debate itself. debates to me are entertainment to see how well candidate handles himself/herself under pressure.
"Most of McCains gaffes have been on foreign policy matters, even in recent weeks. The only thing he is comfortable talking about is being a POW and the surge. Beyond that, he seems equally as clueless on foreign policy as he is on the economy."
Obama's challenge is going to be to keep McCain on the issues, because all the old fart is going to do is repeat his three favorite letters and tell charming-but-irrelevant anecdotes.
Smart people will see through it, but most of them already know he's a fraud. The less informed will think this shows his "expertise".
But they weren't successfully challenged or reversed. We can only go by what happened. If the TV feed for the Reagan/Carter debate had went out, it might be 6/7 :)
I believe we see the same effect at the state polling as well, but we'll just have to wait and see. I'm dying to see someone (Nate? You listening) do a real statistical study on my assertion.
I have been lurking here for a while. I like the site a lot and I will admit that I am an Obama voter. I am 34, married, white and from Iowa. I attended my first Caucus this year and it was great. I have voted in the past for Bush, Perot and Kerry and will vote for Obama this year. I knew a long time ago that Iowa would be an Obama state after witnessing the huge number of first time Caucus goers during the primary. Almost all of them were for Obama.
Right now I work in Iraq and I use my free time to keep up on the stuff going on back in the US. What I don't understand is how the maps at different sites, here, Fox news etc can be so different?
On thing that I will tell you is that McCain wins almost every military vote that I run into over here.
From the "Correlation Without Causation Department"
Since 1988, no candidate going into the debate who was losing has been able to come back and win. 1988 was the first year that the debates have been administered by the Commission on Presidential Debates.
There, I just brought the noise. You bring the funk.
Perhaps we should consider the uniqueness of this election: a relatively more informed electorate, the immediate awareness to the audience that the debates are a means to wade through buckets of media mis and disinformation, and the reliance this time on the "ethos factor." As in the past, people won't be judging the policies, they'll be judging the men--their ability to stand firm in crisis, their honesty, their acumen. But this time, there are multiple sources of information to which they can (if we're lucky?) compare their observations.
"But they weren't successfully challenged or reversed. We can only go by what happened."
That doesn't matter.
The point, again, is that a lead right now in those two elections was not enough to warrant confidence, because the candidate who was leading then arguably won more out of luck than anything else. Definitely the case for Bush in 2000, and even with that one change you are 4/7, hardly anything to crow about.
When did the term ‘town hall meeting’ come into use for candidate appearances? As a New Englander, I find this usage offensive.
The staged Q&A sessions the campaigns call town hall meetings are really press conferences and photo-opportunities.
In New England, local government has deep roots. Almost every piece of ground in New England is part of a city or a town. No one lives outside the town limits. A town meeting is not a discussion group. Historically, it was the local legislative body. Town meetings were held once a year. Every adult resident could vote. They would vote on the annual town budget, by line item.
This form of government still exists in many New England Towns.
Norman Rockwell’s painting for the Four Freedoms series shows a man standing to speak at a town meeting. The document in his jacket pocket that looks like an exam bluebook is the proposed town budget.
http://chawedrosin.wordpress.com/2007/11/23/norman-rockwells-four-freedoms/
The people in the painting are not debating a resolution to support the troops or to have prayers before high school football games. They are voting whether to purchase a new fire truck, or how many miles of local road to be paved this year.
Has McCain used the POW story to the point that it has lost its luster? To the point that repeating it over and over in the debates will not help him and may even hurt him?
"There, I just brought the noise."
Boy did you ever. :0
"That doesn't matter."
Yeah, it kind of does. Because you can call it luck if you want, but they still won, and winning is the only objective measure of who won.
If we start piddling with hypotheticals every time we get data we don't like, we end up being those people who readjust the cross tabs of every poll to make them fit their preconceived expectations.
And besides, does a 4-point lead in most polls at this point count as a "narrow" lead?
"Yeah, it kind of does."
No, it kinda doesn't.
The point, for the last time (because you seem to have a blind spot in this area) is whether or not a small lead at this time is *reason to be confident* about the outcome of the election.
Bush had a small lead at this time in 2000. The election was close enough that it could have gone either way, and came down to luck and legal harrangues. If you think that Bush having had a lead in late September really had anything to do with him eventually winning via the Supreme Court, Floridians voting for Buchanan and so forth -- you're engaging in magical thinking.
I was looking into the "battleground tracking" poll and couldn't find any internals. I was just wondering if anyone knew how this poll could possibly have never shown a single obama lead? Even rassmuten has shown a small yet existant obama lead from time to time.
Eve,I think the POW story could hurt him in the debates. The POW story ahs become the "POW card". Its a joke.
IMHO, whoever wins the debate is the one who makes the LEAST headlines. No gaffes, no missteps, no MSM noise that carries beyond the weekend news cycle.
I think using the POW card more than once in this debate will turn into to a SNL sketch.
The truth is that he is outside of the mainstream in his desire to re-instate the military draft
Is that true? I don't ever remember McCain saying he wants to re-instate the draft.
And I say this not because Rasmussen has it tied, but because he only had Obama +1 during the worst week of the McCain campaign, and yesterday had McCain at 50% in three battleground states. Sorry, ain't buying it, not after 3 weeks of slanted polling for Palin, and the propaganda he's been pushing. Any objectivity was lost as soon as Rasmusseen started pissing and moaning about how Palin was being treated by the media.
If Bush hadn't had the lead at that point, then Gore cutting it close wouldn't have left Gore still short of a decisive win.
Bush led, Gore closed, Bush won. Close or not, you can't change data points because they might have turned out a different way. Especially when we are already dealing with tiny sample sizes.
I hate to be a party pooper but the ABC/WaPo poll is listed as +8 in VA. It should be +3 Obama.
I hate to be a party pooper but the ABC/WaPo poll is listed as +8 in VA. It should be +3 Obama.
As Nate has stated, he uses the RV model at this point in the race, and Obama leads +8 among registered voters when third parties are included.
I think this years debates do matter. I think in a sense both candidates are in a similar position to Reagan in 1980. (and probably for that matter a lot of the candidates who have done well in the debates.) The question people will have in there mind is 'can this guy be President' when looking at both. I think that that is the key to this election now. Which of the candidates can be imagined as President.
I think the interesting thing going forward in this election is that neither candidate has an inherent lead on that. There is no sitting President or vice president running, no-one with a phony connection to doing the job (like George W Bush had in 2000).
I think what Obama needs on Friday is a confident, intelligent performance. Thoughtful but decisive.
Rasmussen can't get out of the trap. He polls the same questions twice. First time he asks party preference (not ID), second he asks voting intentions. Since there is a timelapse between the two polls HIS MODEL STILL CARRIES THE POST GOP CONVENTION BOUNCE.
Debates are going to be watched by the "boxing match" set who already have a horse in the race. It is more about "gotcha" to this crowd than it is the consideration of issues.
That is why the numbers do not move. The undecideds rarely watch debates. They want the drive thru version and the news at 11 will condense such nonsense for them.
I am anticipating there will be a decidedly domestic slant to this foreign policy campaign. Either the questions or Obama will bring it around to how it affects us at home. Truth be told, McCain has actually made more foreign policy gaffes than domestic policy gaffes this campaign.
Sure, we will see the phrase "the economy is fundamentally strong" a million times over the next month. But his reflexive use of "Czechoslovakia", insisting there is an Iraq/Pakistan border, and concurring with the audience member that a draft may be needed would be huge in a year where the economy was not such an overwhelming #1 issue.
I must have missed that. Thanks Scott. Just trying to keep things honest.
Another batch of ARG state polls today:
Arkansas
M: 53
O: 41
Massachusetts
O: 55
M: 39
Oregon
O: 52
M: 41
Pennsylvania
O: 50
M: 46
Vermont
O: 58
M: 38
@ Orin - Welcome. I'm going to try to explain why the maps are different.
Every pollster has different rules that they assemble into an algorithm for measuring the race. Some of these rules are outdated. Some partially reflect voter registration data and some assumptions are made as to who actually votes from the pool of registered voters. Because each of these assumptions by each pollster are very different, they end up with different results.
I like this site the best because from what I have seen of the algorithm, this site best reflects the voting public on Election Day 2008. It's not perfect, but I agree with its assumptions more than any other pollster out there.
Hope that makes sense.
Rasmussen has completely lost ALL credibility. Even his state polls are completely irrelevant. He is no better than Zogby, and should be treated as the shill that he is.
There will be no Joe Lieberman standing behind McCain to tap him on the shoulder and correct him this time.
Yes it's ARG, but it's still good to see Obama at 50 in PA.
"Rasmussen has completely lost ALL credibility. Even his state polls are completely irrelevant. He is no better than Zogby, and should be treated as the shill that he is."
Unless he shows Obama ahead.
Surely the significance of the above data is that things don't move very much.
So, assume that Obama is ahead now, there are 4 possibilities:
1)Obama stretches his lead = Obama Landslide
2)Obama maintains his lead = Obama narrow win
3)Race tightens to a draw = 50/50
4)McCain gets a lead = McCain narrow win.
Overall then that's 2.5:1.5 in favour of an Obama win, or around a 60% chance of his winning.
But of the 4, option 4 is the least likely as McCain would need to move the polls significantly and consistently and the data says that doesn't often happen. So it has to fall even more in Obama's favour than 60%...
Great site, compulsive viewing.
My perspective from across the pond is that IF Obama comes out of his shell, he'll win the debates and the election.
But he needs to find his authentic voice.
See:
http://www.huffingtonpost.com/drew-westen/the-day-the-momentum-chan_b_128227.html
"If Bush hadn't had the lead at that point, then Gore cutting it close wouldn't have left Gore still short of a decisive win."
Yes. So? The point isn't whether the guy trailing can score a decisive win, but whether the guy *leading* does.
Or put another way:
Of the 7 times since WW2 that a candidate had a lead similar to Obama's at this point in the game, *not once* did the leader hold onto that lead through the rest of the season for a clear win.
Looks like PA is moving back towards Obama.
The Palin bump played pretty well there, and I suspect it will dissolve with Palin's favorability.
Ho-hum, another PA poll, another Obama lead...
Sorry Scott, I don't care if Rasmussen ever has Obama ahead again. Obama could be ahead by 15 points in Rasmussen polling and i wouldn't give a flying fuck. Rasmussen is done.
All of those ARG polls very close to Nate's current projections.
Anyone want a Good Laugh??
-- Check out today's Battleground Tracking NATIONAL poll
-- McCain is up by 2!
ha, ha, ha, ha, ha, ha, ha, ha, ha
ha, ha, ha, ha, ha
ho, ho, ho, ho, ho, ho, ho,
Best ever!!
Maybe they will do a spoof on this on this Sat night's SNL!!!
ha, ha, ha, ha, ha
Rasmussen is at least consistent, a pollster offering hope and faith to the conservative flock whether directly or through helping to impart the "right" impression in FoxNews polls.
However, the small sample sizes, and the "overnight" character of his state polls add more volatility than is probably reasonable, and I don't put much stock in those as a measure of reality at that level.
"Yes. So? The point isn't whether the guy trailing can score a decisive win, but whether the guy *leading* does."
Decisive is a semantic point, but my point is whether that guy who led won. And the guy who led, won.
To reiterate, no debate will be "gaffe-free" for Obama. If he makes it through without fumbling a fact or a sentence (unlikely, but possible) Republicans will just pick one of his phrases out of a hat and declare it to be offensive. It worked in 2004.
In the best case scenario for Obama, the three storylines that come out of the foreign policy debate, for example, will be
1. "More undecideds liked Obama's performance."
2. "Critics attack Obama's mention of a 'new birth of freedom' for America. Was it unpatriotic, or just poorly chosen phrasing?"
3. "McCain camp says by referring to Czechoslovakia, the Senator was calling for re-unification of the Czech Republic and Slovakia. An Obama surrogate called such a position 'foolish'."
That's the best case scenario, and it's not a clear win.
scott wrote:
Is that true? I don't ever remember McCain saying he wants to re-instate the draft.
McCain on the military draft.
He "doesn't disagree" with the re-instatement of the military draft.
QUESTIONER: If we don’t reenact the draft, I don’t think we’ll have anyone to chase Bin Laden to the gates of hell.
[Applause]
MCCAIN: Ma’am, let me say that I don’t disagree with anything you said.
Connect the dots.
@ Quantman - Wow. So I wanted to dig a little deeper to see how Battleground/Tarrance got their numbers. I got broken links and redirects instead.
Thing about Rasmussen... Do they follow some pattern as to when they shift their party ID numbers? Does this happen monthly/bi-monthly or what? If I remember, they did one either during the RNC or right after it. Wouldn't he have taken the convention surge into the party identification?
The thing is though that Obama is winning right now and the polls seem to keep resetting to Obama +4. McCain needs a game changer in the debates, so it's not that Obama has to have a clear victory as much as he has to not have a clear defeat.
Obama can lose the news cycle on Saturday; he just needs something that doesn't cause him to still have lost it a week from Saturday.
I have faith that he'll be able to do that.
"Decisive is a semantic point, but my point is whether that guy who led won. And the guy who led, won."
Here's what you sound like.
Two guys are at a Yankees-RedSox game. The Yankees are up 4-2 in the 7th inning, and the Yankees fan says "We're in great shape! 5 out of 7 times that we lead this late, we win!" The RedSox fan scoffs, knowing a 2-run lead is certainly an advantage in the 7th inning but there's plenty more game to go.
The score remains 4-2 into the bottom of the 9th. The Sox get two guys on and their cleanup hitter hits a towering shot to right field. Home run! The Sox win. He rounds the bases and the crowd cheers.
But the first base umpire says he missed the bag. The Yankees toss the ball to first, the guy is out and the Yankees go on to win. Later replays show he actually *did* touch the bag and the umpire was wrong.
The Yankees fan goes, "see, I told you we had it in the bag!"
There has yet to have been an NBC/Mason-Dixon poll that HASN'T shown Obama with at least a 2 point lead, so I will wait for other Florida polls. I find it highly fishy in those polls that McCain is only up 4 among seniors, not up by at least 20 in the panhandle and losing in Tampa Bay-St. Pete
Hi there Real Mike is Back:
Don't waste your time trying to figure out Battleground Tracking poll.
Check OUT all their polls!!
You will fall out of your chair laughing!!!
They should charge for the entertainment!!!
Ha, ha, ha, ha, ha, ha, ha
Battleground Tracking poll is the ALL-TIME BEST!
We should all get together and get them a special award!
I think Sarh Palin's technology and math is the intelligence behind their polls!!
Anyone notice how small Obama's lead is in the 18-34 demographic in the Quinnipiac/WSJ polls today?
Among 18-34 yrs old:
Colorado:
Obama 48, McCain 48
Michigan:
Obama 51, McCain 48
Minnesoata:
Obama 51, McCain 43
Wisconsin:
Obama 55, McCain 38
Is this because there is such a small sample. I have trouble believing 48% of the youth will go to McCain in ANY battleground state.
matthew: Yeah, that's a bit odd. So the numbers are probably better for Obama than the polls suggest, especially in Colorado...
ARG's +4 O in PA sounds about right... too low for my comfort, though.
"Connect the dots."
You may as well have asked him to perform brain surgery. Connecting the dots is just not the strong suit of most republicans.
In your scenario, the possibility of a near-comeback *was already factored into the odds of winning*.
According to fangraphs.com, the road team leading by 2 in a baseball game with the home team about to bat in the seventh will win 76.9% of the time.
You are the Yankees fan who says "That's not the real odds" and then tries to use the near-comeback as proof.
The 76.9% wins *included* the times one team almost came back but failed.
Michael said...
Don't forget this announcement from Ras:
"Results from this past week showed that the number of people considering themselves to Democrats spiked early in the week as the economic problems on Wall Street became visible. Overall, it was the best week for the Democrats since July. It remains to be seen whether this might lead to a lasting adjustment or be more of a bounce like that resulting from a party’s nominating convention."
His party IDs still use an average much less favorable to D's. When he makes announcements like this, he's telling us that the recently favored party is actually polling MUCH better than the ID-weighted numbers show. Presumably BO is +3 or so. We saw exactly the same process in the other direction when Mc's spike didn't show up much at first in Ras, and a similar announcement came out.
This was the problem with Rasmussen changing the PID pre-maturely. Now they are over weight to McCain and Scotty doesn't know whether to switch them to the "NEW" pro Obama figures or keep the numbers he hastily switched to McCain after the conventions. Compromising your integrity Scotty.
It's probabley Obama +2 in Rasmussen right now if Rasmussen were using real PID.
"The 76.9% wins *included* the times one team almost came back but failed."
Yes, and what was the sample size?
Yours is *seven*.
That's the main point.
On the subject of Party ID weighting and tracking polls, here's some info that I found interesting and thought others might.
The R2000 tracker gives the following breakdown by party:
Dems, Obama wins 86-11
Reps, McCain wins 88-6
Inds, Obama wins 46-41
Now, if you apply those numbers to Rasmussen Party weights, here's what you get:
Using Ras's current weights, Obama leads 48-45 (+3).
Using the weights Ras was using two weeks ago, Obama leads 49-44 (+5).
Ras in their article on Sunday giving their new weights said that last week's Dem number was the highest it's been since July. Using Ras's July weights, Obama leads 49-44 (same as two weeks ago). Using Ras's June weights (41% Dem, 31.5% Rep), Obama leads 50-44 (+6).
R2000 is reporting a 49-43 Obama lead, so obviously they're using Party ID numbers consistent with Ras's June. Question: Are the R2000 results weighted by Party ID, or is Party ID organic to their poll?
"Yours is *seven*."
It started at 17, and you picked an arbitrary definition of "close" to decide what should count and what shouldn't.
And I've specifically said my sample is not meant to be conclusive statistical evidence. It is simply corroborating evidence to a phenomenon I think exists.
@ The Real Mike is Back:
Thanks for taking the time to respond to me. I like this site as well because Nate at least tries to use more than just one poll. Is Fox JUST using that Rasmussen poll??
"Anyone notice how small Obama's lead is in the 18-34 demographic in the Quinnipiac/WSJ polls today?
Among 18-34 yrs old:
Colorado:
Obama 48, McCain 48
Michigan:
Obama 51, McCain 48
Minnesoata:
Obama 51, McCain 43
Wisconsin:
Obama 55, McCain 38
Is this because there is such a small sample. I have trouble believing 48% of the youth will go to McCain in ANY battleground state."
Remember, Colorado has a LOT of military - the military rank and file aren't as strongly Republican as in past elections, but that's still a lot of young votes for McCain. Michigan is kinda odd, I agree, but everything else seems legit.
Let's use statistics to build a model of the Average Undecided Voter. He's in his late 30's and considers himself "conservative" though he occasionally votes for Dems. He attends church at least once a month, believes the country is heading in the wrong direction and has seen his net worth erode in the past year. He wants the Iraq war to end, but doesn't want the military to be dissed. He didn't really start paying attention to this election until the conventions.
He also didn't pay much attention to the primaries. He has a firm impression of John McCain as a war hero and straight shooter, but worries about hsi age. He pictures Obama as a quasi-American, a sort of exotic African gazelle who eats arugula, talks in five-syllable words and wants to take people's guns away but is very, very smart.
He strongly thinks the country needs to change. He's voted Republican in past presidential elections but knows they've been careless with the nation's economy.
He's UNDECIDED.
What does this guy need to see from Obama in the debates?
I think what he needs most is reassurance. He needs to see Obama as a down-to-earth, believable person who talks like everybody else, laughs and smiles, seems to connect with ordinary folks. Everybody knows Obama is smart. This is one time where it's more important for him to appear SOLID.
If Mr Undecided sees that Obama (the one many of us are familar with, but the average voter is not) his vote will be won.
The 48-48 for CO is not believable. Even Colorado Springs has plenty of young liberals at Col.College.
Wisconsin may be about right; even Badgerland does have College Republicans.
How can Rasmussen have it as a tie. That guy is on the Republican payroll. Let's just ignore his polls and use the ones from Kos.
election is over
McCain LANDSLIDE !!!
say it with me liberals
McCain L A N D S L I D E !!!
"How can Rasmussen have it as a tie. That guy is on the Republican payroll. Let's just ignore his polls and use the ones from Kos."
Let's use both. The more data points, the better.
"The 48-48 for CO is not believable. Even Colorado Springs has plenty of young liberals at Col.College."
Sure, but to be fair, that's just one small subset of the data that was cherry-picked as being biased toward McCain. It's probably just as easy to find some small subset that's biased toward Obama. These things are likely to balance out.
The bottom line for the poll is Obama winning in Colorado by 4. That seems about right, doesn't it?
Agree on the O+4 overall; it's likely some of the other stats were skewed the other way.
As far as the 51-48 for Michigan: I didn't know Hillsdale College was that large. :)
You have to wonder about Nate and some of his assertions, like the race is pretty much set in stone. I am sure he would like to believ that given his support for Obama, ut once again, partisan ship should not cloud reasoning.
An actual pollster, Scott Rasmussen had this to say today:"With the race so close, the debates scheduled to begin this Friday night could be more significant than usual. That’s especially true since 18% of voters say they are either uncommitted or could change their mind before voting."
Nate, man, ya gotta stop making things up. This site is starting to look and feel like a train wreck and you have as much credibility as your posters!
This Rasmussen poll scares me a little. How is McCain anywhere close to Obama right now? Do the stupid voters realize that McCain is singlehandedly responsible for every economic problem we are currently experiencing. At least the Quinnipiac poll shows what I want to see. Although they always seem to be biased in Obama's favor.
For what it's worth, I think Virginia is the bellweather this time. Sounds strange. No Dem since 1964. The polls there have consistently been in line with the trackers. I think we can look at Virginia and know who won or is going to win. I don't think we can say that with any other state. The other 49+1 all seem to lean blue or red to me.
petekent: If the site is starting to turn into a train wreck, what are you doing hopping around in the carnage? Go somewhere else.
I think the debates will play a bigger factor this year than in past election years.
First of all, you have two candidates not affiliated with the current administration for the first time in forever.
Second, Obama is still an unknown politically on the national scene and people want to see if he can debate and just isn't speeches and appearance. Same goes with Sarah Palin.
Third, there is still a large amount of undecideds, and if anything this is going to be the event to help them mobilize.
I think debates will play a larger role than ever this year.
koz: I think today Obama's lead nationally will settle into 2.5% or so. So I think he'll probably lose a point on Gallup as well.
Charles M. Kozierok said...
This Rasmussen poll scares me a little. How is McCain anywhere close to Obama right now? Do the stupid voters realize that McCain is singlehandedly responsible for every economic problem we are currently experiencing. At least the Quinnipiac poll shows what I want to see. Although they always seem to be biased in Obama's favor.
Rasmussen consistently shows a republcian bias relative to the others. Either almost all of the other polling is wrong or they are. It's possible that Ras is right and everyone else is wrong, but I doubt. It's even reflected in their state polling. At any given time if you changed their state poll to D+2, it's about what you'd guess another polling outfit's number would be.
Pete Kent says: partisan ship should not cloud reasoning.
Pete Kent says: man, ya gotta stop making things up.
In other news... End times arrive. Sarah Palin raptured into heaven in clouds of glory. Sinners flee to Alaska for sanctuary. Film at 11.
Today's polls are MARVELOUSLY MOVING!!! For John McCain!!!
Here's a worrisome bit of news for all the Dem partisans out there. it's from the new Yahoo/AP poll (the same folks who brought us the news on the Bradely Effect earlier in the week:
"Barack Obama's support from backers of Hillary Rodham Clinton is stuck smack where it was in June, a poll showed Tuesday, a stunning lack of progress that is weakening him with members of the Democratic Party in the close presidential race."
Mmmmm . . . seems like a ceiling to me.
Of course none of that should be surprising. Hillary won big in conservative rural areas. Those folks were just parking their votes with her and are never going to go for Obama. That is the reason why PA is such a problem for him, why OH won't tip and WV and KY are safe.
The same poll makes the point that Palin has not really made traction with women voters generally. Should seh catch on there, even in a modest way, it will be lights out Obama!
Palin has a great deal of upside after the unrelenting drubbing she has taken in the media.
Her appearances at the UN this week should help.
@ Orin - Yes, I do believe Fox sponsors Rasmussen only. Fox also has links to Gallup and AP, too. But that is no different that the NBC/Wall Street Journal poll or the CNN poll or ABC/USA Today. They have one model/algorithm that they trust and they run with it.
The Diageo/Hotline poll only polls 300 people nationally each day, but last night it seems they had it about 51-39 for Obama, give or take a couple points. He won there polling by somewhere between 8-16 points last night.
Hopefully you people of subhuman intelligence understood my baseball analogy which I used to "dumb it down" for you. You see, progressives like myself are more intelligent than the voting public in general. I use my higher level of intelligence to try to bring better understanding of what these polls mean to the dumb masses and how they prove that Obama is the savior of mankind. It's hard being so superior to you but somehow I manage.
"Pete Kent says: partisan ship"
That sounds like a ship Pete sailed on long ago!
" I think we can look at Virginia and know who won or is going to win. I don't think we can say that with any other state. The other 49+1 all seem to lean blue or red to me."
Just out of curiousity, which way do you see Colorado leaning, because if it's blue, then Obama can win without Virginia (assuming all states go the way you see them leaning, of course - I'm assuming you see PA/MI/MN/WI/IA/NM/NH all leaning blue)? I agree that, given the current state of the race, VA is the closest thing to a pure tossup, but, as I said, the current state of the race has Obama leading in enough places to win the election without VA.
"How can Rasmussen have it as a tie. That guy is on the Republican payroll. Let's just ignore his polls and use the ones from Kos."
Let's use both. The more data points, the better.<<<<<<
I think that is correct. I take every National Poll that exists (no interactives) - throw out the top and bottom , average the rest. Not perfect , but a good approximation.
The "Charles Koz" poster on here today is a sockpuppet.
"Barack Obama's support from backers of Hillary Rodham Clinton is stuck smack where it was in June, a poll showed Tuesday"
Look at the Super Tracker, Pete. June was Obama's best month of polling. If he polls in November like he polled in June he wins fairly easily.
A comment upthread made a very important point: the winner of the debate is whoever the MSM says won the debate, and the MSM are unduly influenced by the likes of SNL and the late shows.
SNL is the best explanation I've seen yet for why the media decided, several days after the fact, that Gore had lost the first debate. And of course the "can I get you a pillow" sketch from earlier this year undoubtedly shaped primary coverage in the 2nd half of the Dem campaign.
I don't think complaining about the refs gets you very far, but Obama (and McCain as well) need to know who the VIPs in the audience are. Their take will be accepted by all of the people who don't watch the debate, and sadly, even many who do.
Of course Pete Kent chose to quote Scott Rasmussen, over Nate who obviously is far ahead of Scotty when it comes to electoral prediction.
Pete, Scotty is a Hannity loving member of the GOP. While I believe he tries to keep his numbers legit, his commentary certainly is not. He broke his own rules in switching the PID numbers mid month after the RNC convention, but now refuses to change them back when the evidence is so clear. How do you reconcile this?
Diageo/Hotline/FD tracker up on RCP:
47-43 Obama. McCain gains a point.
tomthress said... Just out of curiousity, which way do you see Colorado leaning, because if it's blue, then Obama can win without Virginia (assuming all states go the way you see them leaning, of course - I'm assuming you see PA/MI/MN/WI/IA/NM/NH all leaning blue)? I agree that, given the current state of the race, VA is the closest thing to a pure tossup, but, as I said, the current state of the race has Obama leading in enough places to win the election without VA.
light, light blue. Probably the closest blue state to the center, either that or New Hampshire.
As I mentioned Rasmussen has a +2 or so bias in favor of Republicans. I've looked at CNN, Quinnipiac/WSJ, and many of the other pollsters and they seem to be about +5 in favor of Obama. The Daily Kos/Research 2000 seems to be about +8 Obama due to the party affiliation numbers that they use. So right now Rasmussen is probably the closest although in my heart I know that Obama is leading.
petekent--"you have as much credibility as your posters!"
Did you just accuse yourself of having no credibility?
realistxxx said...
Diageo/Hotline/FD tracker up on RCP:
47-43 Obama. McCain gains a point.
you're right, bad info on my part, sorry!
@filistro,
"Let's use statistics to build a model of the Average Undecided Voter. He's in his late 30's and considers himself "conservative" though he occasionally votes for Dems."
Great idea -- why is it a "he"? Why doesn't he consider himself "moderate", because if he has actuall voted for some Dems and some Pubs, he is likely in the middle?
SERIOUSLY, is there any "model of the average undecided voter" defined by any blog, or political pundit?
Puppet Charles also posted as "charlice" above (same blogger ID).
Two gems from PeteKent:
"Palin has a great deal of upside"
"Her appearances at the UN this week should help"
I worship at the alter of petey.
Koz:
the reason why Obama is not doimng better is b/c he fails on the leadership and expereice dimensions. People might like his positions better, but they don't trust him to carry them out and don't believe that he would be a good Commander in Chief, whether of the army or the economy.
This is a problem you have when you nominate a nobody as your candidate.
Obama has most likely hit a ceilling that he is incapable of breaking. McCain would have to implode in order to lose, like Gerald Ford did in 1976. It's possible, but not likely.
Unless Obama hits a homerun in the debate, he will have hit his high water mark in the polls. Too bad it is 48% and not 51%.
I automatically discount Scotty Ras because his polls despite their history of prior accuracy don't benefit my candidate. I much prefer to cherry pick those polls like DK/R2000 that show Obama with a comfortable lead of 20 points or so. On election day, I'm hoping the exit polling will show Obama ahead like it showed Kerry with a sizeable lead in 2004 so that we can supress Republican turnout as much as possible, as we did in the FL panhandle in 2000.
It's not just that Virginia is a pure toss-up. It's that it seems to move directly with the national polls. Today it's probably at Obama +3 or 4, the nation is at the same place. 10 days ago, it was probably at McCain +3 or 4, same as the nation.
Let's see: JM gains a point on Ras, a point on Diageo, no point on R2000, - I'd say he's due to gain a point on Gallup, but experience shows that at least one of the four national polls will always do something completely off-the-wall. So while logicaly, I expect McCain to gain a point or two on Gallup, no result would surprise me: a tie or O +5.
Actually KTH, the media decalred Gore the winner of the fist Bush degbate and then the polls showed the public felt the reverse was true.
In this year, more so than ever before, the people will discount anything pro-Obama the media says. They recognize the bias and discount it. Indeed there is likely to be a backlash from the liberal view expressed by the chattering classes.
@filistro said...
The "Charles Koz" poster on here today is a sockpuppet.
I thought so also, After you listen to Charles a while, you can tell when the comments are off the standard, At least most of the time
"The Daily Kos/Research 2000 seems to be about +8 Obama due to the party affiliation numbers that they use."
Does R2000 weight their results by Party ID, or are the Party ID numbers that they report just whatever their respondents happen to tell them? Either way, their Party IDs are consistent with what Rasmussen says he saw last week - most Dems since July (R2000 shows about 5% fewer Dems and 5% fewer Reps than Ras was using as weights in June).
The party ID weighting is pushing Obama's lead down by 3 points in Rasmussen if this week's party ID numbers are matching last week's. That may be right, but it's worth understanding if you're comparing Rasmussen to polls that either don't weight by Party ID or that use different weights.
Gaining or losing a point could just be random MOE shifts too.
I'll officially start worrying when the majority of polls show McCain leading or one poll shows him outside of the margin of error.
...and yes, I was worried during the Palin bump...
Matt JH:
You are fool. Rasmussen is a POLLSTER. Nate is a baseball analyst. One does this thing for a living, the other has made a hobby of it.
I scratch my head at the assertions that Rasmussen's numbers aren't credible. No one has been closer to the mark in the the last two presidential elections.
I understand there's a lot of speculating about the reasons that Dem vote might be undersampled, party ID, cell phones, GOTV, no Bradley, etc., but the numbers in support of those arguments are far from decisive.
Anyone who's been watching these things over the years recognizes that there's aways been a tendency to underestimate the Republican performance in close elections, and to overestimate Dem performance in GOTV and
Surely, the Obama GOTV effort this year is bigger and more focused than historically, but history indicates less effect than anticipated. Plus, the aggressiveness and vitriolic style this time could produce a backlash response, which may be showing up already in the pollling numbers for the youth vote.
The tendency around here (starting at the top) seems to be to latch onto any evidence in support of preconceptions, no matter how tortured the numbers must be to make them dance, and reject any evidence that doesn't fit the desired outcome.
We can only look at history as the best guide for the future. In this kind of election, virtually all of the mainstream pollsters have had pronounced error in favor of Dems. So, be careful before asserting that Rasmussen is the shaded outlier just because the usual suspects are all out in left field.
petey, petey
Obama is not a nobody. He is a very gifted speaker who has experience as a community organizer register voters of all shapes and sizes and various levels of alive or dead for the Chicago political machine. He attended a very well-respected church in the Chicago area and he has written two books describing his amazing story. If you are one of those ignoramuses who think Palin's two years of executive experience make her more qualified to serve as President than Obama then you clearly don't understand the size and scope of his campaign. It is frickin huge and is a multi-million dollar operation. This guy knows executive experience better than anyone.
George Will has had ENOUGH!
http://www.realclearpolitics.com/articles/2008/09/mccains_temperment_fails_again.html
The rats are deserting the sinking SS McCain.
Imposter is back, so I guess I'll have to go back to using my signature. Thanks for the excuse to advertise my blog! :)
--
Charles M. Kozierok
If you see someone post something stupid here using my name, it’s an imposter. Click the name to go to the person’s profile, and check the blogger ID in the address line: mine is #10019303035321620019, and my profile contains a link to my blog, AABW.
Vern:
You will be crushed (like Obama) when you finally realize that Governor Palin is not Dan Quayle!
New crowd estimates for Palin in Orlando now take the number up to 70,000.
petekent: Of course she's not Dan Quayle. Dan Quayle got elected. :)
Fla GOP... when I built my Average Undecided Voter, I went with simple majorities from various studies. I won't supply links to all of them but you can easily find them through Google.
There is no composite that I know of (though mine is available for free :-) but the average American considers himself "conservative" rather than "liberal." Church attendance in America avarges once per month per capita. More men than women are undecided in this election. Etc etc etc... (That's a quote from Yul Brynner ;-)
petekent: Did a "fire marshall" give those estimates, or did the McCain campaign use a random number generator again?
PeteKent, though you're obviously one of the top 2 or 3 most biased Republican posters on here, you have a point. Obama has a ceiling due to some not willing to take a chance because they think he's too liberal, inexperience, or Black. That ceiling when the undecideds are pushed on election day is probably 52-53% if he does everthing perfect, but really it's probably more like 51% as a reasonable expectation. Figure 2-3% for third party candidates, Obama's upside is plenty. In my opinion, though McCain fits the mold, noone is particularly excited to vote for him and his ceiling has appeared to be substantially lower. This election is a referendum on Obama, but I think the electorate on the whole is more likely to give him the benfit of the doubt than you do. Especially given the alternative.
"New crowd estimates for Palin in Orlando now take the number up to 70,000"
as sworn to by sean hannity on a stack of korans.
Do you think what we are seeing is the dissapation of the economy crisis bump that Obama got last week?
Perhaps if he had shwon real leadersip and actually decalred his position, he might have been able to consolidate his lead. Now it will fritter away back to a tie by Friday.
This electorate is looking for an excuse not to vote for Obama. Palin proved that. The debate might seal the deal.
McCain supporters:
What do you think of George Will's conservative credentials and his recent thoughts on John McCain?
PeteKent said...
Koz:
the reason why Obama is not doimng better is b/c he fails on the leadership and expereice dimensions. People might like his positions better, but they don't trust him to carry them out and don't believe that he would be a good Commander in Chief, whether of the army or the economy.
This is a problem you have when you nominate a nobody as your candidate.
Obama has most likely hit a ceilling that he is incapable of breaking. McCain would have to implode in order to lose, like Gerald Ford did in 1976. It's possible, but not likely.
Unless Obama hits a homerun in the debate, he will have hit his high water mark in the polls. Too bad it is 48% and not 51%.
----------------------
Pete, Pete, Pete.
This is below your intelligence. Usually your commentary is at least amusing. This is just lame. We know Obama loses 2-3 points because of race. We know Palin hit her peak and now is nothing more than an unqualified shiny object for the GOP base.
Obama has hit his ceiling? Really?
Desperation Pete, get back on your talking points, you seem to be cracking under the pressure of the Obama machine rolling over McCain's incompetent ass.
McCain has to hold IN,VA,CO,FL,OH. Obama is surging in Virginia and Colorado and McCain's only recourse is to try and win a state (PA) he has not had a single poll go his way ever, besides Zogby interactive.
Its bleak. The country is seeing it's new leader with his steady leadership in the face of the economic crises while McCain flails about with his advisers lashing out in anger at the NYT's and the media. It's all coming apart now for McCain. His attacks are falling on deaf ears while Obama makes McCain live the failures of the past eight years. Obama has solidified his strength as "Change" while McCain is seen as "Bush". It's over.
rudy
The reason we disparage Scotty R's polls is because this is a left-leaning site that wants more than anything to see Barack Hussein Obama (yes, I'm proud of his middle name and I will tell it to anyone who asks) in the white house. He is the lightbearer sent from God as Nancy Pelosi said and he is the single most transformative candidate to ever appear on the national stage. He is far better than Reagan, Lincoln, FDR, Jefferson or Washington. He has a profound level of experience and has not lost his message of "Hopey-Changy-Happiness" despite all the unfair attacks leveled at him by Karl Rove and the Rethugs. I don't care how much he raises my taxes, destroys our foreign policy like my other hero Carter, or how unprepared many Americans think he is. He is the single greatest thing that has ever happened to me.
Koz;
You overplayed your hand to today and I fell for it!
Sock Puppet.
There is one who apes me, but I am inimitable so he or she is clearly seen through!
"axmxz" @ September 23, 2008 10:33 AM - LOL, my first impersonation - yey!
Seriously, isn't there a way to secure our IDs or something? Last couple of days it's gotten to the point where you don't know who the fuck you're talking to anymore.
George Will is one rat who has never been on the McCain ship. He is dirt to me!
Look at Nate. Look at George Will.
Mccain has alienated the baseball guys.
In America...that means it's all over.
Post a Comment