Comments on FiveThirtyEight: Politics Done Right: Bad Math and the Bradley Effect

艾葳酒店經紀公司提供專業的酒店經紀, 酒店上班小姐,八大行業,酒店兼職,傳播妹,或者想要打工兼差、打...

2009-09-23T01:39:51.992-04:00

艾葳酒店經紀公司提供專業的酒店經紀, 酒店上班小姐,八大行業,酒店兼職,傳播妹,或者想要打工兼差、打工,兼差,八大行業,酒店兼職,想去酒店上班, 日式酒店,便服店,制服酒店,ktv酒店,禮服店,整天穿得水水漂漂的,還是想去制服店當日領上班小姐,水水們如果想要擁有打工工作、晚上兼差工作、兼差打工、假日兼職、兼職工作、酒店兼差、兼差、打工兼差、日領工作、晚上兼差工作、酒店工作、酒店上班、酒店打工、兼職、兼差、兼差工作、酒店上班等,想了解酒店相關工作和特種行業內容,想兼職工作日領、假日兼職、兼差打工、或晚班兼職想擁有鋼琴酒吧又有保障的工作嗎???又可以現領請找專業又有保障的艾葳酒店經紀公司!

艾葳酒店經紀是合法的公司工作環境高雅時尚，無業績壓力，無脫秀無喝酒壓力，高層次會員制客源，工作輕鬆，可日領、現領。
一般的酒店經紀只會在水水們第一次上班和領薪水時出現而已，對水水們的上班安全一點保障都沒有！艾葳酒店經紀公司的水水們上班時全程媽咪作陪，不需擔心！只提供最優質的酒店上班,酒店上班,酒店打工環境、上班條件給水水們。心動嗎!? 趕快來填寫你的酒店上班履歷表

水水們妳有缺現領、想要兼職、有缺錢的煩腦嗎?想到日本留學缺錢嗎?妳是傳播妹??想要擁有高時薪又輕鬆的賺錢,酒店和,假日打工,假日兼職賺錢的機會嗎??想實現夢想卻又缺錢沒錢嗎!??
艾葳酒店台北酒店經紀招兵買馬!!徵專業的酒店打工,想要去酒店的水水,想要短期日領,酒店日領,禮服酒店,制服店,酒店經紀,ktv酒店,便服店,酒店工作,禮服店,酒店小姐,酒店經紀人,
等相關服務幫您快速的實現您的夢想~!!

^^ very nice 徵信, 徵信社, 徵信, 徵信社, 徵信, 徵信社,徵信, 徵信社, 徵...

2009-03-12T02:06:00.001-04:00

^^ nice blog!! ^@^徵信, 徵信網, 徵信社, 徵信社, 徵信社, 徵...

2009-03-12T02:06:00.000-04:00

情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,...

2009-01-19T09:08:00.000-05:00

Let's call the Bradley effect what it is...racism....

2008-09-29T10:30:00.000-04:00

Let's call the Bradley effect what it is...racism. Worse than racism...closet racism. Racism among people who know it's wrong and won't admit it.

At least overt racists have the courage of their convictions. They are full of hate and proud of it. But closet racists are not only racist...they are hypocritical and dishonest. They pretend to be enlightened, but follow their baser instincts. I have nothing but contempt for them.

What we need to take into account, to be realistic...

2008-09-21T18:41:00.000-04:00

What we need to take into account, to be realistic, is the Jeb Bush Effect, which is state and local officials preventing black people and poor people from voting, on the assumption that they will vote for the Democratic candidate.

I have lived in Maryland my entire life and I clea...

2008-09-20T20:32:00.000-04:00

I have lived in Maryland my entire life and I clearly remember when Doug Wilder won the gubernatorial election in Virginia by a hair, despite being up considerably in the polls the day before.

I sincerely hope that the Bradley effect is going away, however, I am not entirely convinced. I believe that for every white guy that secretly votes FOR Obama, there are two guys voting against Obama because he's black and these two guys won't even tell their wives.

I'll believe when I see it.

It has been ageneral observation that the Bradley ...

2008-09-20T14:46:00.000-04:00

It has been ageneral observation that the Bradley effect has dissapeared in recent years. Perhaps it has been partially eclipsed by the cell phone effect (see the subsequent post). Calculations of the Bradley effect usually involve comparing polls that don't call cell phones with the election results. If the cell phone effect is included this suggests that the Bradley effect never really dissappeared, it just got concealed by technological change.

As for the "Reverse Bradley Effect" that is probably due to pollster models not capturing the rise in African American turnout this year as previously noted.

It will be interesting to compare the poll averages with the actual returns the day after the election. I suspect that the margin in Pennsylvania will shift for McCain by about 2% more, and that the margin in Virginia will shift for Obama by about 5% more than the RCP averages.

It doesn't have to be that people are not willing ...

2008-09-19T20:38:00.000-04:00

It doesn't have to be that people are not willing to vote for a black man.

There are just as likely people who will tell a pollster that they are supporting Obama because they want to be seen as racially progressive, even though they agree more with McCain's positions. So it's not really racism against a black man, it's more wanting to gain social approval. Racism is really looked down on in our society, and people will do a lot to avoid being considered racists.

Will, that argument is really pushing it. He wasn't saying the Bradley effect is a good thing, just that it will help his candidate win. The British and French were more than happy to benefit from the racism of the Nazis by accepting recruits who were persecuted by the Nazis, that doesn't mean the French and British were a bunch of racists. Cool down on the partisan attacks.

Nate - here is a study from a Harvard prof that Ma...

2008-09-19T18:00:00.000-04:00

Nate - here is a study from a Harvard prof that Marc Ambinder is posting on ..
http://people.iq.harvard.edu/~dhopkins/wilder13.pdf

"It's Wisconsin and Michigan in particular that, w...

2008-09-19T17:01:00.000-04:00

"It's Wisconsin and Michigan in particular that, when I've talked to strategists on both sides, that's where they're nervous. It's those--they're sitting in undecided. These are folks that if you probe them on issues, they tell you they--the country's moving in the wrong direction, they tell you that the economy stinks, they'd like to see a little more money in their pocket. They're voting for Democrats for Congress, they're voting for Democrats for Senate. And then you ask them about the presidential race, and they say, "I don't know yet. I'm undecided." They don't tell you why they're undecided. And it's that voter--it--Obama's got a little magic number that I think people need to start watching in these states. On the Sunday before the election, he better be at 48 or above. Anything less than that--because he's going to lose 70 percent of the undecideds. I think the McCain folks know this; I think the Obama folks know this. So the key now is to get his numbers to 48 or above."

Chuck Todd
NBC News
Meet the Press
September 14, 2008

@Pacific moderateI Was originally from Fresno Cali...

2008-09-19T16:52:00.000-04:00

@Pacific moderate
I Was originally from Fresno California area before i moved to Central Missouri and i can say yes there are still some that will not vote for Obama just because he is black. I will also be happy the day the Bradley effect is completely gone who cares if the best person for president is old, young, black, woman or Hispanic.
I also think that the Bradley will not be present in some states and in others it will be there.

cangratulations to president-elect mccain!!! said....

2008-09-19T16:29:00.000-04:00

cangratulations to president-elect mccain!!! said...

Idiotic, is that you?!

I wonder if the Bradley effect might not pop up in...

2008-09-19T16:17:00.000-04:00

I wonder if the Bradley effect might not pop up inconsistently in post election data, showing in some states but not others? But I also think that applies to other X factors, such as the youth vote, or the Palin effect.

I have been following Presidential elections relatively closely for 20 years, and never have I seen such an inconsistent set of polls in terms of the ups and downs inside each state.

Slightly worrying thought for McCainiacs. Maybe the late conventions have pushed the 'point of no return' moment (when the leader is most likely going to win) back a couple of weeks. Till about now coincidentally. It is becoming harder to imagine a scenario in which McCain wins now. (Maybe disaster in Iraq, maybe OBL being captured, maybe stellar debate performances??) But I am not sure any of those are either possible or actually going to turn the election in McCain's favor.

Michael said... "Come election night, you guys wil...

2008-09-19T16:13:00.000-04:00

Michael said...
"Come election night, you guys will all be shocked to see at the results. The Bradley Effect is real and will cost Obama the election.
Maybe in some states it won't be a factor, but it will be a factor in the midwestern states of OH-MI-WI-MN-IN.
Nate, you can try to denounce it, but it is true. He got ripped to shreads in OH-PA, because of race. Polls may give him the edge in PA, but it's closer than all you think."

I've heard and seen some deplorable opinions during this election season, but the comments (such as those listed above) that yearn for a "Bradley Effect" are among the most troubling because they come from people who openly hope to reap the benefits of latent racism in the country. Truly, a ghastly allignment.

Any Republican or Democrat during any election cycle who openly or covertly hopes for a Bradley Effect is a dispicable human being. Abstracts aside (and this site certainly has clinical grounds to hypothesize thereon), there is only so much that a person should feel comfortble arguing (amorally) on direct behalf of their opponent.

As an aside: unlike many of my associates, I do not believe that John McCain has run a racially-tinged race so far. Hillary Clinton, during her Primary efforts, has so far proven to be more divisive along those lines than the Senator from Arizona (Her desperate and deliberate "hard-working white americans" line should remove all doubt). Perhaps McCain's surrogates would be happy to oblige with a "White Hands" ad in October, but I haven't seen it from the man himself.

Politics is all 'bout gittin' money!

2008-09-19T15:45:00.000-04:00

Politics is all 'bout gittin' money!

"One big difference between polling in past electi...

2008-09-19T15:43:00.000-04:00

"One big difference between polling in past elections and this one is that polling is now done almost exclusively with robo-calls. In other words, the voter isn't talking to a human being, but a machine."

Right. The more that the polling process resembles the privacy of the voting booth, the less likelihood of a "Bradley Effect". It can only manifest when the person polled wants to seem more politically correct to that nice college kid on the phone or doing the exit poll. The "Bradley Effect" does not measure racism - it measures LYING for the sake of political correctness. By its very nature, a "Bradley Effect" will be more pronounced in a democratic primary than in a general election, because it would be DEMOCRATS who would be the Bradley Effect voters.

My new favorite sport: stat nerd bodyslams. Great ...

2008-09-19T15:38:00.000-04:00

My new favorite sport: stat nerd bodyslams. Great post Nate.

Please. This just proves that math has a liberal b...

2008-09-19T15:23:00.000-04:00

Please. This just proves that math has a liberal bias. If you seriously needed any more proof that arithmetic is firmly in the tank for Obama, there it is.

One way to look at a the possible "incumbent effec...

2008-09-19T15:17:00.000-04:00

One way to look at a the possible "incumbent effect" is this.

In a normal incumbent election it's up to the challenger to make the case for himself. The incumbent is the default candidate.

Thus, undecideds tend to break predominantly for the incumbent.

However, not only is there NO incumbent in this election, McCain is running as far away from Bush as he can get, even taking the extreme measure of not letting Bush/Cheney speak or even appear at the convention.

Republicans universally deny the obvious truth that "McCain = Bush III."

So, how can there be an incumbent effect?

But, perhaps racism would play a similar role. McCain is the "safe" (read "white" old conservative) choice. Thus, he might be the "default" candidate similar to an incumbent.

The problem here is the miserable unpopularity of the Republican party. McCain is having to run as a "Maverick" just to keep the race close at all.

Thus, any late-breaking undecideds who flip to McCain can virtually ONLY do so because of racism.

But, the majority of voters don't think McCain will be any different from Bush (see this week's NYT poll). Thus, there's probably two competing trends -- undecideds who vote for change because the status-quo is unacceptable, and those who vote for McCain because of racism, even though they don't think he'll change anything and they're unhappy with Bush.

IT's tough to call.

Nate has done an excellent job of showing that the...

2008-09-19T15:16:00.000-04:00

Nate has done an excellent job of showing that the Bradley effect was smaller than sampling uncertainty for Democratic primary voters.

It is fair to ask, however, whether this is true for all other voters. That is, how many GOP and independent voters will say they are voting for Obama when polled but really plan to vote for McCain? This question is harder to test empirically, although given that McCain has about 90% of GOP voters and that the two split independents who are about a third of the electorate, it's hard to see this number rising above uncertainty either. Still, it'd be nice to see Nate take a shot at it.

One big difference between polling in past electio...

2008-09-19T15:12:00.000-04:00

One big difference between polling in past elections and this one is that polling is now done almost exclusively with robo-calls. In other words, the voter isn't talking to a human being, but a machine. Are people really going to be embarrassed to tell a machine that they aren't voting for the black candidate? Perhaps a few, but the vast majority, who might otherwise pretend to be voting for Obama, won't care, because they won’t have to worry about a machine passing judgmenmt on them.

I would guess the Bradley Effect, if it still exists at all, affects polling results at approximately 1/10th of one per cent. In other words, it would have to be a 49.5-49.5 kind of deadlock for it to play a significant role in the outcome.

One way you might see racism play a hidden role in...

2008-09-19T14:56:00.000-04:00

One way you might see racism play a hidden role in this election (rather than the obvious fact that the election should by all rights be a blow-out and would be a blow-out if the last 8 years were any judge).

That is that undecideds are likely to break for McCain at the last minute. They are reluctant to vote for McCain, but they can't bring themselves to vote for the black man.

Thus, they whaffle until time runs out and reluctantly pull the lever for McCain.

If there are a LOT of undecideds (5% or more) on election day that might swing the election to McCain if something like 75% of them broke to McCain at the last minute.

EX: .75 x 5% = 3.75% (McCain's share of late deciders) - Obama's share: .25 x 5% = 1.25%

3.75% - 1.25% = 2.5%

So, in this worst case scenario, Obama would have to be up by 3% on election day in this scenario, or else his GOTV effort would have to net him at least 2.5% for him to win the popular vote.

Of course, the fewer undecideds there are by November, the less of a problem this will be for pollsters to predict.

Large undecideds = trouble. How can anybody predict how they will jump in the end?

And will it be 75%-25% or 60%-40% or some other number?

I hope that the Bradley Effect is no longer releva...

2008-09-19T14:50:00.000-04:00

I hope that the Bradley Effect is no longer relevant, but here in North Carolina, I've seen two senatorial campaigns in the past 12 years where the Effect was clearly documented. Two points suggest that discounting the Effect is dangerous, at least here in the Upper South:

1. I am not persuaded that data from the primaries are relevant to the general campaign. The Bradley Effect is not likely to show up in a primary (and certainly not in a Democratic primary down here), but will be more likely in a general election where people with "closet" racial attitudes are more likely to vote.

2. I don't expect to see the Bradley Effect in the Deep South, and in fact a reverse Effect may be more likely there. But in potential swing states like North Carolina and Virginia, you often see some subtle racial attitudes that may significantly contribute to a Bradley Effect.

I hope I'm wrong, but we'll see in November.

Our right-wing trolls are such idiots!"You'll be s...

2008-09-19T14:46:00.000-04:00

Our right-wing trolls are such idiots!

"You'll be sorry on election day when the Bradley effect costs Obama the election!"

B.S. Pure total B.S.

Don't these idiots know that for there to be a "Bradley effect" voters have to LIE to pollsters? Open racism shows up in polls. Closet racism shows up in polls as long as the voter can find a legitimate reason NOT to vote for the black candidate.

The ONLY time there can be a Bradley effect is if the voter is put into a corner and forced to lie. There's NO plausible/acceptable non-black candidate, he's under serious pressure to vote for the black guy, but he just can't do it.

When alone in the booth. He balks and secretly pulls the lever for the other guy.

Thus, for voters to LIE they have to be EMBARRASSED about voting for McCain.

Well, how many McCain supporters do you see who are embarrassed about voting for him? He's a perfectly credible candidate, and he might even win.

From a liberal Democratic perspective he's an aging fool who will usher in Bush III, but that's hardly a universal opinion! (Alas).

The stronger his candidacy the less possible any "Bradley effect" is. Thus, unless Obama blows the election out as a result of the debates (unlikely) there's little chance of any Bradey effect at all.

White working-class Racists in Michigan keeping the race close? Far from a "Bradley effect" they are openly supporting McCain. Thus, the narrow margin.

Hillary-bots who'd rather see an anti-abortion activist appoint the key Justices who will permanently overturn Roe than allow Obama to get away with beating Hillary? Open PUMAs. Another Hillary delegate just openly endorsed McCain.

Women who just want a woman candidate, even if she's a right-wing nut-job who wants to force victims of rape and incest to carry their rapists' fetus? "Palin's a soccer mom!"

There IS NO BRADLEY effect, because nobody in this country right now has any strong pressure to lie.

Nobody. There are hundreds of excuses to vote for McCain! He's "more experienced," "Maverick"(tm), "reformer", "war hero."

Obama's a "tax & spend liberal," "secret Muslim," "too inexperienced," "too different," "too unknown/too much of a risk in these dangerous times," blah, blah, blah, etc.

No reason to lie when you have a thousand excuses other than that you just won't vote for a black man for President. Period.