Objective clarity on what is happening with the polls

[OCinBuffalo]

http://www.nationaljournal.com/columns/on-the-trail/parties-wonder-which-side-s-polls-reflect-reality-20121101

 

My takeaway:

 

1. Somebody is right, somebody is wrong? This is not good. Internal polls for the candidates should not be drastically different. We are supposed to be able to agree on the scoreboard, and then debate the score.

 

2. We have a real problem with polling in general, and no, that is not limited to Nate Silver. Silver is only as good as the polls. The polls have problems, so everybody, including Silver, has problems. (Silver is bad for his own reasons: re-weighting polls after they've been weighted, and using an "ideology score" as though it's = as meaningful and predictive as an On Base %. facepalm )

 

3. We can't keep pretending this is 1965, everybody has a land line, and they all answer their phone immediately. But, we also can't pretend that this is...never, and we can just make whatever electorate demographic models we want.

 

4. Like it or not, polls, and polling have a real effect, and they do shape opinion, if not create it. We are better off not having any polls than having ones that are broken, one way or another.

 

5. There's nothing like being vindicated, albeit by Republicans(but I have been saying this for 4 months now):

Republicans say their party is a victim of media bias—but not in the standard "Lamestream Media" sort of way. Pollsters on both sides try to persuade public surveyors that their voter-turnout models are more accurate reflections of what's going to happen on Election Day. This year, GOP pollsters and strategists believe those nonpartisan pollsters are adopting Democratic turnout models en masse.

Remember, I reviewed the data myself, and said BS. I did not have some guy do the research, read his article, and start passing off both his work and his opinion as mine.

[Duck_dodgers007]

Romney is still going to Virginia and Florida this late in the race. He is playing defense when he desperately needs to be playing offense. He is losing. There is five minutes to play and Obama has a nine point lead. Obama has the ball and is running it between the tackles for first down after first down. tick, tick, tick...

[OCinBuffalo]

Romney is still going to Virginia and Florida this late in the race. He is playing defense when he desperately needs to be playing offense. He is losing. There is five minutes to play and Obama has a nine point lead. Obama has the ball and is running it between the tackles for first down after first down. tick, tick, tick...

This has no bearing on this thread at all....but...Mrs. Obama is going to be in NC on election eve. What does that mean?

 

Romney is going to be in PA, eastern suburbs, big rally, the whole weekend. That might actually mean something...because of the data. You do realize that in PA, rich suburbs of Pittsburgh and Philly are 25% of the vote...and Obama is polling at 34-38% amongst whites, Romney at 62-65%, right? I've spent plenty of time, and lived in both suburbs. I can assure you: lily white.

 

Do that math, if you are capable...and...there's no way Obama is not in deep trouble in these areas, and therefore, these states. If the minority turnout in PA is down from 2008, by just 1-2 pts, then Romney makes it a photo finish. If they are down by 2-3 points, Romney wins easily. You do understand this, right?

 

There's no guarantees...but given the data, that seems like a very fine use of resources.

 

That's the difference: I'm looking for data. Show me some data, stop cheerleading, or STFU.

Edited November 2, 2012 by OCinBuffalo

[Duck_dodgers007]

Obama in North Carolina means he has a chance to take another Southern state away from Romney.

[3rdnlng]

Obama in North Carolina means he has a chance to take another Southern state away from Romney.

 

And Romney in PA means what then?

[dayman]

I have never seen such obsession with polls when the election is less than a week away. Nobody knows the turnout, Obama does have a good "ground game" in swing states, and it's safe to say it won't be 2008 or 2010. But rest assured good men of PPP, there is an actual day where people will vote and it's less than 7 days away.

[OCinBuffalo]

I have never seen such obsession with polls when the election is less than a week away. Nobody knows the turnout, Obama does have a good "ground game" in swing states, and it's safe to say it won't be 2008 or 2010. But rest assured good men of PPP, there is an actual day where people will vote and it's less than 7 days away.

And this response is a long, long way from what I was hearing in August = "you don't have a real point here, you just don't like what the polls/Nate Silver/Intrade are saying".

 

As for the ground game?

http://thehill.com/b...-in-ground-game

 

Obama's "ground game" is not the advantage it has been sold as. This comes right after we find out that Obama's "early vote advantage"...isn't, so far. The R "ground game" was able to deliver OH for Bush in 2004. Thus far, with Obama's lead in early voting from 2008 virtually erased? It appears the R ground game is no joke either. Especially if we are to believe that Ds are cannibalizing their election day voters by making them vote early.

 

There was a laundry list in August of "why Obama can't lose". Over time, we've seen item after item being scratched off that list. The "blue wall" was the first casualty. It seems the "line of death" keeps getting closer and closer to the shore...and the last thing on the list is "structural electoral college advantage".

 

That depends solely on which electorate demographic model is accurate, and we'll see, won't we?

[dayman]

You problem is you are way out of your mind bullish about so called "low propensity voters" just slinking back to silence b/c of reasons they know nothing about (hence why they are low propensity voters in the first place). Additionally, you discount historical growth with naive wishes. Is it possible the Dems have a view of the electorate that is a bit beyond reality? It is possible, of course. Is it likely that the historical developments with the electorate that have consistently built up over years just go away in a presidential election year? Probably not.

 

I saw a post where you advocated taking the last 4 electorates EXCEPT 2008, then averaging them...and proclaiming that is what 2012 looks like. You are out of your mind to say such a thing. There's a reason people who's jobs depend on being as close to accurate as possible don't do stupid things like that...and it isn't b/c they just want to make slow gentle love to Obama it's b/c there is an election next week and everybody will know if they were credible or not.

[/dev/null]

And Romney in PA means what then?

 

Romney is trying to lure the Copperheads that cling to their religion and guns into supporting the Confederacy in Virginia, North Carolina, and Florida, that voted for Obama in 2008 but are magically now Confederate Racists®

[OCinBuffalo]

You problem is you are way out of your mind bullish about so called "low propensity voters" just slinking back to silence b/c of reasons they know nothing about (hence why they are low propensity voters in the first place). Additionally, you discount historical growth with naive wishes. Is it possible the Dems have a view of the electorate that is a bit beyond reality? It is possible, of course. Is it likely that the historical developments with the electorate that have consistently built up over years just go away in a presidential election year? Probably not.

 

I saw a post where you advocated taking the last 4 electorates EXCEPT 2008, then averaging them...and proclaiming that is what 2012 looks like. You are out of your mind to say such a thing. There's a reason people who's jobs depend on being as close to accurate as possible don't do stupid things like that...and it isn't b/c they just want to make slow gentle love to Obama it's b/c there is an election next week and everybody will know if they were credible or not.

White decrease election over election is historically is 1.3%. The model your pals are using for this time? -4%.

 

Who is dreaming again? Who is being "bullish"?

 

2008 is an outlier for 3 reasons:

1. electing a black dude fired up lots of people over things that have nothing to do with a normal election

2. inordinate increase in minority voters, that was not in line with the longitudinal growth rate that had been established. It was a spike. There's 0 reason to believe that it will remain. And, the polls that say Obama is up, also inexplicably show a decrease in minority enthusiasm, while also showing an uptick in white. Explain that.

3. inordinate decrease in white voters. The polls then showed that this was directly due to Bush exhaustion, and the fact that McCain inspired...few. In order for you to be right, and this turnout model to be right, we have to believe that not only ALL of them will stay home again, but that even more white voters will.

 

Good luck with getting all 3 of those things to show up again in 2012.

 

Dude this is common f'ing sense. IF you honestly believe that all 3 of them will show up again, to the same extent?

 

You are delusional. There's no other nice way to say it.

Edited November 2, 2012 by OCinBuffalo

[dayman]

White decrease election over election is historically is 1.3%. The model your pals are using for this time? -4%.

 

Who is dreaming again?

 

2008 is an outlier for 3 reasons:

1. electing a black dude fired up lots of people over things that have nothing to do with a normal election

2. inordinate increase in minority voters, that was not in line with the longitudinal growth rate that had been established. It was a spike. There's 0 reason to believe that it will remain. And, the polls that say Obama is up, also inexplicably show a decrease in minority enthusiasm, while also showing an uptick in white. Explain that.

3. inordinate decrease in white voters. The polls then showed that this was directly due to Bush exhaustion, and the fact that McCain inspired...few. In order for you to be right, and this turnout model to be right, we have to believe that not only ALL of them will stay home again, but that even more white voters will.

 

Good luck with getting all 3 of those things to show up again in 2012.

 

Dude this is common f'ing sense. IF you honestly believe that all 3 of them will show up again, to the same extent?

 

You are delusional. There's no other nice way to say it.

 

Think about what you are saying. Let's take 2010, 2006, 2002, and 2004, and average them out. That's 2012. In other words lets take 3 non-presidential races (where the dynamic is different and turnout is lower obviously), and the presidential race last time the GOP won, but lets leave out 2008 b/c Obama is black and nobody cares anymore....if we average these 4 years where only 1 was even a Presidential race then we magically have the 2012 electorate. Forget any growth in minority participation that's just a Dem pipedream. It's ok we leave out the electorate that happened most recently for a President, b/c Obama was black that time and it was new, and b/c I am personally so disgusted with him that I can't imagine everybody else doesn't feel the same it makes sense that blacks will not turn out as they did nor will other minorities grow as expected and white people are back ready to make 2010 look like the new normal! Red State supreme! Woo!

 

I'm not sitting here vowing for every pollster or even getting into the minutia...but I know you have been deluding yourself and too into Dick Morris and other people who are much too bullish on this GOP surge and fail to recognize the reality of certain struggles the GOP has while emphasizing struggles the Dems have.

 

Romney can win. Obama can win. If Romney wins it will be an upset, not a historical upset but an upset. Why? B/c your nonsense isn't reality. It's the reality distortion field you think you can materialize to boost your enthusiasm. Settle down, stop thinking you know what you are talking about, and accept reality.

Edited November 2, 2012 by TheNewBills

[OCinBuffalo]

Think about what you are saying.

I clearly have. Extensively. The question here is: have you? The other question here is: where the F is your common sense?

Let's take 2010, 2006, 2002, and 2004, and average them out. That's 2012. In other words lets take 3 non-presidential races (where the dynamic is different and turnout is lower obviously), and the presidential race last time the GOP won, but lets leave out 2008 b/c Obama is black and nobody cares anymore....if we average these 4 years where only 1 was even a Presidential race then we magically have the 2012 electorate. Forget any growth in minority participation that's just a Dem pipedream.

The putting words in my mouth/strawman arguments on this board have become so tiresome.

 

Not what I am saying. Not even close. I am saying: let's use the NORM for determining minority growth. NOT THE OUTLIER. Did you take statistics in college? Do you know what these terms mean?

 

There is a historic growth rate for minority voters relative to the electorate as a whole, over the last 40 years, since they've been tracking this. It's ~1.5 %. Your pal's model relies on a 4% increase. Just stop....and think about that for 3 seconds. It also relies on the electorate going from white 76% in 2008(when the white vote was inordinately depressed)...to white 72%

 

What could possibly make you think this is right? Seriously. Even if we include 2008, there's no way in hell.

 

The rest of your post is babble, as is whatever else you have to say, until you explain how the F white voters are going to sit home, especially ones that voted every damn year BUT 2008, in even larger #s this year than in 2008.

 

That's the only way Axelrod's, your buddy's, demographic model makes any sense.

 

If you want to stand by that model, be my guest. Why would I want to stop you from signing up for something I get to mock you for...eternally?

 

EDIT: Liberal Projection Alert!

 

I almost missed this. Hmmm. Why do you think I need to boost enthusiasm? I am perfectly fine with R enthusiasm. In fact, all signs show it's better than expected. The early voting #s in Ohio, and especially Nevada and Iowa, show that R enthusiasm is over performing expectations. Given this, why should I have an enthusiasm problem, or even a question about it?

 

Do you have the enthusiasm problem, and your coping mechanism, as with all projectors, is telling me that I have....your problem?

Edited November 2, 2012 by OCinBuffalo

[fjl2nd]

I clearly have. Extensively. The question here is: have you? The other question here is: where the F is your common sense?

 

The putting words in my mouth/strawman arguments on this board have become so tiresome.

 

Not what I am saying. Not even close. I am saying: let's use the NORM for determining minority growth. NOT THE OUTLIER. Did you take statistics in college? Do you know what these terms mean?

 

There is a historic growth rate for minority voters relative to the electorate as a whole, over the last 40 years, since they've been tracking this. It's ~1.5 %. Your pal's model relies on a 4% increase. Just stop....and think about that for 3 seconds. It also relies on the electorate going from white 76% in 2008(when the white vote was inordinately depressed)...to white 72%

 

What could possibly make you think this is right? Seriously. Even if we include 2008, there's no way in hell.

 

The rest of your post is babble, as is whatever else you have to say, until you explain how the F white voters are going to sit home, especially ones that voted every damn year BUT 2008, in even larger #s this year than in 2008.

 

That's the only way Axelrod's, your buddy's, demographic model makes any sense.

 

If you want to stand by that model, be my guest. Why would I want to stop you from signing up for something I get to mock you for...eternally?

 

EDIT: Liberal Projection Alert!

 

I almost missed this. Hmmm. Why do you think I need to boost enthusiasm? I am perfectly fine with R enthusiasm. In fact, all signs show it's better than expected. The early voting #s in Ohio, and especially Nevada and Iowa, show that R enthusiasm is over performing expectations. Given this, why should I have an enthusiasm problem, or even a question about it?

 

Do you have the enthusiasm problem, and your coping mechanism, as with all projectors, is telling me that I have....your problem?

 

How do you know it was depressed? It seems more like minority voters came out more than usual. That would lower the percentage of whites voting. (BTW, I have 74% for white turnout in 2008)

 

How do you know 2008 is an outlier and not a trend? There hasn't been a Presidential election since then. You are making a lot of assumptions here. Your "model" is nowhere close to perfect. It's more of what you hope happens. We won't know anything until election day.

[OCinBuffalo]

How do you know it was depressed?

I have linked that info from various sources about 50 times in these threads. How about this for fun: find me a link that says white voter turnout in 2008 wasn't depressed, especially in Ohio.

It seems more like minority voters came out more than usual.

Ya think? Now why would that be? What "never in my lifetime" thing was actually happening "in their lifetime"?

That would lower the percentage of whites voting. (BTW, I have 74% for white turnout in 2008)

Yeah...and so would white turnout being depressed...in fact, imagine what would happen if BOTH things occurred?

 

Yawn. This, again: http://www.brookings...-elections-frey

Why minorities mattered in 2008.

Minorities mattered in 2008 for three reasons: first, their relative sizes compared with whites increased in each state; second, their enthusiasm for the Democratic candidate was greater than in 2004; and third, white support for the Republican candidate (John McCain) waned in comparison to the previous election.

 

74% = wrong. See, now they are even lying about the 2008 #s. The white vote was 76.3: http://pewresearch.o...ential-election That is from the US Census Bureau, not OFA HQ in Chicago.

 

It's going to go down to 72% this election, even though 1.7 million white voters who sat last time, are guaranteed to return. The 4 pts necessary to overcome that and still move it down to 72? Where are they going to come from? Apparently: magical fairy land.

How do you know 2008 is an outlier and not a trend? There hasn't been a Presidential election since then. You are making a lot of assumptions here. Your "model" is nowhere close to perfect. It's more of what you hope happens. We won't know anything until election day.

Re-read all the things I have just linked. Stop. Re-read all the things I have just linked. Stop. Keep doing this until that stupid question is permanently erased from your mind.

 

There are 0 assumptions here. Only facts. The only assumption, is that magically, white voters will make up 72% of the electorate this year. It's based on nothing, nothing at all, other than "because David Axelrod said so".

[fjl2nd]

I have linked that info from various sources about 50 times in these threads. How about this for fun: find me a link that says white voter turnout in 2008 wasn't depressed, especially in Ohio.

 

Ya think? Now why would that be? What "never in my lifetime" thing was actually happening "in their lifetime"?

 

Yeah...and so would white turnout being depressed...in fact, imagine what would happen if BOTH things occurred?

 

Yawn. This, again: http://www.brookings...-elections-frey

 

 

74% = wrong. See, now they are even lying about the 2008 #s. The white vote was 76.3: http://pewresearch.o...ential-election That is from the US Census Bureau, not OFA HQ in Chicago.

 

It's going to go down to 72% this election, even though 1.7 million white voters who sat last time, are guaranteed to return. The 4 pts necessary to overcome that and still move it down to 72? Where are they going to come from? Apparently: magical fairy land.

 

Re-read all the things I have just linked. Stop. Re-read all the things I have just linked. Stop. Keep doing this until that stupid question is permanently erased from your mind.

 

There are 0 assumptions here. Only facts. The only assumption, is that magically, white voters will make up 72% of the electorate this year. It's based on nothing, nothing at all, other than "because David Axelrod said so".

 

Even with your increased voter turnout for whites, Romney isn't getting all of those votes. Obama will get some. Romney will need a big increase to really see an effect. And I don't see minority voters not turning out for the President (unless Republicans have their way to suppress them. )

 

Whites will probably make up 74-75% of the electorate. As you said, it decreases just a tad each election. Latinos were the ones who really made the jump in 2008. I don't see that changing this cycle. Blacks actually made up the same % of votes as they did in 2000. Romney doesn't win with either group. I think you are really underestimating the Latino vote.

[Cugalabanza]

Objective clarity

 

[John Adams]

I've been saying Obama was in a runaway for a long time but if you Dems don't see that the landscape changed dramatically over the last 8 weeks, you're nuts. Obama still has an advantage in the electoral mapping, but he's gone from a wide lead in a lot of states to a close lead in many. Romney's return to PA is not good for team Obama. Romney probably doesn't realistically think he can win PA, but it's close enough for him to give it a try, and it's not a Hail Mary anymore.

 

Romney needs to pull off some surprises on Tuesday and win in a state that it seemed improbable he would. I still think it's Obama's win but at least Romney has a shot. Given my dire predictions earlier in the campaign, I am really excited that Romney actually could win. (This is the worst news ever for the Romney supporters, since I've never backed the winner.)

Edited November 2, 2012 by John Adams

[Nanker]

Even with your increased voter turnout for whites, Romney isn't getting all of those votes. Obama will get some. Romney will need a big increase to really see an effect. And I don't see minority voters not turning out for the President (unless Republicans have their way to suppress them. )

 

Whites will probably make up 74-75% of the electorate. As you said, it decreases just a tad each election. Latinos were the ones who really made the jump in 2008. I don't see that changing this cycle. Blacks actually made up the same % of votes as they did in 2000. Romney doesn't win with either group. I think you are really underestimating the Latino vote.

 

And you think BO will get anywhere near the support in the white community that he enjoyed four years ago? Put down the crack pipe. Tens of millions of adults have been unemployed for nearly his entire administration. It doesn't matter who's fault it is - he's been in charge and has been largely ineffective in turning around the economy. As others have said before, he drove an ideological dream and spent all his capital on it in his first two years instead of fixing the problem plagued economy. People see an alternative and are backing it - celebrate success and build business instead of demonizing the wealth and jobs creators.

 

BO has no business sense. None at all.

 

I've been saying Obama was in a runaway for a long time but if you Dems don't see that the landscape changed dramatically over the last 8 weeks, you're nuts. Obama still has an advantage in the electoral mapping, but he's gone from a wide lead in a lot of states to a close lead in many. Romney's return to PA is not good for team Obama. Romney probably doesn't realistically think he can win PA, but it's close enough for him to give it a try, and it's not a Hail Mary anymore.

 

Romney needs to pull off some surprises on Tuesday and win in a state that it seemed improbable he would. I still think it's Obama's win but at least Romney has a shot. Given my dire predictions earlier in the campaign, I am really excited that Romney actually could win. (This is the worst news ever for the Romney supporters, since I've never backed the winner.)

 

I saw the first Romney ad in weeks today. He's definitely making a play for the state. He'll make a stop there this weekend. He's got a lot more cash than BO does, so even if he doesn't win PA - although I agree there does seem to be a possibility, however slight that he could - he's drawing attention and resources away from BO's efforts in other battleground states.

 

I do share your skepticism though.

[OCinBuffalo]

Even with your increased voter turnout for whites, Romney isn't getting all of those votes. Obama will get some.

How profound. Yes. That's true. Romney is only getting 59-60% of them. Hmmm. What happened to the last guy who only got 60% of the white vote? Oh that's right, Ronald Reagan "only" beat Walter Mondale 49 states to 1 in 1984. He didn't get all of the white people, either. He should have gotten more, and gone for the shut out.

 

The only chance Obama has, if Romney's white performance holds? That Romney does worse amongst minorities than McCain, AND, that Axelrod's model is right. Both must be true, or this is a close Romney win. If neither is true: this is a Romney blowout.

 

EDIT: Now that I see John Adams and Nanker are here...consider this: These two guys may end up being the reason Romney wins PA. If enough of them switch their vote from Obama to Romney, or, vote this time, or, don't vote for Gary Johnson(John Adams historically votes for silly), Obama can lose PA.

 

Think of it this way: What is 60% of 75% of the total electorate vs. what is 90% of 11%, in terms of # of votes? The first is the Romney support among whites, the second is Obama's support among blacks...if the polls are right. See? They have to tinker with the white vote...because a single point in that wipes out a 5 point swing in the black vote. IF Romney's support goes up to 61, that wipes out Obama's support with blacks going from 90 to 95.

Romney will need a big increase to really see an effect. And I don't see minority voters not turning out for the President (unless Republicans have their way to suppress them. )

 

Whites will probably make up 74-75% of the electorate. As you said, it decreases just a tad each election. Latinos were the ones who really made the jump in 2008. I don't see that changing this cycle. Blacks actually made up the same % of votes as they did in 2000. Romney doesn't win with either group. I think you are really underestimating the Latino vote.

In effect Romney being at 59-60 with whites has ALREADY wiped out the historical minority historical increase, it's even wiped out Obama's 2008 minority support. So, the only thing left for Axelrod to do? Say that whites aren't going be 76% of the electorate, or even 75. He says they are 72.

 

No. Now you are being Axelrod. Where the hell is +4 latino going to come from, that pushes the white vote down to 72%...when latinos supported Obama ~65%, not 80%, in 2008, and every poll all has them down in both support for Obama and enthusiasm this year? Forget the individual poll results, that's what the trend says.

 

This latino argument = emperor's new clothes. But, it's the only one Axelrod, and apparently you have.

 

On election night, you better have your sombrero on, make some burritos, and get yourself some Corona...perhaps do some human sacrifice to an Aztec god? With the early voting in Ohio, and everywhere else? Here: http://www.washingto...38d93_blog.html If we throw out half of these results, and only use the other half? The Aztec thing is looking to be your only chance at this point.

Edited November 2, 2012 by OCinBuffalo