As Polls Blow It (Again), O Has Mo

You'd think with 42 races behind them, the pollsters would have the Dem primaries nailed, but again yesterday they proved their utter worthlessness as prognosticators.

Yesterday, as the polls in North Carolina and Indiana opened, the RCP average of the big-time polls showed Obama up by 8 in North Carolina and Clinton up by 5 in Indiana. Oops, again, as Obama doubled that in NC, trouncing Clinton by 16%, and pulled within two points of her in a last-minute surge in Indiana, as the northwestern counties with their big black populations came in.

Obama's performance yesterday is tantalizingly close to a closer -- except that when it comes to calling it "it," the Clintons have a different definition of "it" than most of us. She was looking for Hoosier double digits to balance out what was certain to be a drubbing in NC, but she got just two digits, which isn't the same as double digits by a long shot.

And speaking of long shots, Obama's 200,000+ margin in voters yesterday now puts him back on top of the popular vote, which Clinton previously could lay claim to by counting Michigan and Florida's screwed-up primaries. So the only cloak left for Hillary to wrap her hopes in is that she can win in the big states, especially the ones with a lot of old-line Dems who aren't exactly in the front lines of the tolerance movement. Not much to cling to.

So it looks like the primaries turned out pretty darn well for the GOP. Hillary played her part in bruising the party and playing up Obama's many weaknesses, and we ended up with the one candidate most likely to appeal to Dems who are mature enough to fear an Obama presidency.

But GOP campaign chief Tom Cole says the party is facing a disastrous election in November. Is he just fundraising -- gotta have a big problem to raise big funds! -- or is he right?

Poll Check

In the final minutes before voters in North Carolina go to the polls, the RCP average has Obama up by 8.0, dominating all the polls RCP scans, from a low of +4.0 to a high of +14.0. (The high, BTW, is Zogby, which is continuing this election season in its quest to prove itself inaccurate and inconsequential.)

In good ol' Indiana, home of most of my relatives and my alma mater, the RCP average shows Clinton up by 5.0 after being behind in the third week of April. The swing is almost as dramatic as in NC, from Obama up by 2.0 to Clinton up by 12.0.

Guess who has Obama up by 2? Zogby, natch.

So there's a swing of 18 points in the NC polls and 14 in the Indiana polls, this after pollsters know pretty much exactly what they'll get from each demographic. So it appears obvious to me that the polling operations that aren't playing fair are cherry-picking their demos (blacks for Zogby, white women for SurveyUSA which has high numbers for Clinton in both polls).

We'll check back this evening to see how the pollsters fared.

Classic Polling Mish-Mash

Here's the lead on a story reporting the lastest USA Today poll:

Barack Obama's national standing has been significantly damaged by the controversy over his former pastor, a USA TODAY/Gallup Poll finds, raising questions for some voters about the Illinois senator's values, credibility and electability.

And here's what a story on an NYT poll has to say:

A majority of American voters say the furor over the relationship between Senator Barack Obama and his former pastor has not affected their opinion of Mr. Obama, but a substantial number say it could influence voters this fall should he be the Democratic presidential nominee.

So who are you going to believe? Once again, we see the only poll that truly counts is the one the voters fill out in the election booth.

hat-tip: memeorandum

Polls Off In PA: Biased? Incapable? Inept?

Three point four. It doesn't sound like much, but at this stage of the game, the polls should not have missed the Pennsylvania call by this much -- the high end of the margin of error.

The pollsters had all but eight states behind them going into PA , with a mountain of data on the candidates and the electorate, yet on average (the RCP average) had her at a 6 percent lead while the voters carried her across the line 9.4 points ahead of Barack Obama.

And it's not like there were a lot of surprises in the vote; it went pretty much as expected, as RCP's Horse Race Blog points out:

What we see, then, is what we have seen again and again in this contest. Clinton continues to do well with "downscale" whites. Obama does well with "upscale" whites and African Americans. What is intriguing about this result is not just that it is similar to Ohio - but also that it is similar after seven weeks and millions of dollars in campaign expenditures. Clearly, these voting groups are entrenched.

Check this out and see just how predictable PA was:

Obama carried the black-and-lib urban county and the elite-and-lib university town, and Hillary won everything else.

Granted, the RCP average was skewed by a horrific PPD poll that showed Obama up by 3 (can we say "push poll?"), but Rasmussen had Clinton by only 5 and Survey USA only by 6.

Are the pollsters biased toward Obama? Incapable? Inept?

All of the above?

Pollsters Misunderestimate Hillary

We know how Hillary and Obama did -- but how did the polls do? Looking, as we always do, at the RCP averages here's what we've got:

First, to Texas, where on the eve of the election the RCP average said it would be a squeaker:

Clinton: 46.8%
Obama: 46.5%

It turned out to be close, but not that close, as Clinton pulled away by more than th 3% margin of error:

Clinton: 50.9%
Obama: 47.4%

In Ohio, RCP's summary of the big polls said Clinton would win pretty big:

Clinton: 49.3%
Obama: 42.9%

Actually, the polls called it closer than it actually turned out:

Clinton: 54%
Obama: 45%

Once again, Clinton outpaced the polls by 3%.

What's up? Two things, probably.

First, Independents broke more heavily for Clinton than expected. I've always seen Independents as a negative-thinking bunch, and it's likely that as Hillary started going more negative, i.e., talking frankly about Obama's weaknesses, they listened.

Second, the polls probably misread the number of Hillary-hating Republicans who crossed over to vote for her, in order to keep the campaign running.

So Mr. O. lost the Big Mo, the folks who were being dispatched to tell Hillary to quit are checking to see if their tickets are refundable, and the Donkey's back to biting his tail for a couple more months.

Mission accomplished.

Poll Check: Florida Vote

CNN is beating Fox on reporting returns, having just broken through the 90% barrier, so the results of the Florida GOP primary are pretty much set in stone.

Here are this morning's RCP polling averages followed by CNN's actuals:

McCain: RCP - 30.3%; Actual - 36%
Romney: RCP - 29.8%; Actual - 31%
Giuliani: RCP - 15.0%; Actual - 15%
Huckabee: RCP - 12.7; Actual - 14%
Paul: RCP - 3.8%; Actual - 3%

McCain's final tally was outside the 3% margin of error we expect from polls, but just barely, and no one else was outside the margin. McCain just picked up a bit here and there from each of the other candidates to solidify his narrow win.

This all leads to one (almost) known conclusion: Rudy's out. It's a shame he ran such a pathetic campaign, because he deserved much better. But I don't think he had a choice of any other strategy. If he had had the money, he would have run in more states.

He did tell supporters he's going to California ... but the good money is on the trip being about endorsing McCain, not running himself.

The results also tell us that McCain was able to get away with dirty politics this time around. His contention that Romney waffled on the a timeline for Iraq was a stretch of Reed Richards proportions.

So McCain (and the pollsters) leave Florida with their heads high. But with the warm-up states behind them and Super Tuesday a week away, McCain only holds a 21-delegate edge over Romney (95 to 74). To win, 1,191 delegates are needed ... so you can liken what we've done up to now to a hitter taking a couple practice swings while the pitcher looks in at the catcher for a sign.

UFOs And Global Warming

The League of Conservation Voters is getting hot under the collar due to global warming -- more specifically, due to the media's lack of focus on the Big Hot Button Issue of the Greenie movement.

Would you believe the MSM have asked presidential candidates about as many questions about global warming as they have about UFOs? That might be excusable -- after all, neither phenomenon has been proven to exist.

Salon isn't happy about this state of affairs, quoting a League of Conservation Voters quickie study of transcripts of TV interviews and debates conducted by CNN's Wolf Blitzer, ABC's George Stephanopoulos, MSNBC's Tim Russert, Fox News' Chris Wallace and CBS's Bob Schieffer. Through January 25, 171 interviews with the candidates and 2,975 questions asked, only six mentioned the words "climate change" or "global warming."

The League is all uppity:

"Global warming is unequivocally one of the biggest issues facing the nation and the planet, and one of the issues that the next president will have the greatest impact on. And yet we've gone through the longest presidential primary in our nation's history, and these reporters are ignoring the most pressing issue," says Navin Nayak, director of the global warming program of the League of Conservation Voters.

Nayak (rhymes with "kayak," which we'll all need if his hysterics prove founded) is missing one little point: No one knows more about what is on the mind of the American voter than presidential wannabees, who spend millions of dollars probing what they should talk about and what they should say.

That global warming is not an issue in the campaign should tell any intelligent observer (Nayak apparently fails this test) that all the presidential pollsters have found the same compelling reasons to not talk about global warming: The people of America don't want to hear it.

Maybe after the campaign's wrapped up, the now-hidden details of this story will come out and we'll begin to see what I've expected all along: A strong cynicism of the extremist position staked out by the Gorites.

Poll Check

As Palmetto State voters turn out, possibly in record numbers, to vote in the Dem primary, let's take a look at what the pollsters have to say. Here are the Real Clear Politics polling averages on the eve of voting:

Obama -- 38.4
Clinton -- 26.8
Edwards -- 19.2

We'll check back after the results are in to see if the pollsters were up to snuff. Until then, some thoughts:

If Edwards can't carry the state he was born in, the state next door to the state he represented in the Senate, why is he still running? Even his VP wishes will then be dashed, since one of the roles of the traditional VP (Cheney nothwithstanding) is to deliver some states.

The LAT is trying to generate some hype over the idea that Edwards is gaining traction in the final days (just like Fred Thompson did ... not) and just might edge out Clinton ... that would shake things up a bit, but it's just desperate pundritry, and she's really tossed out the state anyway, focusing on Super Tuesday states instead.

What do the polls tell us about racial voting in the South? Well, nothing from these superfluous numbers, but when we get a chance to look at the details, will we see a big black vote for Obama and a split white vote between Edwards and Clinton? If that's what happens, why do we bother to listen to the Dems about civil rights? Blacks, by the way, make up 50 percent of the SC Dem electorate, by the way.

And finally, if Obama finishes with a lead of these proportions, what does it tell us about his future as a prez wannabe? Not much. He can't win unless he can win a big-delegate state, and we won't know that until the night of Feb. 5.

Poll Check: Update

Now the results are in from South Carolina, so I'm updating my earlier post on the Nevada results, which you'll now see below the discussion of South Carolina.

This morning I posted the latest polls' predictions; using Real Clear Politics' average of several polls; they are repeated below, this time followed by the actuals:

So, South Carolina with 98 percent of precincts reporting:

McCain: Poll - 26.9%; Actual - 33%
Huckabee: Poll - 25.9, Actual - 30
Thompson: Poll - 14.7, Actual - 16
Romney: Poll - 14.6, Actual - 15
Paul: Poll - 4.4, Actual - 4
Giuliani: Poll - 3.4, Actual - 2

In South Carolina, the polls got the race right, although the numbers were off a bit. But all in all, the pollsters earned their keep. Unfortunately the hoped-for Thompson surge some polls sniffed out failed to amount to much.

And here's the earlier materials, starting with the Nevada GOP, with RCP averages, then actuals:

Romney: Poll - 25.7% , Actual - 52%
McCain: Poll - 20.7 , Actual - 13
Huckabee: Poll - 12.3, Actual - 8
Giuliani: Poll - 11.7, Actual - 4
Thompson: Poll - 10.7, Actual- 8
Paul: Poll - 7.3, Actual - 13

Nevada Dems, RCP averages

Clinton: Poll - 37.8%, Actual - 51%
Obama: Poll - 33.8, Actual - 45
Edwards: Poll - 18, Actual - 4

These results are with 90 percent of precincts reporting, so they're pretty fixed.

Once again, the results are far outside the standard five percent margin for error. The misread of the both the GOP and Dem results was equality bad, to the point of showing the polls to be nearly worthless.

Perhaps the worst news is Paul's 13 percent grab, for an apparent tie for second. That means he will be able to say that he's a credible candidate -- in Nevada at least, which is perhaps the most atypical state in America. Loony desert rats for Paul!

Polls in South Carolina may close soon ... may not. McCain's pushing for the polls to be kept open longer due to bad weather conditions. As soon as they do close, I'll update again with those more interesting and meaningful results -- but for now, another bad day for the pollsters.

Poll Check

Here's your handy-dandy polling summary, showing the Real Clear Politics averages as polling wrapped up and voters began going to the polls in the South Carolina primary and caucusing in Nevada.

Be sure to come back tonight as the results come in to see how the pollsters compared to the people.

South Carolina GOP, RCP averages

McCain: 26.9%
Huckabee: 25.9
Romney: 14.7
Thompson: 14.6
Paul: 4.4
Giuliani: 3.4

South Carolina Dems, RCP Average

Obama: 43.2%
Clinton: 33.6
Edwards: 13.2

Nevada GOP, RCP averages

Romney: 25.7%
McCain: 20.7
Huckabee: 12.3
Giuliani: 11.7
Thompson: 10.7
Paul: 7.3

Nevada Dems, RCP averages

Clinton: 37.8%
Obama: 33.8
Edwards: 18

If the polls turn out to be correct, my favorite candidate, Fred Thompson, will certainly have to fold up the ol' campaign, since he has to be strong in the South if he's ever going to get traction. Lesson for the poli-sci majors: Candidates who diddle away valuable campaign months may not be able to overcome the lost time and the image of indecisiveness.

Note: Captain's Quarters says a "Thompson surge" may be growing in South Carolina, quoting an American Research Group poll that shows Thompson running up, hitting 22 percent, compared to the RCP average of 14.6.

And if the polls are correct, my least favorite candidate (well ... there is Ron Paul, too), John Edwards, should hang up the campaign and tend to his dying wife, since if the former one-term, punk senator from North Carolina can't win in South Carolina, it's obvious that the American people are not as dumb as he thinks we are.

Dixville Notch Deuce

Today, the 19 folks who voted in Dixville Notch, NH, are back in hibernation, removed from the national spotlight for four years, when New Hampshire once again greedily grabs the national spotlight by tenaciously clinging to its role of host of the nation's first primary, and they crawl, locust-like, to the polls.

As you may recall, here was their call:

Obama - 7
Edwards - 4
Richardson - 1
Clinton - 0

McCain - 4
Romney - 2
Giuliani - 1

The media dutifully reported this political Puxatawney Pete-watch as if it mattered, just as they reported all the polls, as if they mattered. They were both equally accurate ... or inaccurate, naming McCain, but totally missing Clinton.

ABC's polling whiz, Gary Langer, writes this a.m.:

There will be a serious, critical look at the final pre-election polls in the Democratic presidential primary in New Hampshire; that is essential. It is simply unprecedented for so many polls to have been so wrong. We need to know why.

Sure, like it's never happened before and inaccurate polls are a troubling new phenomenon. The efficacy of Langer's "critical look" is undone a few sentences later when he says:

In the end there may be no smoking gun. Those polls may have been accurate, but done in by a superior get-out-the-vote effort, or by very late deciders whose motivations may or may not ever be known.

If they were "done in," no matter how they were done in, they were not accurate, so defending their accuracy is a refusal to understand that it's impossible to predict something as dynamic as the American political process, so there can be no fix. And therefore, no reason for over-coverage of polls ... or high-paid positions like polling director for ABC News.

Despite the failure of the New Hampshire polling, what will we see next? South Carolina polls!

Imagine an economic model that pegs the economy right half the time -- would we use it?

Or a set of business metrics that predicted a company's performance in half the quarters right and half the quarters wrong -- think its authors would hold onto their jobs long?

Or a global weather prediction model that got half ... oh, forget that one.

hat-tip: memeorandum

Mr. Whipple And Global Warming

Mr. Whipple, the strange man with the fixation to squeeze toilet paper rolls, proved the old advertising adage that you can't just tell somebody to buy something. Instead, you have to bludgeon them into it. Or, as they say in ad school, repetition builds recognition, and recognition builds sales.

So year after year, Mr. Whipple got caught in the aisle white-handed, squeezing his beloved Charmin. And people bought it, and life was good at P&G or whatever corporate demon foisted this heinous campaign on us.

Mr. Whipple, metaphorically speaking, is whispering in my ear, telling me this is quite simply the most amazingly unbelievable chart ever produced:

We have BBC to thank for it, because they stuck it on the bottom of a story about low recognition of global warming in Russia. Scan over to the right, to the green "don't know" numbers and you'll see that absolutely everyone in Brazil, Australia and France has heard of global warming.

Yup, there's not one up-Amazon indigenous tribe member walking about with a bone through his nose who has not heard of global warming. Way out in the Outback where people know a thing or two about heat, there's not even one Fosters-swilling town drunk who couldn't tell you why the appropriate term is "climate change." And all the epicures and snobs in Paris, not to mention every single French sheep herder down in the Pyrenees, can sit down over a croissant and discuss the finer points of urban heat islands and polar ice cycles with you.

Why, even in dumb ol' America, where we're all too fat and sassy in our SUVs to ever actually learn anything, a full-blown 99% of us know about global warming. That 14-year-old gal Warren Jeffs wed up maybe didn't get a chance to find out about it yet, I guess.

To cover up the sheer audacity of making such claims as these, BBC tells us:

The survey was conducted for the BBC World Service by the international polling firm GlobeScan together with the Program on International Policy Attitudes (Pipa) at the University of Maryland. GlobeScan co-ordinated fieldwork between 29 May and 26 July 2007.

The question itself is straightforward enough on the survey instrument:

M1. How much have you heard or read about global warming or climate change?

The methodology section of the survey is suspiciously unrevealing, just giving the number of people surveyed per country, when they were asked and a few sketchy details. There are no clear reasons in the information available that would give us a hint at why such improbable results were posted. I see two possible reasons:

1. The data are true. We may indeed have reached a point where the media have such compelling reach that if they all get singing from the same page, people will hear about it. I find this impossible to believe. There are just too many people how are too remote, too unplugged or too stupid for the data to be true.
2. The countries were selected carefully. I don't see Senegal there, or Peru, or Myanmar, or Yemen. Instead, countries with better than average education and communications systems were selected. Still, the data for the 20 countries selected is simply not believable.
   

Neither idea really pans out. But a second look at the methodology tells the tale: In virtually every country, a phone survey was conducted. Phone ownership biases surveys heavily. Phone owners have dwelling places, money, connections, technology. And (surprise!) when face-to-face interviews were used, they were conducted in urban markets. Guess what? Urban residency biases surveys heavily. Urban dwellers -- especially the ones who will stop and talk to someone on a busy downtown street -- have money, connections and technology.

So what we've learned is that a lot of people know about global warming and there are a lot of ways to bias a survey.

Oh, we also learned that in Russia, they really don't care all that much about the negative effects of global warming. As one meteorologist in the frozen town of Arkhangelsk told the BBC, "I know global warming is a problem, but I would welcome a bit of warmth up here. Then I could grow my own tomatoes."

He was standing in the middle of a frozen river, with the temperature hovering at -25C.

hat-tip: Jim