I’ve owed PJM another polling piece for a couple of days, but frankly there’s not much left to say that’s particularly deep. Look, it’s like this: for all the reasons I’ve mentioned before, the polls this year are more unreliable than ever. When the pollsters have to call eleven people to get one person to cooperate, then nobody has any idea what the real population is thinking.
That said, there are some things to look at that can perhaps help you sleep tonight.
I. Michael Barone
There’s not a lot of people smarter about American politics than Michael Barone; he’s worked for both parties over the years, he’s made a life study out of American politics. And he’s been saying for weeks that he thinks Romney will win. He said it today, again.
Bottom line: Romney 315, Obama 223. That sounds high for Romney. But he could drop Pennsylvania and Wisconsin and still win the election.
II. Today’s Washington Post
In a story today titled “Obama’s Defectors,” the WaPo looks at a long series of their polls. Rather than averaging them — I just drove a bunch of people screaming from the room by showing a proof of the problems of averaging, so I promise I won’t, but believe me, it sucks — they looked at a long series of polls of various different subpopulations. In each case they asked people who had voted for Obama if they were going to again. In every case, there was a fair proportion of “defectors.”
Overall, “[t]wo weeks of Washington Post-ABC News tracking poll interviews find 84 percent of likely voters who supported Obama in 2008 support him this year, while 13 percent say they are switching to Romney and 3 percent are backing others or haven’t made up their mind yet.” Now, from the standpoint of the Republican Party, whether they vote for Romney or for Mickey Mouse, they reduce Obama’s vote. Now, here’s a little bit of math for you: Obama got about 53 percent of the vote in 2008. Obama is only getting 84 percent of those votes this time.
84 percent of 53 percent is 45 percent.
III. Enthusiasm
One of the things that turned the election for Obama in 2008 is that while a lot of people were excited about Obama — from the people who thought he really was the Messiah to middle of the roaders who just thought it’d be cool if we elected a black man — a lot fewer were excited about McCain. Lots of hard-core conservatives thought he was a RINO; a fair number of people believed the hatchet job on Sarah Palin; a bunch of otherwise sensible people got gamed on the financial crisis. The result was that from about the first of October on, the crowds were surging for Obama; early voting was way in Democrats’ favor.
This year, the actual data is much the other way. (There were some polls, of course, that early on claimed Obama was getting amazing proportions of the early vote in, say, Ohio, but they turned out to be what mathematicians call “nonsensical,” with sample sizes of, like, 50.)
I’m going to propose an idea: in the absence of other evidence, I think we should assume that people who vote on Election Day will be much like the people who vote before Election Day. This election is different from previous ones in one significant way — we’ve made it ridiculously easy to vote early or absentee (unless, of course, you’re in the military overseas, but that actually is a fairly small population). You don’t need to be extra enthusiastic to vote early.
If I’m right, then Republican turnout will turn out (heh) to be quite a bit more enthusiastic, and therefore more likely to vote, than Democratic.
If the population is exactly evenly divided, and 53 Republicans show up to vote for every 45 Democrats, what is the actual result?
53 percent Republican/47 percent Democrats
Now, does all this mean Romney is certain to win? Of course not. Hell, all those D+8 polls could come true. Someone could come up with pictures of Romney in bed with a dead boy and a live girl. A New Madrid earthquake could disrupt the whole country. The Mayans could turn out to have been optimistic.
I’m just saying that the way to bet is that Romney has the odds on his side; instead of letting uncertainty discourage you, there are three days left. Stay calm, do what you’ve been doing, and remember the goal.
Viva Christo Rey!
Estimada NC Chica de la Montana,
Yo te acuerdo! Gloria a Dios!
Mr. Martin you are a pleasure to read. I would very much enjoy sharing articles I read on PJ media with my friends on social media sites. As far as I can tell my iPhone application does not allow me to do that. As the science and technology guy I’m wondering if you can help me decide if that’s a bug or feature?
Thank you very much, Joe. I don’t really know much about the iPad app except that it looks like the original wireframes I roughed out before being lured back to a real job; I think I can say, though, that not being able to share to social media sites is more of a bug than a feature. Page views are our friend.
Not to get way off topic, but since this issue was raised, it seems like PJM in general has technical issues (to put it politely). I can’t leave a tab open from more than maybe an hour, because it causes Firefox, Chrome, and just about every other brower I’ve tried tend to leak RAM something horrendous. Are they aware that the site has these problems? Other sites don’t seem to do this, or not anywhere near as much. I’ve been able to slow the process down significantly by installing a flash blocker plugin. I think a lot of the problems have to do with promiscuous use of flash.
“I just drove a bunch of people screaming from the room by showing a proof of the problems of averaging”
Well, I’d like to see it. But then again, I’m the kinda weirdo who gets ticked off when people form the Lagrangian the wrong way. It’s f=λg, NOT L=f-λg dammit!
…I think I need to start drinking again.
Okay, it’s fairly straight forward actually. Each of the poll results is really an approximately normally distributed random deviate with mean at the percentage of the vote given. Assume wlg that the margins of error are all the same; it’ll turn out that in terms of order statistics it’ll all end up in the constant anyway. The mean of the sum of normal random deviates is the sum of the means, the variance is sum of the variances. By the assumption, that means the variance for the sum is n*σ^2 (stupid WordPress won’t let me have a superscript.) Thus the standard deviation is √n * σ and thus we see that the standard deviation of the sum of these independent normal random deviates varies with the square root of the number of random deviates summed. Margin of error is basically just the 2σ band above and below the mean. In other words, the margin of error grows approximately as O(√n).
So, when RCP averages their ten polls with a margin of error of roughly 3 percent, the margin of error of the average is more like 9.4 percent.
I’ve been explaining it intuitively by asking a reader to visualize each poll result not as a crisp number, but as a fuzzy thing, dark in the middle and lighter at the ends. If you overlay a bunch of them with slightly different centers, then you’ll see something with a much wider “blurry part”, from the left end of the lowest one to the right end of the highest.
Um, OK. I’m not sure I followed all that, but I agree RCP averages are worthless.
Different samples, different weights, different sample sizes, and different margins of error all averaged together is the essense of “garbage in, garbage out.”
Not to mix metaphors or anything, but trying to figure out an election by looking at an RCP average is like trying to learn what an orange tastes like by drinking fruit punch.
Charlie’s no doubt a genius, but I like RayRaven’s ‘fruit punch’ analogy better.
Hey, I gave you guys a chance to avoid it.
Well *I* understood it, but I’m a hopeless geek. I actually get excited when they let me do predictive modeling at work. I’ve been trying to think of ways to explain these things to my more, uh…liberal arts oriented friends….and I like your “fuzzy bar” image very much. I might try to use than in an animation.
I’d love to see it!
That first paragraph threw me in the ocean without a lifejacket, the third one brought me back to the ship successfully
. Thanks for the excellent visualization, that actually made it clear for me.
D
You know what’s worse? I really do think like that.
I think I’ll ask my mathematical economics professor to explain it to me after class next Thursday.
…did I just type that out loud? Rats.
This should be patented
“In other words, the margin of error grows approximately as O(√n).
So, when RCP averages their ten polls with a margin of error of roughly 3 percent, the margin of error of the average is more like 9.4 percent.”
Uhm, no: the margin of error of the SUM grows approximately as O(√n), but the margin of error of the AVERAGE (sum divided by n) grows approximately as O(1/√n).
That does not mean that I give more credence to the polls than you do, but for different reasons… which I actually explained in a comment that seems to have disappeared. If it does not pop up, I’ll rewrite it.
I think you forgot to divide by n. The error of the sum is of the order O(√n), but of the average it is O(1/√n). Assuming, of course, that the samples sizes are comparable, which is indicated by the assumption that the fractional error is roughly the same in all n polls. I expect in this sort of thing systematic errors are hard to account for and could be the dominant feature, which has the practical implication that the carefully done good polls are probably as good or better than the averages. If we only knew which polls were the good ones we’d be home free
Charlie,
It’s not the sum of the means that matters, it’s the mean of the means. The dispersion of the sum of the means does indeed grow by root-N, as you point out, but then you have to divide by N to go from the sum of the means to the mean of the means. When you do that, your root-N growth becomes root-N contraction.
I’ve got to think about this a bit more, and today isn’t going to be the day.
How are the results from early voting and turnout models taking into account party defectors? For example if you say 60 registered Democrats voted and 40 registered Republicans, how do you account for who those people actually voted for? I know in a lot of states the primaries are open only to people with that party affiliation, thus forcing many moderate voters to pick a party they may not have a strong commitment to if they want to vote in the primaries. I know a lot of people in my home state of NM who are registered as Democrats for that reason, but are planning to vote for Romney this year. At a guess you statistically won’t see any registered Repblicans voting for Obama, but probably at least 5-10% of registered Dems voting for Romney, probably more in coal states. So if a model predicts a +8 turnout among registered Democrats, and 5% of registered Democrats defect to Romney, would the resulting 10 point swing give a R+2 outcome? I’m not sure how they’re accounting for all this.
Badly.
And on purpose.
Obama will win, California, New York, Illinois, Vermont and the District of Columbia.
Obama May win Washington, Oregon, Maine, and Hawaii.
All other states will vote Romney.
The republicans will gain 12 in the Senate and hold the House with 7 to 11 seats gained.
I can now state what I think the Popular Vote outcome will be;
The Popular Vote Split will be – Romney 54 Obama 45 (+/- 2 emphasis on Minus)
You are going to be very disappointed soon.
All evidence points to the opposite.
I think maybe that word doesn’t mean what you think it does;
http://en.wikipedia.org/wiki/Evidence_%28law%29
IMHO, if you can’t prove it in a court of law, it is opinion.
Meanwhile, here is the post election, Democratic party theme song;
http://www.youtube.com/watch?v=1HsgfQux-ho
“Oh, people can come up with statistics to prove anything, Kent. Forty percent of all people know that.” — Homer Simpson.
I think I can prove this in a court of law…
http://abcnews.go.com/Politics/OTUS/chick-fil-supporters-gather-appreciation-day/story?id=16904664
Tell me it is only my opinion that people (on the RIGHT) flocked to support a Chicken Sandwitch Shop and that those same people won’t support a Candidate.
Evidence: The available body of facts or information indicating whether a belief or proposition is true or valid
Note the word OR between body of facts and information.
That word means EXACLTY what I think it does.
You, on the other hand, taking it to be defined as “A body of facts or information admissable in a court of law” makes you the one that hasn’t a clue as to what it means.
A pleasure to read. But the expression is “found in bed with a dead girl or a live boy.” You can’t mix and match or add apples and oranges; rhetoric ain’t polling.
Who says I can’t?
Now 84% of 53% is more my kind of math Charlie but I never left the room.
Two more points, Rove is not only predicting a Romney win, he is relaxed and confident about it. That suggests to me the proprietary polling really looks good.
The crowds. Those people in WI and Ohio yesterday think they have come to see the next President and that becomes contagious. This may be the weekend of cascading preferences as people go to parties and discover people they respect are defecting. Making it Ok for them to as well. Or they hear the troubling facts about Benghazi despite the media blockade as they are out and about over the weekend.
Nothing is there this weekend to make the break to Obama.
I saw Karl on the tube last night and he said yeah, he thought Romney would win but that it would be a squeaker. Also that we wouldn’t have a result until the wee hours of Wednesday morning. Not exactly sounding like a shoo-in.
Who cares what that RINO/Establishment hack Rove says? There is a movement afoot in this country that started when Obamacare was passed.
Obama WILL BE defeated and it WILL NOT be a “squeaker”.
The FoxNews poll just released is also garbage. They cannot just come out and say that Romney is way ahead, because that might hurt their numbers of eyes watching them. They need it to be a horse race, not a blowout, so folks will still tune in. It’s gonna be an early night, when PA goes for Romney, and the Libs will shut off the TV and go cry.
The libs are already crying. There is a stench in the air – they are stupid but they can smell.
I’m just wondering if this narcissistic fool has bothered to start on his concession speech. Or if he’ll even give one. Not that it matters – listening to him is akin to hearing two cats mating. Either drive me nuts – and it’s a short drive in my elder years.
I may just take a listen to his concession speech though. It just might be the prettiest sound on earth to hear this fool announce for the first time in his life that he’s a loser. The only downer to that is we have 2 1/2 months of this loser before he takes up his new role – the mantle of Jimmy Carter and (arguably) title of the worst x-president alive. The only bright spot here is Carter’s days are numbered. We’ll only have #1 and #2 around for a short time – best enjoyed with the mute button!
Thank God for the mute button! Its gotten me through these past 4+ years.
Loser? LOSER? Obama will claim that his friends stabbed him in the back,
that he was framed to take the blame for Benghazi – It may even be true.
Janet Reno once said that she preferred to predict the jury AFTER they returned the verdict.
Mutatis mutandam, I prefer to wait until Tuesday night to make my predictions. but I sure do hope that you, Charlie, are correct, and Mr. Barone (the real Brainiac).
But forget about that Lagrange multiplier business and, like Barone counsels, just stick to the fundamentals: the path integral and good old F – ma.
Janet Reno once said that she preferred to predict the jury AFTER they returned the verdict.
Mutatis mutandam, I prefer to wait until Tuesday night to make my predictions. but I sure do hope that you, Charlie, are correct, and Mr. Barone (the real Brainiac).
But forget about that Lagrange multiplier business and, like Barone counsels, just stick to the fundamentals: the path integral and good old F = ma.
I’ve been tryingall morning to think of a comeback, but I’ve given up; I don’t want to force it.
Who would have thunk that the tables would turn so dramatically? It was October 3, a month ago, when Mitt Romney was able, in debate 1, to address the American people without the media bias and the Obama lies!
Romney’s obvious intelligence, knowledge, and decency came to the fore and he has been on a roll since. Just four more days to go in this vile campaign.
This all presumes there won’t be Obama win with Romney at 52 %and Obama at 52.1%
Well, that’s mathematically impossible unless the dead really are voting
David
Ever been to Cook County?
What is impossible for me to understand and accept is how this could even be a horse race!
I wholeheartedly agree with you, but I’ve come to realize that I”m a stranger in today’s America, where freedom is sold for free lunches, guilt free abortion, sloth, lust, hedonism, and self-centered privelige at the expense of moral imperatives. The rotten economy is but a byproduct of our cultural demise.
Booing God, and cheering for abortion, deviancy, and “victimization be us” at a convention that could better be named “Condom’s “R” Us” This is a far better window as to why we are where we are.
The left planned this well–taking decades to acheive their take-over our our once proud and responsible God-fearing citizenry, and now we argue about where the deck chairs ought to be placed, or how much they cost, on this Titanic called America.
Four years of economic responsibility won’t fix the real problem–our misguided secularist culture.
“
Excellent comment, Don. Fixing the economy is the top priority as a place to start, however. Even the most uninformed can relate to that and the positive effect that flows from it. The next priorities must include turning the economy into a double-edged sword as we choke off the economies of the mainstream media, the educational establishment, Hollywood, the political locus of most of the rot that infects our society……the modern Democrat party. An improved economy will only drive the above mentioned cockroaches back into the crevices where we must continue to pursue them. My hope is that Obama is removed from office and the great cockroach hunt has just begun. That is an imperative and is the only lasting chance for our survival as a free people.
Just think how wonderful it will be to have a man in the White House who sets a higher tone!
BINGO.
We need more than Romney and the Senate. We need a return to God, otherwise we will have just slowed the descent.
When they booed the Creator, I knew they were done.
Don’t over think it Soft. The MEDIA, period. They control it, we don’t. The Left got their pseudo Kennedy-esque television performer elected in the midst of an economic crisis and proceeded to prop him up at every single turn on every television network but one.
The media spent 4 years lying about the true state of the economy (we should call it “the unexpectedly economy” as every piece of bad economic news was inconceivable to them.) They lied about who he was from day one, they smeared anyone who opposed him, and they are covering up for him now on the attack in Libya, as well as, many other scandals.
It really is that simple. When Romney assumes the office in January the economic reporting will change overnight, literally. Romney and Ryan will be called every name in the book as they attempt to deal with the structural problems that the Democrats have deliberately ignored for the last four years.
The battle to save our country will only be started when the conservatives come up with a long term, aggressive, systematic and detailed strategy to combat and then control at least half the media. Without an ability to provide a day to day narrative of what is really happening within our country, we will be left watching, in disbelief, as the left lies our miraculous nation into the dustbin of history; which seems to be their underlying intent.
One example? DEFUND PBS ENTIRELY if they refuse to reform their leftist propaganda pseudo news division. Why the public is forced to subsidize a 6th leftist “news” organization is as incomprehensible as it is indefensible. Attack now. Any new hires at the public network must be conservative, right wing, libertarian, you know alternative, to the dominant liberal media class. Got to fight harder people…
god bless
Thank you, Mr. Martin. At this point this is the only sort of essay I want to read. After so much fawning adoration from every traditional source of information till this point I don’t want to see any more pessimism on our side. It certainly isn’t going to help. Also, I love the sign!
Charlie,
Your response to #3, esp. the last graph, was damn near peotic. Lagrangian math “transformed” into visual/sensual gelatinous masses-let’s hear it for fuzzy math, ironically the only kind that Democrats seem to comprehend.
Heh. I actually did a couple papers in graduate school on fuzzy set theory and its’ applications — like fuzzy arithmetic. Anyway, thanks!
The Japanese are way ahead of the rest of the world in using fuzzy logic for automation. Or as they say, Fuzzy Rogic. It really puzzles me why this hasn’t caught on in the ROW. We round eyes’ keep trying to futz with neural nets, and keep failing (for the most part).
I don’t understand either — fuzzy systems theory, fuzzy logic, etc., is not well thought of here. When I did my first school paper on it, I got an A but the professor I did it for actually called down the hallway to his colleague “Hey, we’ve got a student here doing a fuzzy logic paper!”
It got more respect in Electrical Engineering (I’m Computer Science) but even that, at least at Duke, the professors who were interested were all Chinese of Japanese.
That’s a topic for a whole article – how culture impacts technological choice and preference. It’s really interesting how China has made strides in manufacturing tangible goods, while India has tended to produce much more idea-type products, such as software.
As these countries develop in this century, this is going to emerge as a huge deal. I’m running into a lot more Brazilians than a few years ago on electronics/software boards, to the point where Portuguese is emerging as a discussion language. I’m even seeing a smattering of Africans doing various interesting things. It’s going to be an interesting century, if we don’t completely hose it up.
You can prove anything with statistics. The average person has one t**t and one t’******l.
A 315 electoral win won’t just be devastating for Obama in that it would be too large for a dead voters and court challenge overturn, but it would also kill a lot of the credibility of the MSM pollsters.
Kiss even more ad revenue goodbye.
This nugget is hidden in Gallup’s latest findings: “The largest changes in the composition of the electorate compared with the last presidential election concern the partisan affiliation of voters. Currently, 46% of likely voters identify as Democrats or lean Democratic, compared with 54% in 2008. But in 2008, Democrats enjoyed a wide 12-point advantage in party affiliation among national adults, the largest Gallup had seen in at least two decades. More recently, Americans have been about as likely to identify as or lean Republican as to identify as or lean Democratic. Consequently, the electorate has also become less Democratic and more Republican in its political orientation than in 2008. In fact, the party composition of the electorate this year looks more similar to the electorate in 2004 than 2008.”
H/T National Review
Charlie, If I make take the liberty.
Excellent article. Excellent response. Now I know why I struggled so in graduate school statistics. Though, being 77 and trying to cope with the realization that I am finally missing a step or two in a failing attempt to keep up, I still recognize a keen intellect’s illustration/explanation of a tough topic (for most of us) when I read one.
Good show!
As mentioned above, it is nearly impossible to believe that this nation has sunk so low it is even a horse race at all.
Keep calm…and remember.
Much of “the world” has a vested interest in America not righting its ship of state, which translates to re-electing Obama.
In addition to the Chinese & Russians who have endorsed Obama, consider Islamists, welcomed at the White House and, reportedly, having done fundraisers with Pelosi.
Cretin Oogoe Chavez has endorsed Obama.
Certainly all those creeps we support with American tax dollars at “The UN” (about 25% of total UN budget is paid by us) who are hellbent on using the organization to destroy Israel and, more recently, have laid claim to a right to send monitors to US voting venues….are pumping for another 4 years of an abject and pathetic United States, led by a guy who disses Israel and its president at every turn.
Great article!
You’ve probably answered these questions before, or else your readers are sufficiently learned in statistics not to need an answer, but I want to know how the Margin of Error should be interpreted.
Does the margin of error simply mean that the two numbers are accurate with a certain percentage of probability (e.g. MOE +/- 3% means that the results are considered accurate with 97% confidence)? Or, do you add/subtract that 3% to the results (e.g. MOE +/- 3% for R: 48% O: 48% could be R: 51% O: 43% or other combination of add/subtract)?
At any rate, when media reporters say a poll result is withing the MOE, are they saying something meaningful or not? I suspect they need a course in statistics just as I do.
They probably need that course more than you do: you’re asking the right questions. The newsers I know (I’ve met a fair number of them in the six years I’ve been doing this) are mostly in that group that gets queasy when asked to do long division.
Here’s the way to interpret MoE. If you have a poll value (which is a mean or average) m with an MoE of k, then 19 times out of 20, another poll taken at the same time over the same overall population will have its mean in the range m-k ≤ m ≤ m+k.
One possible poll is called “an election” and counts a whole population. That gives a more or less exact figure (although because of spoiled ballots etc there’s even some randomness there.)
So if you have a poll like Rasmussen with Romney at 48 percent MoE 3.5 percent, then you’d expect the actual election to have Romney between 51.5 percent and 44.5 percent. Now, notice that both Romney AND Obama are at 48 percent in today’s Rasmussen: that means the results could be anywhere from 52 R/48 O to 52 O/48 R and agree with the poll results.
In other words, today’s poll numbers tell you exactly nothing about who’s winning: the poll can’t tell. What’s more, if it were 49/47, the poll still tells you exactly nothing.
When you have a poll that changes from day to day, and the difference between two days is within the margin of error, then you can’t say the poll is doing anything. So 49R/47O changing to 48R/48O gives you no new information either — it could easily be just randomness.
When you have two polls that differ within the margins of error, that means having two polls tells you nothing more than one poll did — there’s no new information there.
Thanks for the info.
I’m 70, a ME, excelled at math (back then)…BUT you’ve got my head spinning.
November 6, 2012 will be the only poll I’ll be able to believe as a (maybe) true reflection of what the majority of Americans want: a (sort of kind of) REPUBLIC or SOCIALISM
I think it all comes down to motivation. The people who are most motivated to make a change will win. And as far as I can see, the Republicans and the conservatives that I know are really, really, motivated to vote. Wild animals (or ACORN members) couldn’t stop them from showing up and throwing Obama out of office. So let’s just say the people who hate Obama have been motivated from election day 2010 to get their chance to send Obama home to Chicago next week. And I think that is going to make all the difference. And don’t be surprised if Obama doesn’t get as many supporters from young people or African-Americans. There is a lot of anger, bitterness, and disenchantment out there among former Obama supporters, especially among African-Americans (14%+ unemployment, higher amonger younger blacks) and college graduates (roughly 50% can’t find a job that has anything to do with their major, if they can find a job at all). So it will be interesting to see what happens on Tuesday. I think Romney will win and by a decent margin.
Exactly. That’s what that WaPo data tells us too: in every group polled Obama has at least some defectors; over all Obama voters, 16 percent of them say they’re voting for someone else this time.
Here’s my high school dropout math: If 16% defect from Obama, and he only won by 7% last time, and virtually no McCain voters defect from Romney….. ROMNEY WINS BY NINE PERCENT! ..(I think, if my maths right?)
Well, no, because that’s 16 percent of *his voters*. So you want to take off about 16 percent of his 53 percent, which is the same as taking 84 percent of his 53 percent, which is 44.5 percent. Except all these numbers are rough, so call it 45 percent. So he loses 8 percent, not 9 percent. not that big a difference admittedly.
Not correct either. The math is correct, but the problem is wrong. This all presupposes the vote totals remain the same. They won’t. Further, even if 84% do not defect, they may not vote at all. This is the enthusiasm gap. Less people on the Left will vote. More on the Right will. Does anyone think a single Republican voter from last time will not vote this time? Ha! I expect Obama to lose a fourth of his 69mil votes from last time, so he may come in at 52mil-ish. Yes, that includes fraud, because there was as much fraud last time. Romney is going to crush him, just crush him! It’s not going to even be close. The polls are way, way off.
Thanks for the answer. I particularly like your explanation of how much information we really have. If I had understood this weeks ago, life would have been less tense.
A further clarification, please.
When I apply the MOE to create a range of values for both R and O, those values would obviously overlap (unless the poll results imply a blow out which, of course, is not the case in the battleground states). So, it appears from your example that I should calculate the maximum value in the range for either candidate and then assign the remainder (100% – that maximum) to the other candidate. In that way, I can come up with the best and worse case for each candidate. Have I got that right?
Thanks again.
Yup, exactly. We know the results have to total to 100 percent, so for example, 52 percent Romney and 53 percent Obama isn’t possible.
Except, of course, in Cook County.
OK, love the math (and I do get it) but the title says it all. I can’t put all those numbers on a bumper sticker, but I just zeroxed the “Keep Calm and Finish Him” sign; it is now taped to the side and rear windows of my car.
And while I am a Vermonter (but not Blue) I will be in NH today; and will be making a lot of people smile, If I don’t get arrested by the thought police on the way first
And the thrill of pissing off my progressive in-state neighbors….priceless
It’s always a sort of hard decision. They say in popular science publishing that every equation costs 10 percent of sales.
Anyway, I’m glad you like it; I should mention that I stole the sign from Facebook, Gods alone know who started it.
McCain was a loser. The only reason he got as many votes as he did was due to party loyalty and enough people able to see through the Obama BS. This time we have a decent candidate and a knowledge of what Obama really is. You could throw a live goat in with the dead girl and live boy and I still don’t see how Romney could lose.
Just another Bob ‘Viagra’ Dole
‘Rather than averaging them — I just drove a bunch of people screaming from the room by showing a proof of the problems of averaging, so I promise I won’t, but believe me, it sucks ‘
Certainly averaging is not even statistically valid. In a proper meta-analysis a poll with fewer data points has to be weighed lower.
But there are so many systemic problems with polls that at best they can only give you a general idea of what people think.
I trust yard signs and the attendance at the candidate’s events. When Mitt is drawing 10,000 and turning folks away because they are out of room, while the Won is getting 800 including the paid union thugs, Blood is in the water.
Axelclod reminds of Custer’s lieutenant talking about a comeback in minute now.
Riiiighttt…..
Give the people light and they will find their own way.
Really a pleasure to read Charlie, thanks. And just what I needed today. Great tone and a great sign.
It frankly infuriates me, and it must you too, to hear the pundits spend so much time on discussing polls but they never discuss the underlying premises such as sample size, party ID and who is really picking up the phone.
Thanks for this. I can get out of bed now.
thought you’d appreciate this..:
http://www.youtube.com/watch?v=P5b7OjXrYhc&feature=relmfu
http://www.youtube.com/watch?v=jBR9RRKvP1o
“In each case they asked people who had voted for Obama if they were going to again.”
From what I’m reading, I assume you mean again in THIS election.
Vote early vote often.
The discourse at HuffPo is not at this level…
I wonder why…
“Keep Calm and Finish Him”
That’s ok. But “voting is the best revenge” is somehow beyond the pale.
Ding-a-ling. One of those was said by a sitting president (or facsimile, anyway).
“That’s ok. But ‘voting is the best revenge’ is somehow beyond the pale.”
Some maybe happy to seek revenge on their fellow Americans, seeing them as capitalist-roaders, running dog dupes of the TEA Party, kulaks, etc. who must have evils inflicted upon them. Let them vote.
So, Yes, a political party encouraging it adherents to seek revenge on their neighbors (hmm, sound like something from the 20th century) is beyond the pale. Finishing Obamunism and Obama’s career is a positive good.
I, for one, will be happy when Obama’s finished.
Very Interesting article. While I am not a math person it was wonderful to finally read something where I could read all the comments on a post that were intelligent, thought provoking and stayed on subject. Math aside your article here was intelligent and interesting. In the end the polls will tell it all and I can’t wait for it to be over. The most important thing now is to Vote!!!!!!!!!!!
Romney can win this if we have a good turn out at the polls. Here in southern Florida the obama supporters are standing in lines two blocks long to cast their vote. We must meet the same determination.
Vote for America not Obama. We can’t have both.
“Europe isn’t working in Europe and Europe won’t work here.” -Mitt Romney-
Let’s roll…………Romney/Ryan 2012
This article and the comments have given me hope. I trust Obama will have many defections from his ranks. It is amazing that he has so much support. I pray that God will open eyes and minds to see glimpses of truth and to act according to the dictates of conscience.