Republicans say Nate Silver is a biased liberal, and reading his blog this election cycle makes it seem that way to me even.
But I would like to know - is there a Republican analog to Nate Silver? Someone who predicts 90% of the races at the end, does the meta-math, and has a computer model?
If there is no Republican analog, can you possibly say Nate Silver is liberal or biased? Does using a computer model and math, and blogging about your results make you a Democrat?
It's just a time tested tactic of discrediting a result you dislike in order to prevent your constituents from becoming apathetic and potentially not voting.
If Republicans actually concede that Nate Silver's results are legitimate and unbiased (if they are), than the party would have a much harder time raising money and getting out the vote.
It's better to frame it as a conspiracy so less people take it seriously.
If you follow much of his writing/tweeting, he doesn't disguise much opinion, which has a liberal slant - but the pure math of his project is pretty hard to argue against.
Well the proof is in the pudding. His performance on the 2008 and 2010 election indicate to me he's not some partisan hack. He accurately predicted all of the EVs Obama obtained in 2008 with the exception of Indiana (which was quite close - only a few hundred votes in the margin IIRC). He did exceptionally well in 2010, coming in 3rd for accuracy behind Survey USA and PPP (A democratic shop, incidentally) in house race prediction.
Silver is politically liberal but he's a stats guy at heart.
The republican equivalent would be Unskewed Pollshttp://unskewedpolls.com/ which re-weights partisan affiliation of polls to match Rasmussen (The result: Romney is WAY ahead). It's author isn't a big stats guy and goes more off gut. This comparison isn't entirely fair though. Republican polsters and aggregators are probably private, working inside campaigns and not for a public news organization.
You can be politically liberal or conservative and not have it directly spill into your work.
Rasmussen is well known to have a decent conservative slant because of the way they select their population and perform their polling. I'll leave it as an exercise to the reader to decide if this is intentional or not.
While Rasmussen seems biased in the strict technical sense, it's otherwise relatively low-variance high-quality data, so the bias can be corrected for, leaving you with useful information.
No, not directly. RealClearPoltics (http://www.realclearpolitics.com) aggregates polling data, but it doesn't analyze it. Sean Trende does some analysis of the polls, but he is a lawyer and political scientist, not a mathematician.
Other than that, pollster Scott Rasmussen (http://www.rasmussenreports.com/) has a solid track record forecasting outcomes and is the closest thing to a Nate Silver analog.
Edit:
> does the meta-math
BTW, I think one of the most interesting things that will come out of this election is the impact of cell phones on polling data. The meta-math is based on the sampling models for each pollster, and the models have to take land line / cell phone demographics into consideration. Many pollsters only call land lines and try to accommodate for the demographics in their polling, but it's not certain how accurate their models are.
There has been a shift in land line demographics in the last four years. Lower income people used to have land lines only and mid to upper income people had land lines plus cell. land line usage has dropped a lot. In recent years, lower income people tend to have cell phones only. It will be interesting to see how the the election results compare to the polling data.
I don't think he has a leaning one way or the other - he tries to be unbiased, and arrive at data-driven conclusions. What have you seen from Nate that makes him seem a "biased liberal" to you? I'd be very interested to know. Also, we'll see in about a month how good his meta-analysis is.
He states straight out in his FAQ that he supports Obama.
"What is your political affiliation? My state has non-partisan registration, so I am not registered as anything. I vote for Democratic candidates the majority of the time (though by no means always). This year, I have been a supporter of Barack Obama."
And the next question -
"Are your results biased toward your preferred candidates? I hope not, but that is for you to decide. I have tried to disclose as much about my methodology as possible."
Sure, but I'm having a hard time imagining the answer to either question has changed. If it had, I'm sure he's capable of having that information removed from a post bearing his name.
His name is right there on the FAQ. He's a media figure, of at least some note, who makes his living predicting political outcomes. If he had changed his stand, he would have the information removed from something bearing his name. It has direct impact on perception about his work.
Disagree; you can't even get to www.fivethirtyeight.com anymore (it redirects to NYT) and Silver is pretty transparent; I think if that affiliation was dependable information, it'd be somewhere in the masthead of his blog.
I don't really see why it's wrong that Nate has political opinions. The alternative is Jim Lehrer who values his "objectivity" so much that he doesn't even vote at elections.
Sometimes, facts favour liberals, sometimes they don't.
I can't put a finger on it, but you know which way he hopes things go, and I think that shows through, despite his attempts to stay unbiased, and despite the NYT style guide. I think a TV news anchor probably gives away his bias with body language too.
I like to tell myself that too, so I regard his work with at least a little skepticism, even though I gobble it up and bet on it.
Humans who are not biased probably do not exist. Or at least do not care or know about politics in the slightest. I think there is nothing wrong with that. That doesn't invalidate his work.
I know too much about human biases to even try to claim that his do not influence his interpretations and work at all – I just don't think that's possible – but that is general and in all likelihood inevitable problem. I think it's pretty clear, however, that it's in his best interest to make accurate predictions, not partisan predictions.
I do not think it's really helpful to start out with that question. It puts too much focus on what's actually important. Yeah, bias can be hard to detect, but a little bias does not make models useless and when there is a lot of bias you should actually be able to point at what's wrong with the model instead of having to resort to some diffuse claims of bias.
Writing this made me think of Elisabeth Noelle-Neumann (http://en.wikipedia.org/wiki/Elisabeth_Noelle-Neumann). I disagree with many of her views and I think it would be fair to call her biased after reading her most famous work (The Spiral of Silence), but I do respect her work. There is a difference.
He is a biased liberal and reading any of his commentary should be viewed through that lens. He doesn't even make much of an effort to hide it, in 2008 on 538 he was basically rooting for Obama.
However that doesn't mean his models are wrong. When they show that Republicans are going to win (or have improved), he states that just as clearly as when the same is true for Democrats.
I think Republicans are more likely not to want to mention or distribute condoms to teens in sex education, but I don't regard this as quite the same general disregard for science and math.
This is a little silly. Intelligent people could easily vote for the candidate of the party that is toughest on public-sector unions for one office and the candidate who is most supportive of a carbon tax for another office, believing both that public sector unions and their pension commitments are out of control and in anthropogenic global warming, and make those choices dependent on the impact those beliefs have on the role. The governor of Wisconsin is poorly situated to influence climate politics, while control of the Senate might not have much influence on whether employees get defined-benefit pensions.
The political parties themselves would very much like you to believe that you're along with them for the ride on all their issues, and this no doubt accounts for much of the strongly held beliefs of both Democrats and Republicans. Push comes to shove, most people have not given much thought to the issues that politics have the most control over.
The Republicans are the ones responsible for the ridiculous union pensions, because those were what they offered in concession during the 60s and 70s instead of paying the workers a fair wage at the time. So why the hell would you vote Republican if you were against the pensions?
Is that a serious question? Scott Walker is a Republican, not a Democrat. Also, the GOP of 2012 bears little relationship to the GOP of 1965. Barry Goldwater was practically on the fringe of the 1960's GOP. What party would Nelson Rockefeller belong to in 2012?
You can vote a specific interest, you can vote a straight party line, you can vote out of ignorance. But one thing that makes zero sense is to vote party-line in 2012 to punish a party for what it did 60 years ago. Doing that has literally the opposite impact as you'd ostensibly intend.
It's not about punishing one party for the past, it's about which party would be most likely to make future spending commitments without raising taxes now to pay for them. (Probably the one that's been doing it consistently for the past 50 years and is campaigning on a platform of doing it now.)
> The Republicans are the ones responsible for the ridiculous union pensions, because those were what they offered in concession during the 60s and 70s instead of paying the workers a fair wage at the time.
What are you babbling about? Public employees weren't unionized in the 60s and early 70s.
The greatest trick political parties ever pulled was making this kind of thing part of their images.
With only two choices, people are roped into supporting all kinds of things that make no sense for them. But you get "My party believes in science!" and "My party believes in God!" and it becomes part of people's political identities. And then all sorts of other issues are piled on top of it and you find yourself wondering how we got here.
If it truly is science vs. religion, then the Republican party will die and be reborn as something new. Religion loses to science over the course of history.
Is that intended as a recommendation of a scientific source about long-term trends cross-culturally? I got a big volume of the collected "future history" stories by Robert Heinlein out of the library a year ago, and I was disappointed by how poorly those stories have aged. I also figured out, by rereading Heinlein (a favorite author of mine when I was a kid) that he never informed any of his stories with much perspective from non-Western cultures (although I give him credit for being aware of that kind of issue). Since I spent three years in the 1980s and later another three years spanning the turn of the last century living in a non-Western country, I don't turn to Western science-fiction stories anymore for my social commentary. I like to read real science based on verifiable history to puzzle out social trends. I can recommend a source
Not extremely well. You've cherry-picked your data to support your theory.
Specifically, I notice you don't touch on those who reject vaccinations, practice homeopathy, or believe Feng shui goes beyond aesthetics.
As someone living in a very large, liberal city and hailing from a rural, conservative town, I can tell you the Republicans of the world are not the ones debating if vaccinations cause autism. The ones I know think it's moronic & hideously irresponsible not to vaccinate your child.
> I think Republicans are more likely not to want to mention or distribute condoms to teens in sex education, but I don't regard this as quite the same general disregard for science and math.
It's a fairly settled issue in social science, though, that abstinence-only sex ed doesn't work. Not a hard science, perhaps, but still a science in that it's data-driven and rejects hypotheses based on new data.
> It's a fairly settled issue in social science, though, that abstinence-only sex ed doesn't work.
So? The alternative to abstinence-only sex ed isn't perfection, it's a bunch of stuff that doesn't work all that well either.
Your links are all behind paywalls. That said, they're probably just like the other work on the subject. Their "measure of success" is curious at best. (No, reducing teen pregnancy isn't the only goal.)
As to "telling teaching inaccuracies", no other teaching is held to that standard, so why is abstinence-only sex ed held to that standard?
Note - I think that abstinence-only sex ed is a bad idea. My point is that the arguments are not "scientific" or fair. That said, they are doing a great job in discrediting science. If that's your goal....
> As to "telling teaching inaccuracies", no other teaching is held to that standard, so why is abstinence-only sex ed held to that standard?
Because kids going around with misconceptions about what really caused the American Revolution is less dangerous than kids going around with misconceptions about reproductive issues.
> Because kids going around with misconceptions about what really caused the American Revolution is less dangerous than kids going around with misconceptions about reproductive issues.
Umm, if it doesn't matter whether it is true, why are we teaching it?
And no, sex ed isn't the only dangerous thing that we mis-teach kids about.
But, thanks for demonstrating your priorities.
However, I'm still unclear on why I should be paying for them.
> Does following the scientific consensus on issues like anthropogenic global warming, evolution, and sex education make you a Democrat?
The correlation is with misunderstanding all three and coming to bogus conclusions.
Let's take evolution. There are lots of folks, typically Democrats, who insist that it is fundamental to all of science. (Hint - that's clearly false. Disagree? How does evolution matter outside biology? Too hard? Name three biology subfields that don't work without evolution? Too hard? Name three things in biology that were predicted using evolution?)
Democrats like to think that they're scientific, but ....
Let's take the suggestion that state of mind affects conception in humans. Clearly false when Akin said it, but what does everyone say to a couple trying to get pregnant? Yup - "you've got to maintain a positive mental attitude." (I don't know whether mental attitude matters - I'm just pointing out the double-standard.)
There are lots of folks, typically Democrats, who insist that it is fundamental to all of science
You can't, on the basis of zero real evidence, dismiss the theory that best fits the known facts, without causing harm to the understanding of the scientific method as a whole. So it is not that the one theory is so critical to all of science, but the act of throwing it out because it offends religious sensibilities is damaging.
I don't see the point in finding three of anything (is this Super Mario Bros?) but I was reminded of a great prediction made possible by the theory of evolution and thought it would be fun to share with people interested in the history of science.
There are lots of folks, typically Democrats, who insist that it is fundamental to all of science. (Hint - that's clearly false. Disagree?
Of course evolution is not fundamental to ALL of science. For instance chemistry is based on current measurements, not what happened in prehistory. But you are overstating your position.
How does evolution matter outside biology?
Evolution matters to geology because specific fossils are often used as a cross-reference for the age of specific rocks. Indeed such "indicator fossils" are important economically because of how they are used in oil exploration.
There are also applications of fossils to astronomy, for instance fossil corals have been used to verify how the number of days in a year have changed over time.
Too hard? Name three biology subfields that don't work without evolution?
1. The study of evolution itself.
2. Ecology. A very large part of ecology is based on understanding the Evolutionarily Stable Strategies of living organisms. For everything from mating to parenting behaviors, what animals do is explained by how it is selected for evolutionarily.
3. Taxonomy. We classify living organisms based on the best guesses we can come up with for where their ancestors were in the evolutionary tree. We reconstruct those trees using everything from physical characteristics to molecular genetics. NONE of it makes any sense without common descent.
Too hard? Name three things in biology that were predicted using evolution?)
1. (Prediction made by Charles Darwin. Confirmed many decades later by Leakey's discovery of Lucy and other fossils.) Fossils for early humans would be localized to Africa. Based on homology, humans are clearly an ape. Since existing ape species are mostly found in Africa, our ancestors are likely to be found in Africa as well.
2. (Prediction made by Edmond O. Wilson, later confirmed by him) In slaver ants, male and female babies should be treated equally if there is a good supply of slaves, and should be treated unequally (as they are in most ants) if there aren't. The cause of the unequal treatment of males and females in an ant colony is that a quirk of genetics causes females to share 75% of genes with sisters, and only 50% with parents or brothers. Therefore female ants are willing to devote their lives to the production of more sisters. This causes preferential behavior for female larvae. In slaver ants this disappears because the slaves are controlled by the queen who is equally related to sons and daughters. Remove the slaves, and the daughters express their genetic preference.
3. (Made and confirmed many times by a variety of people.) Pylogenetic reconstructions of evolutionary tree are robust to the specific set of characteristics chosen for that reconstruction. The branching in characteristics reflects the actual ancestry of the animals, and so two sets characteristics will tend to reveal the same ancestral branching pattern. (There are exceptions where common evolutionary pressures cause disparate animals to have similar adaptations. But these are well-understood.)
Now let me guess. You don't believe in evolution, believe that you understand science, and it surprises you that I'd actually find your challenge easy. If so I would urge you to open up your mind, go to http://www.talkorigins.org/, and educate yourself.
> Now let me guess. You don't believe in evolution,
Wrong.
> Evolution matters to geology because specific fossils are often used as a cross-reference for the age of specific rocks.
Evolution doesn't tell you how old a fossil is. Evolution says that there's a path, but doesn't say how long each step took.
> There are also applications of fossils to astronomy, for instance fossil corals have been used to verify how the number of days in a year have changed over time.
You're confusing evolution with fossils.
Slaver ant behavior isn't dependent on evolution. (Also, everyone believes in selective breeding. Evolution is more.)
Evolution tells you that a fossil of a given type cannot be older than ancestral types, and is unlikely to be younger than descendant types. That sandwiches the age fairly nicely.
But if you want a prediction that applies to geology, evolution predicted - long before physical mechanisms were known for it - that specific land masses must have been connected in the distant past. And even put approximate dates on the connection. This lead to endless speculations of land bridges, etc - and also a crazy theory called continental drift. (Which was eventually proven correct.)
There you have it. Relevant predictions in the field of geology which depend in critical ways upon Darwin's theory of descent with modification.
As for slaver ant behavior, you asked for a prediction from evolution. The detailed study of ant behavior was undertaken exactly because evolutionary theory predicted specific behavior. Thus it is a prediction of evolution, regardless of the fact that you can try to explain it with selective breeding. But that said, the time it would take selective breeding to match the exact ratios predicted by theory is sufficiently long that you wind up with evidence of evolution. Evolution further predicts that the actual genetic mechanism by which this behavior is controlled is likely to be conserved across many ant species. I do not know whether this has been studied (it probably has, but I cannot confirm that), but there is another prediction for you.
> But if you want a prediction that applies to geology, evolution predicted - long before physical mechanisms were known for it - that specific land masses must have been connected in the distant past.
Evolution did no such thing.
I can't tell if you're confusing fossils (same creature, same time, must have been same location) or thinking that land masses "evolve", but either way....
> The detailed study of ant behavior was undertaken exactly because evolutionary theory predicted specific behavior.
Actually, it doesn't (there are non-slaver ants, so if their behavior is also evolutionary, you have evolution predicting X and not-X).
> Thus it is a prediction of evolution, regardless of the fact that you can try to explain it with selective breeding.
The claim was that evolution was necessary to predict something. If said something can be predicted by other means, evolution is not necessary.
As I've written before, I believe in evolution, but most of the arm-chair supporters offer false arguments.
I can't tell if you're confusing fossils (same creature, same time, must have been same location) or thinking that land masses "evolve", but either way....
This line indicates, based on the distribution of related species today, what islands were connected with Asia in the last versus which other nearby ones weren't. The line was drawn before fossil data or underwater topography could confirm or explain the past connection. Without the hypothesis that similar species are similar because they share a recent ancestor, this data on extant species would not have lead to any conclusions.
Moving forward, in decades after evolution and before plate tectonics, there were many examples of scientists drawing connections between once connected masses based on related species - but not the same species in both places. For instance both India and Madagascar contain rich lemur fossils that are different but clearly related. That relation was explained by a past land mass called http://en.wikipedia.org/wiki/Lemuria_(continent). The conclusion was incorrect, but it is true that Indian and Madagascar once were connected and did exchange fauna, including lemurs.
Actually, it doesn't (there are non-slaver ants, so if their behavior is also evolutionary, you have evolution predicting X and not-X).
You clearly have failed to understand the example.
Ants generally feed female larvae something like 3x as much as they feed male larvae. (I think the ratio is 3 to 1, but this is pulled from a course 20 years ago, and I can't swear to that figure.) Wilson came up with an evolutionary explanation for this. Based on this explanation HE PREDICTED that if the queen was in control of the care of the young, this discrepancy would disappear. Slaver ants provide a test case, because (at least in some species) the queen does control the behavior of the slaves.
In slaver ants he found, AS PREDICTED BY THEORY, that when the queen was in charge (ie there were slaves), male larvae received as much food as female larvae. But when a slaver colony had no slaves, they revert to the typical 3 to 1 ratio that is common among ants.
There is no X and not-X. There is behavior when queen is in control versus behavior when workers are in control. Non-slaver ants differ from slavers in that the workers are always in control of care of the larvae.
> Thus it is a prediction of evolution, regardless of the fact that you can try to explain it with selective breeding.
The claim was that evolution was necessary to predict something. If said something can be predicted by other means, evolution is not necessary.
Scroll back. Your demand was, "Name three things in biology that were predicted using evolution?" This is, in fact, an example that was predicted using evolution. The existence of other possible explanations does not change the fact that evolution was, in fact, used to predict this particular thing.
As I've written before, I believe in evolution, but most of the arm-chair supporters offer false arguments.
This is true of most topics. A lot of armchair supporters know the orthodoxy, but do not understand it and cannot defend it. People believe in all sorts of alternative theories usually have an easy time dispatching these armchair supporters because while they are wrong in a big way, they actually know more than the armchair supporters do.
But I would like to believe that I know more about evolution than the average armchair supporter. After all your average armchair supporter hasn't done things like audit a graduate seminar about evolution theory. :-)
Just a nit-pick: Continental Drift is the theory that the continents were floating on the ocean and has been replaced by plate tectonics. It's not right to say that theory is correct, even though it does align with the facts about Pangea, etc.
If you're going to nitpick, choose your words carefully. :-P
http://en.wikipedia.org/wiki/Continental_drift puts it, Continental drift is the movement of the Earth's continents relative to each other by appearing to drift across the ocean bed. Note that "appearing to drift" and "floating on the ocean" are very, very different things. A trivial comparison of rocks and water demonstrates that "floating on the ocean" would be inconceivable for something natural that is as big as a continent.
That said, plate tectonics is a much better theory. But I would say that it has the same relation to continental drift that Newtonian mechanics has to Special Relativity. The theory is clearly wrong, but the succeeding theory explains why the less detailed theory appears to work.
Furthermore plate tectonics is itself a theory that in due course of time will be (and to some extent already has been) replaced by something more complete. For instance consider Hawaii. We have a chain of islands that is produced by the movement of the Pacific plate over a hot spot. What creates these hot spots? What role do they have in pushing plate tectonics? A theory that explains all of this will be undoubtably differ in detail from the original theory of plate tectonics, though it will have to explain why plate tectonics seemed to work.
Are you being deliberately obtuse? The reason prospective parents trying to conceive are told to keep a "positive mental attitude" is because it can be a long, frustrating process, and it doesn't help if you're too frustrated to want sex.
> Hint - that's clearly false. Disagree? How does evolution matter outside biology? Too hard? Name three biology subfields that don't work without evolution? Too hard? Name three things in biology that were predicted using evolution?
Gish Galloping wins you no support. It's a dishonest tactic that attempts to flood the opposition with questions that require more time to answer than they do to ask. It's a Denial of Service attack, not an argument.
> Clearly false when Akin said it, but what does everyone say to a couple trying to get pregnant? Yup - "you've got to maintain a positive mental attitude." (I don't know whether mental attitude matters - I'm just pointing out the double-standard.)
Because Akin was minimizing rape. He was trying to de-legitimize the experiences of victims of a violent crime. Do you really not understand how people get upset over that?
No, he wasn't. He was trying to respond to the "what about rape" argument wrt abortion.
I'm pro-abortion, but he's correct in saying that only a small fraction of abortions have anything to do with rape. (That said, age of consent and under the influence rape don't involve "negative mental state" that his argument requires.) I don't think that that fact is relevant, but it's dishonest to say that it isn't true.
Three of those are policy issues (how much testing should be done on GMO's, how risk adverse should we be on nuclear power plants given the potential for shoddy construction like in Fukushima, where should we build energy plants and refineries). On these issues, the democrats are being, ironically enough, conservative, as in risk adverse. They are not denying that you _can_ grow GMO's, that you _can_ split the atom, or that you _can_ perform hydraulic fracking (which Obama has supported, so I don't see how this is a democrat issue). This is in stark contrast to the previous issues where the mere existence of a phenomenon (evolution, global warming, and the big bang) is denied. With luck, global warming will move into the policy disagreement realm, and out denial box soon.
Anti-vaccination is a fringe movement, and party independent (Michelle Bachmann got on the anti-vax wagon). The democrats aren't putting anti-vaxxers on the house science committee.
I'd deny that anti-vaccine idiocy is part of the Democratic platform the same way anthropogenic global warming denialism is part of the Republican party platform. Do you have any evidence it is?
Reminds me of the meme "reality has a well-known liberal bias". I don't know or suspect whether that's true or not. But I have noticed personally in life that whenever someone is both intelligent, educated, is compassionate/empathic and does long-term thinking, that person also seems to be "liberal", progressive or left-leaning, politics wise. Perhaps not always. But most of the time, based on my observations. I think if you have those traits but take away compassion/empathy, and focus more on short-term/short-range thinking, you're more likely to be "conservative", or right-leaning.
What would you say about the tens of millions of Americans who are educated, religious, and give hundreds of millions of dollars to build schools, clinics, and missions around the world?
Are they short-sighted or lacking compassion?
I think this kind of binary political thinking is the most insidious thought virus we face. Far more dangerous than whether or not someone believes in evolution, because it separates us and makes us believe that the "others" have traits that they really don't have. It's awful.
You write as if you've somehow refuted or contradicted my words, but you haven't. I recommend looking again, more closely, at what I said. I choose words carefully.
But I do understand the idea that religious people can be compassionate and give money away. So can atheists. So can intelligent people. So can both more and less educated people. It's orthogonal. And I never cited religion. And I was speaking in general terms based on direct, personal observation. Also I for one do not engage in "binary" political thinking -- not when I'm actually thinking internally -- though it's often useful when speaking in order to average out details and identify patterns and then express them concisely in public conversation.
And as an additional point, if by 'mission' you mean a religious mission, then I'd say that I don't consider that to be a good thing. It's a bad thing. It spreads superstition, fear, make believe, childish mental states, obedience to arbitrary authority, silly rituals and fuzzy irrational thinking. Yes it also teaches relaxing and addictive chants and tunes, and feel good phrases. And the good parts, anything dealing with peace, compassion, charity, with community interaction, can all be done without the negative that comes from the religious baggage attached to it. Also, like your complaint about binary thinking being a bad thing. I don't think education is a binary quality. You're not either educated or not. There are many intermediate degrees and shades of grey, and types. So while I do think someone can be somewhat educated and religious, I don't think they're extremely well educated or intelligent. At least not both. Otherwise one would see that that particular emperor wears no clothes. But of course this is just the conclusion reached by many of us, not all of us. To each his own.
Yes. Precisely. It's orthogonal. Like compassion and conservatism or short-term thinking and liberalism. Honestly, I disagree that those things are at all related. I think that's the sort of thing you get in political echo chambers, but not in reality.
I meant mission more in the sense of centers intended to relieve poverty, hunger, etc, but it is a word with different meanings, so I see your take on it as well.
I do think, though, that if you think belief in religion is a matter of whether the emperor has any clothes that you're looking at it wrong. There are deeper meanings and purposes to faith that many intelligent and well-educated people choose to pursue. They aren't doing it blindly, they know what they're doing, and they have their reasons.
I'm strongly agnostic (to the point of being ridiculous about it), but I've encountered a wide enough variety of religious people to respect their views.
Nate Silver came from the fantastic world of the statistical analysis of baseball, specifically Baseball Prospectus. You can check out his body of work here (some articles require a subscription):
His big contribution was a model used to predict player statistics. When he first started writing about politics, I assumed the name was a coincidence.
The book is pretty good, sort of like a cross between Moneyball and Freakonomics. It's less political than you'd think it'd be, and as an attempt to preach the gospel of Bayesianism, should be congenial to most HN readers.
I like that your site shows the 2008 history right below the 2012 (although it is distracting that the scale in time is different between them). I couldn't even find that comparison when I looked for it on Nate Silver's site.
I'm curious about the methodology as well. Why didn't you choose more standard gaussian statistics for the differences, instead of the combination linear/step model? How much difference would it make?
In this plot http://imgur.com/BAEie you can see the difference between a gaussian model and the model I use for the translating state poll differences into probabilities of winning that state.
I prefer to have the flat region of 50% probability when the state poll difference is small, which the gaussian model doesn't have, to take into account (a) the various errors in the polling data and (b) the likelihood that people's minds can change by the time the election comes.
As to how much difference it would make, with the current model, Obama's probability is now at 81%, whereas with the Gaussian model it would be around 95%.
Doing the math as "right" as he does it is incredibly hard. You can tell because academics do it totally wrong on a regular basis, which leads to many garbage papers, which thankfully sometimes get caught in peer review.
It's true he's a good writer, but the proof is in the pudding. He could be foaming at the mouth, but if he's predicting the outcomes, people will tune in.
> Silver's final 2008 presidential election forecast accurately predicted the winner of 49 of the 50 states as well as the District of Columbia (missing only the prediction for Indiana). As his model predicted, the races in Missouri and North Carolina were particularly close. He also correctly predicted the winners of every U.S. Senate race. The accuracy of his predictions won him further acclaim, including abroad,[49] and added to his reputation as a leading political prognosticator.[50]
If he has a good method, if he's on a right track, that method should generally be improving with more work and data so recent predictions should generally be better than distant ones.
Also, "You're only as good as your last envelope."
But I would like to know - is there a Republican analog to Nate Silver? Someone who predicts 90% of the races at the end, does the meta-math, and has a computer model?
If there is no Republican analog, can you possibly say Nate Silver is liberal or biased? Does using a computer model and math, and blogging about your results make you a Democrat?