Nate Silver's The Signal and The Noise

Nate Silver's been in the news a lot these last few days: looking at some stories, you'd think he'd won the election, not Mr. Obama. A statistician, his rigorous polling analysis riled, then humiliated political pundits, whose imaginary political horse-race was rejected by Silver's cold, hard numbers.

And what numbers they were. His "prediction"--though really just the most likely probability among many scenarios offered by his model--nailed the electoral college total on the night.

I've just read his book, The Signal and The Noise, and while it isn't as approachable as you might expect, that's what makes it interesting. Silver's insistence is that the quintillions of bytes of data at our disposal actually make prediction harder, not easier, and that we should not place too much stock in forecasts. Nor is it enough for us to to indulge brilliant but unanchored insights of the type offered by Malcolm Gladwell and Freakanomics. Planning for the relative probabilities of many possible outcomes is more useful than going all-in on specific predictions.

He takes us through our failures and successes. We're still lost at sea when it comes to disaster prediction, but (as he demonstrated last week) quite good at forecasting election results. Silva homes in on the work of mathematician Thomas Bayes as the right approach to understanding how statistics are most rationally interpreted to human ends.

By the end, though, it's surprising how unfulfilling it all seems. The Insight Industry's easy answers sure do go down easier than a dissertation on Big Data's ineluctable disinterest in them. Here, we are instead shown how prediction-makers screw up by treating their work as esoteric, by hiding uncertainties and inexactitudes to appear precise and decisive.

Which brings us back to the pundits, red-faced but curiously immobile in the stew of their failures. I don't quite buy that accepting these complexities will fill the chasm between their inanity and numerate reason. Though a place that seems rich with human promise, what emerges from the dark is a kind of interpretive accounting: much less interesting than whatever probabilities may finally be offered, and harder to sell than pompous old men just telling their peers how it is. It's such a difficult story to tell: even if you accept it, it's always less fun than the one where Joe Scarborough was made a total fool of by a brilliant young statistician.

I'm not helping, I know. But the pundits aren't going anywhere.

The Signal and the Noise: Why So Many Predictions Fail-but Some Don't [Amazon]



  1. And good for Nate Silver for doing quality work, and also being an entertaining enough writer and speaker that his predictions achieved some notoriety.

    I also feel that others who predicted nearly as well have gotten less press. I’ve followed this guy for years, he’s done a great job including this time: 
    It is definitely still interpretation, and not an exact science. It is just is so much better than what otherwise passes for analysis in the major media.

  2. Look, I like Nate, but Prof. Sam Wang has a better record over the last several years:

    Also, Nate’s chapter on climate change in his book is shockingly bad,. For more information see:

    FiveThirtyEight: The Number of Things Nate Silver Gets Wrong About Climate Change

    1. I read that first article there on thinkprogress, and to be blunt, I find it awful.  It appears that the author suffers from a bad case of an eagerness to worship specialists and ostracize outsiders who are attempting to develop interdisciplinary worldviews.  Who can blame Nate for trying to look at the big picture?  It’s what James Maxwell did when he combined light to electricity and magnetism.  What you will find is that when a problem is ill-structured and “wickedly” complicated, the specialist’s approach becomes just another perspective (in other words, possibly wrong!), for there oftentimes is no right answer given the data (there are only good and bad arguments), and all that we can hope to do is to map out all of the various lines of argumentation in order to size up where we personally fall.

      I am a Democrat myself, but this never-ending nonsense I see coming from Democrats which conflate these ill-structured, wicked problems with the highly structured, single answer problems of problem sets which we all endured in high school and college, is causing a lot of problems.  It’s naive.  The more proper approach would be to brush up on your Vygotsky, and enter the domain of computer-assisted collaborative learning — where we focus on perfecting the visualization of argumentation, instead of pretending that one side is right and the other is not.

      Also, keep in mind that climate modelers accept all of the assumptions and hypotheses offered to them by astrophysicists, in order to construct their models — such as the hypothesis that gravity is the universe’s dominant force and the assumption that cosmic plasmas can be essentially modeled differently from laboratory plasmas — as if they are superconductors with no electrical resistance (not in the lab); as if they cannot sustain electric fields, since their lack of resistance leads them to (theoretically) instantly charge-neutralize (not in the lab); and as if the magnetic fields are “fossils”, instead of the result of electric currents (again, not in the lab).  All of these approximations have been adopted by astrophysicists despite warnings by Nobel laureate Hannes Alfven, the originator of the MHD cosmic models that astrophysicists use to this day.

      The astrophysical models simply don’t work, failing to identify 95% of the universe’s matter — an astounding failure which would fail to get published in other scientific disciplines — and failing to explain why the interstellar medium is highly filamentary (a feature common to plasmas which are electrically conducting).  These mistakes are likely related, considering that 99.999% of what we see with our telescopes is matter in the plasma state (few seem to realize this).

      Who imagines that a specialist approach, hyper-focused upon climate modeling, would ever uncover such debates lingering unseen beneath the debate we see over climate modeling in the papers?  Right …

      The specialist view is that those apparently faulty assumptions and hypotheses somehow become “cleansed” the second that they cross a disciplinary boundary (like astrophysics –> climate modeling).  In the real world, mistakes in our astrophysical models trickle down into our climate models in ways many of us simply don’t yet understand.  But, of course, perhaps some people do understand, and the specialists simply refuse to listen to them …

      I find Nate Silver’s warnings about the role of uncertainty to be completely accurate.  His prediction paradox seems to almost perfectly explain the problem of humility which we see amongst the climate modelers.

      1. Amazing. We must have read completely different articles.

        The one I read was by a man who actually greatly looked up to Nate Silver, but who was extremely disappointed that he didn’t bother to learn the subject matter well enough to recognize that he was making strawman arguments that had already been repeatedly shown to be false.

        I don’t know where you’re getting the idea that an economist learning about climate science in a few months and telling all the scientists why they’re wrong is remotely like James Maxwell “combining light to electricity and magnetism.” The first is a fairly typical example of someone assuming that their own field of study should just trivially map onto any other field, the second was a scientist who spent his entire life studying the subject in question.

        Also, you are barking up some bizarre tree with your notion that the conclusions of climate study rely on the study of “cosmic plasmas” and dark matter. Is cosmic plasma referred to in any climate science article? No of course it isn’t — plasma cosmology is even categorized in Wikipedia as “fringe physics.” Where the hell are you getting your information from?

      2. This makes no sense (and I’m an astrophysicist). Astrophysics and climate modeling both use physics to model complex systems, but the big unknowns in astrophysics are irrelevant. Climate change models don’t care that we don’t know what dark matter and dark energy are, because dark matter and dark energy have a negligible effect on the climate.

  3. Fact (statistical analysis) vs. Faith (The Right)

    It is always refreshing/sad to see conservative theocrats loose it when faced with science/reality. Some of them hold out longer than others and will fight to prove their faith is “right” long after facts prove otherwise. Trump for instance has a religious belief that nobody can see his origami-esque comb over.
    The same is true of most Fox News viewers… many of them believe that Satan delivered victory to Obama… never mind the fact that there are more “other voters” than there are old white men.

      1. I don’t think I’ve ever seen Mrs Stewart (though I’m happy to assume she’s a “babe,” if that’s the word), but Jon’s voice is a far cry from high and reedy like Silver’s.  Though he has excellent range as a mimic, I find his lower register to be FM-quality chocolate.

    1. Considering the fact that Silver is an out gay man, I imagine that the pool of babes from which he would otherwise be scoring remains yet unrippled. Surely this presents an opportunity for you high-voiced, frail, nerdy types to step up your games.

  4. I don’t think it was much of an issue between sound Science and wishful thinking.

    I actually think it’s more likely that the GOP pundits KNEW their numbers were completely imaginary, and yet that’s not the point.

    What matters in modern political races is keep yelling your message as loud and often as possible, so people will start believing it. The objective of polls is first and foremost to influence undecided voters.

    “If you tell a lie big enough and keep repeating it, people will eventually come to believe it. The lie can be maintained only for such time as the State can shield the people from the political, economic and/or military consequences of the lie. It thus becomes vitally important for the State to use all of its powers to repress dissent, for the truth is the mortal enemy of the lie, and thus by extension, the truth is the greatest enemy of the State.” ~ Joseph Goebbels

  5.  I still think that’s why so many in the Right-leaning Media were seemingly shocked on Election day.  They believe that if you keep telling the lie over and over, it becomes true.  Then it didn’t work and it shook their entire worldview. 

Comments are closed.