Monday, October 29, 2012

Tim Stanley's Anti-Silver Tantrum

In an earlier post I mentioned Tim Stanley's piece "Nate Silver is partisan and wrong. The voters will decide Romney v Obama, not The New York Times."*

I have better things to be doing, but Stanley's piece really is terrible, so I couldn't help myself...  Here it is, paragraph-by-paragraph:

Stanley writes:
In the history of presidential elections, has there ever been such an effort by one side to poll their way to victory? While the Republicans have spoken this season about jobs and debt – willing themselves to a moral victory – the Democrats have talked constantly about how well their guy is polling in one or two states. The goal is to create a sense of inevitability, to convince the public to vote for Obama because he’s a winner and who wouldn’t want to vote for the winner? We’ve witnessed the evolution of polling from an objective gauge of the public mood to a propaganda tool: partisan and inaccurate.
Is there anything sketchier--or more irritating--than a rhetorical question?

Rhetorical questions, of course, are thinly-veiled ways of making statements. Mr. Stanley here makes roughly the following assertion:

(A) Democrats are trying to poll their way to victory.

But this doesn't make any sense, and it isn't what he really wants to say. Rather, he wants to say:

(B) Democrats are trying to win by lying about/distorting polls.

Which is different than (A). One might "poll one's way to victory" by, e.g., doing a lot of polling, and using it to direct one's political efforts. But that's not what Stanley means. Now, (B) is, of course, false. "Democrats" are not trying to do that. Even Mr. Stanley believes that one Democrat--namely Nate Silver--is trying to do that. But I can assure Mr. Stanley that the rest of us aren't. Rather, we're hanging breathlessly on the analyses of Mr. Silver--as well as Sam Want at Princeton Election Consortium, Nate Cohn at Electionate, and Drew Linzer at Votamatic. (I used to check Real Clear Politics, but I don't anymore. I have to hear enough from the Right-Wing Noise Machine as it is, and if I check RCP, I'll just have to see some of the rabidly anti-Obama titles listed there, and then I'll want to read them, and I'll get annoyed, and waste time stewing about them, I don't. But I wish I could.) Nate silver trying to "create a sense of inevitability, to convince the public to vote for Obama because he's a winner"? So far we only have Mr. Stanley's assertion-masquerading-as-a-question to that effect. It doesn't seem too likely. First, Silver's reputation and livelihood are on the line. Second, unlike conservative dittoheads, liberals generally don't much like being fed their own talking points...especially if they are false. Conservatives and liberals really do differ significantly in this respect. Silver has little to gain and much to lose if he skews things in a direction liberals would like. If he's way wrong, I certainly won't be reading him anymore, and neither will most Democrats. He'd likely lose his NYT gig. Stanley's contention is extremely implausible.

Step forward Nate Silver of the New York Times. Nate has been an open supporter of the President and his newspaper just endorsed Obama (although it also went for Dukakis, so it ain’t that good at picking winners). But context doesn’t matter because maths is maths and maths can’t lie – and Nate says that, according to his model, Obama has a 74.6 per cent chance of winning. You might find that figure a little odd given that on the same page you’ll see that Obama is ahead by less than 3 per cent nationally and his advantage lies in one state, Ohio. It’s even odder when you consider how it conflicts with other polls that emerged this weekend giving a virtual tie in Wisconsin and Minnesota. It’s damn near-surreal when you discover that Gallup puts Romney ahead by four points among (and this distinction is critical) likely voters.

Translation: if you just look at the numbers and (a) don't know anything about statistics and (b) don't do any calculations, you will draw a different conclusion from them than will someone who (a') does know something about statistics and (b') does do the calculations. About this I must agree with Mr. Stanley.

He continues:
Meanwhile, Obama’s job approval rating is heading downwards. Does Nate know something that the rest of the world doesn’t?
Well...he knows something that you don't know, Mr. Stanley...that's for sure...

A former business and baseball analyst, Nate came to fame in 2008 when he correctly predicted the outcome in 49 of 50 states in the presidential election. Frankly, a headless chicken could have done that. It was a wave election and we all knew Virginia and North Carolina were in play. Plus Nate had access to internal Obama polls that gave him an advantage over his market rivals. Nevertheless, this success turned Nate into a star – despite his own admission that his analysis technique is not modelled exclusively for politics. In an interview about his life’s work, he wrote:
But the other thing too is on the blog I mostly focus on politics. And I think elections are a really interesting thing to study and to try to predict. But I don't particularly like politics. I find some of the people involved in politics, I don't think they're the most well-rounded or pleasant people necessarily, right? So I want to broaden my focus a little bit and say, look, by being data-driven and looking at how predictions go, doing analysis from statistics and everything else, we can look at business or sports or a lot different fields or science. And there's nothing about politics in particular that my interest and skill sets are uniquely suited to. [Italics are mine.]
That noise you hear is the sound of the cat being let out of the bag. Appreciating that Nate’s system is rather more generic, interpretive and partisan makes sense of its central paradox: that while the rest of us are talking about Romney’s post-debate momentum, Silver still gives the race to the President by a huge margin. Here are some of the problems with his stats.

Sigh. Wow. This is getting rather embarrassing for this Stanley fellow.

What Silver says, roughly:

     (C) My skills, methods and model are really pretty generic, not specific to politics

What Mr. Stanley says that Mr. Silver said:

     (D) My skills, methods and model are generic, interpretive and partisan

Note that (C) and (D) are inequivalent to say the least.

And Mr. Stanley adds, roughly: now we see why Mr. Silver believes different things than do people who don't understand math. Though I'm starting to think that Mr. Stanley might have trouble not only with math but with reading comprehension as well...

 1. Nate isn’t very good at calling close elections. In 2010, he correctly predicted the outcome of the senate elections with the greatest leads. But in the 5 genuinely close races, he got it wrong in 3. For the House elections, Nate ran this extraordinary headline: “House Forecast: G.O.P. Plus 54-55 Seats; Significantly Larger or Smaller Gains Possible.” So, this oracle predicted that the results could have been “larger” or “smaller” – how prescient. In fact, they were much larger. The Republicans took 63 seats.
Whoa! An actual argument that isn't embarrassingly terrible! Now, it's not clear that this argument is much good...but it isn't puerile, which makes it a giant leap forward. Someone more familiar with Silver's work would need to respond here--perhaps Silver himself. There are a  few things we can note, however. For one, we can't tell from this information how much confidence Silver had in the three races he allegedly failed to call correctly. For another, note that Stanley tries to turn a point in Silver's favor against him. If Silver did, indeed, note that larger gains were possible in 2010, then this makes his error less damning. He in effect included an informal estimate of probable error in the summary of his position--he pointed out that he couldn't/shouldn't be that confident in his prediction. That's how you deal with such things scientifically. Stanley pretends that this counts against Silver when, in fact, it counts in his favor.

2. People make their minds up at the last minute, which confuses the outcome of close elections. Historically, voters have tended to break towards challengers, and particularly Republican challengers, in the last week.
Now that's weak. First, some people make up their minds at the last minute...and some don't. Silver, like pollsters, is, as I understand it, primarily interested in the former. That is, he's interested in the question: how much support does each candidate currently have, and how much support is he likely to have on 11/6? As it turns out, there apparently aren't that many undecideds this year. But do undecideds tend to break for Republicans and challengers? Some say yes, some say no. Stanley, firing from the hip--the right hip apparently--only gestures at half of the story. This is the M.O. of the contemporary right, of course--they get their conclusion first, and then assemble whatever evidence they can to convince themselves that they are right, ignoring anything that doesn't support the proposition they prefer to believe.


3. Nate weights polls, meaning that he picks and chooses which data sets to run through his model. He has shown particular affection for Democrat-leaning pollsters like PPP, and this bias is evident in his use of state-wide polls. Silver embraces polling organisations that other writers avoid like the plague. Apparently, the New York Times isn’t as discriminating.
A cheap, abusive and unsupported ad hominem there at the indication that I'm spending way too much time taking an unserious person seriously...but Silver does weigh polls differently...but, then, polls should be weighed differently, as some are better than others. He does rely on state polls for predictions about the states...  Beyond that, we'd need to hear from Silver himself or someone more knowledgeable about his methods. Stanley launches another sloppy criticism here, so it is of limited weight...but it's not weightless as it stands. Bravo, Mr. Stanley!

4. Nate ignores polls that contradict him. So PPP is right and Gallup is wrong.
 Uh...Mr. Stanley should read his own links. Silver has a method for dealing with outlying results. When Gallup changed its methods and produced a radical outlier, he applied that method to the poll. His method might be wrong...but Stanley's own link does not sustain his accusation. Had Silver had to deal with Gallup in an ad hoc way, he might be open to Stanley's criticism. But that was not the case.


5. Politics is even riskier than baseball and “stuff happens.” As columnist David Brooks put it in a critique of Silver’s polls: “Obama turns in a bad debate performance. Romney makes offensive comments at a fund-raiser. These unquantifiable events change the trajectories of tight campaigns. You can’t tell what’s about to happen. You certainly can’t tell how 100 million people are going to process what’s about to happen. You can’t calculate odds that capture unknown reactions to unknown events.” Nor can we determine turnout – and a lot of the polling in 2012 has presumed that as many Democrats will vote today as they did back in 2008. If that’s wrong, many predictions will be confounded.
Gosh, this is getting tedious. Stanley is just wrong here, and wrong in a way that seems to betray a complete misunderstanding of his topic. Any prediction of the kind Silver is making basically has the form: if things continue on their current trajectory, then here is what is likely to happen... Now, of course, things might not continue on their current trajectory. Sandy might change the entire complexion of the race. The final jobs numbers might be extremely bad. An asteroid might hit the Earth. But none of that is relevant--Silver is telling us what we should expect in the absence of such an event. Mr. Stanley asserts that you "can't tell how 100 million people are going to process what's about to happen," but, of course, you often can, and to deny that is just sophomoric skepticism. We know, for example, that a good jobs number is likely to help President Obama, and a bad one is likely to help Governor Romney. We know that another "47%" tape is likely to hurt Governor Romney. And so on. Some uncertainty is not equivalent to total uncertainty. We don't know exactly how people will react to every contingency--but, then, no one anywhere has ever claimed otherwise.

Brooks’ point is really the most powerful of all. Politics is not a science and it doesn’t lend itself well to predictions. Voters lie, natural disasters happen, scandals rock the White House. No one could have predicted – including Silver – that the debates would radically transform Romney’s image in the eyes of the voter. In some cases, that transformation happened in spite of Romney losing one or two of those debates.
Again, this is just embarrassing. Consider the ambiguous:

(E) Politics is not a science

Well...doing politics--e.g. running for office, persuading voters, etc.--may not be much of a science. But predicting elections---that is to say, political science--certainly is. It may, in fact, be rather less complicated than predicting the weather--sometimes known as meteorology... And, interesting fact: political scientists and psychologists actually know quite a bit about when and in what ways people lie, for example. And, furthermore:

We can't predict everything

is not equivalent to:

We can't predict anything.

Silver’s stardom tells us two things about the Democratic Party. The first is that its obsession with numbers is part of a cold, mechanical way of looking at politics that divides the electorate up into blocks of voters who can be cobbled together into a winning coalition. Team Obama went out of its way in the 2000s to recruit professors and statisticians who would turn politics into something like baseball: hire the best players, master all the tricks, bet and gamble your way to victory. Grand narrative is gone. In its place are talking points designed to achieve a 51/49 per cent advantage: war on women, 99 per cent etc
This tells us that Mr. Stanley is either not very bright, or he is blinded by partisanship. Silver's fame lends zero weight to Stanley's assertion here. How one should try to predict electoral outcomes has nothing to do with how the political players view their path to victory. Mr. Stanley's puling is the wail of the innumerate when faced with all that frustrating mathy, sciency reasoning. Although Stanley here slips into postmodern-speak--something I myself eschew--with his talk of "grand narrative"...well, alright I'll go with it. But: what recent American politician has had a grander "narrative" than Obama? Maybe Reagan? By this point, Stanley's piece has become little more than an aimless tantrum, entirely unconnected with the facts.

Second, Nate’s success shows that Democrats are panicking. Losing the war of ideas, they’re resorting to bad maths. Last night a friend posted this on Facebook: “I want Nate Silver's data made into a blanket I can cuddle up with.” Sorry, but weighted polling served up by a partisan analyst is a very false comfort.
 And this is where we laugh out loud...  The right wing--Mr. Stanley, apparently, included--has constructed its own fantasy world and convinced itself that the fantasy is real. Conservatives, as is well known, deny the results of evolutionary biology, climate science, and contemporary economics when they fail to cohere with conservative preferences. Mr. Stanley's largely aimless anti-Silver flailing above, his desperate lunges for criticisms, his panicky assertion that it is his opponents who are panicking...none of these things should come as any surprise to those who have watched the degeneration and derangement of the right in the last twenty years or so. When conservatives don't like where their candidate stands, they leap to their favorite excuse--liberal bias, in this case, polling bias. When Mr. Obama was far ahead of Mr. Romney, conservative kvetching about polls began. When Mr. Romney made huge gains around the first debate, however, conservatives happily accepted these results...only to again reject careful polling analysis when Mr. Romney did not catch up quite enough. Having themselves lost the war of ideas--or, at least, of good ones--and in danger of losing this election, not to mention the demographic battle that will determine the shape of American politics in the longer term, they have retreated into a world in which they have not only their own media telling them what they want to hear, but even their own politically correct version of Wikipedia, scrubbed free of all those unpleasant, cognitive-dissonance-causing facts.

Liberals have Nate Silver, Nate Cohn, Sam Wang, and Drew Linzer. Conservatives have Dean Chambers who takes polling data and adjusts it on the basis of his gut feeling about what it should look like. We have, roughly, science on the left and they have the equivalent of witchcraft on the right. Science is, of course, a dicey matter, and the witches accidentally hit on the truth a non-zero percentage of the time...but that doesn't mean that it's smart to bet on the witches.

* I reckon Mr. Stanley isn't responsible for that heinous title, so I won't complain about it too much. But honestly, if I hear one more idiot suggest that someone, somewhere, thinks that pollsters believe themselves to be determining the outcome of elections rather than predicting them, I'm going to...get quite annoyed...


Anonymous Anonymous said...

Tim Stanley: the NeoConservative toady who shamelessly declared that "Michele Bachmann and Rick Perry are not anti-intellectual."

Hee hee hee.

11:27 AM  

Post a Comment

Subscribe to Post Comments [Atom]

<< Home