Review and comment on the launch of Nate Silver’s for ESPN.

Not the most experienced editor but he's had an experience other editors have not had. He's successfully said to the U.S press: "Current practices could be better. These are better. Here, let me show you."

20 Mar 2014 7:23 pm 22 Comments

The protoblogger and software maker Dave Winer, a friend of mine, recently wrote, “Blogging needs your help. There’s cobwebs in the blogosphere.” His cri de coeur

The mission of blogging is to empower all of us to go directly to each other with our expertise. So if you know something as well as anyone else, or you learn something or know something that should be shared, then you should share it on your blog.

This is my niche blog, PressThink: Current events in the way American journalism explains itself to itself. Nate Silver’s new site for ESPN, the re-born FiveThirtyEight, launched this week. That was a current event for my niche.

First the review:

Economics blogger Tyler Cowen (“so far I don’t think this is working”) was not impressed. Paul Krugman (“hoping that Nate Silver and company up their game, soon”) was underwhelmed. Kevin Drum (“its first day didn’t do much for me”) was skeptical that demand for Silver’s goods would ever prove out. Via Mathew Ingram’s assessment comes this tweet, summing up:

Others were more than “mildly disappointed.”

In The Guardian, my colleague Emily Bell argued that sites like fivethirtyeight, Ezra Klein’s, and Pierre Omidyar’s First Look Media (alert: I am on their team, see this for disclosure purposes…) these brave new start-ups weren’t going to change much if they couldn’t manage a more diverse pool of initial hires. See: Journalism startups aren’t a revolution if they’re filled with all these white men.

An interviewer for New York magazine asked Nate Silver about this.

…The idea that we’re bro-y people just couldn’t be more off. We’re a bunch of weird nerds. We’re outsiders, basically. And so we have people who are gay, people of different backgrounds. I don’t know. I found the piece reaaaally, really frustrating. And that’s as much as I’ll say.

Picking up from that remark, Zeynep Tufekci gave a brilliant explanation for how the self-conscious “outs” can become the less-conscious “ins,” drawing on her knowledge of nerdom (girl programmer) and sociology (Pierre Bourdieu fan.) Her piece: No, Nate, brogrammers may not be macho, but that’s not all there is to it.

“Building A Diverse Newsroom Is Work,” said Shani O. Hilton, Deputy Editor-in-Chief of BuzzFeed, in another good response to Emily Bell’s provocation.

Any newsroom in which the black staffer is expected to speak up for blackness while the white staffers only have to speak for themselves is a newsroom that’s failing.

Getting to that level of diversity takes work. It’s something BuzzFeed is OK at — and we’re working on improving. The undercurrent to much of the criticism of Silver and Klein et al. is an assumption that it’s easy to hire a diverse staff if you try, but white dudes just aren’t trying. I’m not a white dude, so I can only speak to the first part of that sentence, as someone who’s done a fair amount of hiring in my year at BuzzFeed. So here goes: It actually isn’t easy to build a diverse newsroom.

One reason this matters even more for FiveThirtyEight: if the newsroom’s “cuts” on the data aren’t diverse enough, that’s a potential source of error… right?

Speaking of error:

But two days was long enough for Leon Weiseltier, who runs the back of the book at the New Republic and always has, to figure out what FiveThirtyEight has wrong: The Emptiness of Data Journalism: Nate Silver could learn a lot from those op-ed columnists he maligns.

The new technology, which produces numbers the way plants produce oxygen, has inspired a new positivism, and he is one of its princes. He dignifies only facts. He honors only investigative journalism, explanatory journalism, and data journalism. He does not take a side, except the side of no side. He does not recognize the calling of, or grasp the need for, public reason; or rather, he cannot conceive of public reason except as an exercise in statistical analysis and data visualization. He is the hedgehog who knows only one big thing. And his thing may not be as big as he thinks it is.

“He does not take a side, except the side of no side.” Ryan L. Cooper in The Week made a similar point: Nate Silver’s FiveThirtyEight, and the dangers of being ideologically neutral. NateSilverpreso2“By claiming the mantle of pure analysis, Silver is falling into a familiar journalistic trap…” Cooper brings in my own writing to make this point: “Everyone, without exception, has some kind of ideological-theoretical perspective that informs the way they interpret the evidence they see.”

Silver has ripped into lazy columnists for consulting their “ideological priors” first, but he’s not saying he has no priors… is he? Cooper’s warning: failure to answer that leads directly to Slate-style contrarianism.

And that would be disappointing! My own view of FiveThirtyEight’s launch is different from these, but that’s because my “launch object” is different. I take it for granted that the product will be weak at first. What interests me — initially — is the argument for practical improvement on which the new service is launched. You start with some (poorly informed) hunches about what would be worth a try, based on some (firmly held) convictions about what’s presently done badly or not at all in the practices you are trying to make better.

When Bloomberg View (the opinion section of Bloomberg News) launched in 2011, it didn’t launch behind an argument identifying defects in the opinion publishing system that Bloomberg View was just itching to address. It wasn’t like that. Rather, the editors thought they could have a strong entrant in an existing op-ed and editorial page competition. “Get the best people, pay them good money…” is all the strategy you need. Bloomberg had a successful brand, it had the money, and a boss, Mike Bloomberg, who wanted the influence. Great editors sign up the brilliant writers they know. Voilà! Bloomberg View is born. One of the “name” writers they signed, Ezra Klein, had already proven himself at the Washington Post.

So that’s one way of introducing a product: make it like the other products, but try to find better people.

Nate Silver and ESPN aren’t doing that. Neither is Ezra Klein with Vox, or Pierre Omidyar with First Look, or Jessica Lessin with The Information. These ventures may go nowhere. Their work will have to compete with what’s online already, a huge flow of blogging and journalism about politics, technology, sport, economy, world affairs. It is harder — way harder — to come out with something different than it is to describe the different thing you envision making. To illuminate a different starting point is easy. To practice differently by starting from that point: difficult.

As your guide to its pressthink I’m trying to understand the nature of the bet FiveThirtyEight is making. Nate Silver wrote a manifesto, What the Fox Knows, about that. It spells out his pressthink at launch. This is blogging: “If you learn something or know something that should be shared, then you should share it on your blog.” (Dave Winer.)

Why Do We Expect So Much From Nate Silver? by Benjamin Wallace-Wells of New York Magazine— this should be shared, especially for its conclusion:

The hope invested in these projects is that as the industry shrank, perhaps, at the very least, what was left might become smarter. The profession has retreated, but maybe it has retreated to higher ground. Which explains, I think, some of the big talk that has accompanied their introduction. And it may explain too the very slight sense of letdown that accompanied the launch of FiveThirtyEight this week.

For a lot of people in journalism, “big talk” and launch manifestos are on their face ridiculous. What matters is the work you can publish and defend. No one needs to know about your grand intentions. Let the journalism carry the message. It will anyway! That view has a healthy constituency in newsland and after my first five months as a part-time adviser to Omidyar’s First Look, the wisdom in it is… a lot clearer. But Ben Wallace is right, too. There are thousands and thousands of journalists (and young people hungry to be journalists) who know that the news industry is going to be smaller in the years ahead, but it can be way, way smarter. And make better products.

For those people, I think it’s fair to say, the rise of Nate Silver’s FiveThirtyEight has been an important marker of… we can be way smarter. He’s not the most experienced editor, but he’s had an experience that almost no other editor has had. He has successfully said to a particular cohort in the U.S. press: “Your current practices could be better. These methods are better. I am going to show you that they work.”

The part of journalism in which he made this intervention — predicting what’s likely to happen in presidential elections — is an easier problem than you might think, a point that Silver frequently has to make in his public appearances. In What the Fox Knows he isolates three “your practices are weak, I have a better way” opportunities:

* what counts as explanation in advanced news writing today;
* the making of generalizations by journalists steeped in the particulars;
* the art of prediction, the science of likelihood, and products in between.

In these areas he thinks he can make improvements by bringing the discipline of data science to the topsy turvy of news. Whether that will make a big enough difference is the unknown. That’s why we run these experiments.

Finally: My colleague at NYU, Mitchell Stephens, author of a forthcoming book, Beyond News: The Future of Journalism, thinks “explanation” may be a mistaken path to user satisfaction because it takes the fun out of following the news.

[Margaret] Sullivan sounds a bit condescending when she maintains that “the reasonable reader” requires a prominent news summary because that reader can’t be expected to have been keeping up with the news. Can’t that “reasonable reader” just check Google News, or one of a hundred other sites, if momentarily perplexed? There also strikes me as something potentially condescending in Klein’s infatuation with explanation. If readers remain a bit weak on what, say, the debt ceiling is, can’t they just click around the Web a bit and find out? (Siri, on my iPhone, comes up with a pretty good answer.) Do journalists have to spoon feed? Wasn’t an insistence upon aiming at the least common denominator—with a simplified vocabulary and three paragraphs of background in each story—part of what was wrong with traditional journalism? Aren’t the best television shows, the best films, now doing less not more explaining – in part because figuring stuff out on your own is fun?

Stephens is onto something. Past a certain point “explanation” can become a substitute for participation, a barrier to engagement. Another way to say it: The quest to understand is even more involving than a need to know. And mystery is the most engaging metric of all. Explanation that invites participation: what are the odds that with the new FiveThirtyEight we could wind up with something like that?


So if journalists stop explaining, when people do their searches on the Internet for an explanation, what will their searches turn up?

I don’t exactly know how to ‘search the Internet for an explanation.’ For example, I tried “Why is there income inequality?’ Top rank is from the NBER, #2 is from the Institute for Faith, Work and Economics and #3 is Wikipedia’s generic treatise on economic inequality.

Moral – I think we’re always going to need explanatory journalists, researchers and reference librarians. 😉

“the art of prediction, the science of likelihood, and products in between.”

That’s the piece where Nate Silver (and some others, e.g., Brendan Nyhan) shine brightest and IMHO over, say, a ‘pure explainer’ like Ezra Klein.

I’m not excited to judge a new product so soon. But one angle that needs to be explored is a bit simpler and it’s one you’ve dealt with before: Can talented writers scale themselves?

I reread this thinking about Silver’s struggle to scale himself. He’s best known for one thing, despite his insistence about other subjects he covered, his work on the elections is what gave him this opportunity.

I would argue he has to find similar quests for his writers. They have to be bold, big questions to which his algorithm can add tremendous value. And my initial take on it is that he has not found these big questions. Except for with sports. He clearly plans to own the March Madness brackets, for instance.

Finally, on your point about explanation. I think you lead to a point I’ve been thinking about for a while. Explanations aren’t valuable just to bring a reader up to speed. Explanations are valuable, if they’re done well, because they’re entertaining.

People like to learn. If you want proof, you need only look at cable television. Shows like “Mythbusters” are all over. They provide accessible paths to learning.

I’d suggest a less condescending way to pitch explanatory journalism is to say you simply want to entertain people really well.

Scalability is perhaps part of the problem. Or perhaps it’s part of a defense that Silver doesn’t deserve. A number of people beyond Krugman have critiqued him for a number of reasons, starting with but by no means limited to, the Roger Pielke article which Rosen himself doesn’t even mention.

To expand a bit on my quote in Jay’s piece:

I am a big fan of Nate Silver, Ezra Klein and Matt Yglesias. But I think they have to focus on the INSIGHT business not the more condescending EXPLANATION business.

And we have to worry whether the people they hire — now that they are publishers as well as bloggers — can supply the same level of insight we had come to expect from them. Will their sites be as interesting as their blogs were?

As for the “expectations” game, isn’t this part of the dark side of “branding,” whether it’s Nate, or Ezra, or David Pogue, or whomever? Branded Media Guru X is selling us on these expectations, hinting, if not promising, that he will deliver.

As for other issues? Nate had several months since the end of his services with the Times to have at least some kinks ironed out pre-launch.

You say:

“There are thousands and thousands of journalists (and young people hungry to be journalists) who know that the news industry is going to be smaller in the years ahead, but it can be way, way smarter. And make better products.”

There are news readers who know that too. One thing that I find deeply frustrating is the way that news sites pretend that they don’t have vested interests. I’d love to see that convention blown out of the water by a site that makes its interests explicit.

For example, I think it’s nearly indisputable that a wider rotating stable of op-ed columnists would produce more diverse and better coverage — but it’s in news organizations’ best interest to build “brands” of trusted writers (often, so they can syndicate them). So instead we get the equivalent of tenure on op-ed pages.

Or elections. News organizations have a HUGE vested interest in pretending that horse-race coverage is what people want, but I have seen very little data that suggests they actually know this. Rather, I think it is what news organizations *like to, want to, and feel comfortable producing,* and so we get more and more of it.

Or scoops. I could not possibly give any less of a hoot whether outlet X or Y breaks a story. I just do not care. I want accurate news, searchable news, linked-out news. Did the Wall Street Journal publish some actor’s death faster than the New York Times? WHO CARES? What is that a proxy for? Nothing useful to me, that’s for sure. (Doing a major investigation, sure, that’s cool — but it’s the investigation I value, not the scoop worthiness.)

My hope — as a news reader — for sites like Silver’s is that they will scare the pants off traditional media, in a good way. Force them to spell out why they make the priority choices they do, instead of acting like they have the right to make black-box decisions and never deign to let their customers know why.

“One thing that I find deeply frustrating is the way that news sites pretend that they don’t have vested interests. I’d love to see that convention blown out of the water by a site that makes its interests explicit.”

Forbes seems to have that down pat. 😉

Two things:

1) I don’t plan to visit Nate’s website any more than I ever visit the New York Times. Meaning, the stories that matter to me find me in social networks and through email newsletters. So whether every story on his site is inspiring and amazing new journalism isn’t the point. Which leads to,

2) The 538 stories that ARE good and DO find me so far have been pretty good. I wanted to know about the work of Bayesian statisticians in relation to the missing jetliner. Thanks, Carl Bialik:

Getting a journalism site up and running reminds me of when I took up golfing. At first, one good shot occasionally was enough to keep me coming back. Then I tried to have one good shot per hole. Finally, I’d strive to have only one bad shot per hole. That’s bogey golf, the best an amateur can hope for. I’d say Nate is getting pretty close to bogey journalism already.

In terms of the “it’s not easy to create a diverse newsroom,” I’m not sure I agree. Jeanne Brooks (now of WNYC) and Chrys Wu (now of the NYT) and I selected all the speakers for the Online News Association conference in 2013. It’s precisely the kind of news-nerd talent pool I’d shop in if I were hiring for such an organization.

Without fanfare we produced a conference where there were 50% women speakers, and 25% people of color.

The thing is? It wasn’t that hard.

I have two theories about why things are ending up so XY and monochromatic:

1. They’re not asking. Had we stuck to the sessions people had submitted the composition of the conference would have been almost universally male and white.
2. They’re asking, but their network is composed of people exactly like them.

Both of these are soluble problems, and they are trivial in comparison to the problems of creating a successful startup.

Tom Elliot says:

I can understand not explaining some things, flight data recorders for instance which numerous anchor-faces seem to think we need to have explained over and over again. But something like “debt ceiling” which is a phrase key to a particular kind of designed misunderstanding inherent in its usage both politically and ideologically, really does need explaining. Not only explaining but contextualizing. When not only the term but how it is used is key to the news of the day then explanation, which on that particular subject has been woefully incomplete, is not only necessary but imperative.

I find it curious that someone who made their name being able to more accurately analyze the data than just about anyone else would hire Roger Pielke Jr. who has developed a reputation of being more likely wrong that Bill Kristol prognosticating politics. Pielke is the most debunked climate commentator who still gets published by anyone except themselves.

Watts Up With That is self published.

Kevin P. says:

Minor correction: It’s Tyler Cowen, not Cowan.

Thanks, fixed.

Silver says they’re a bunch of nerds. Well, a subset of “nerd” — they’re statistics nerds. It seems to me that they don’t really see the forest because they’re too busy counting the trees, and they don’t really see the trees for what they actually are, because it’s the numbers that fascinate them, not the trees themselves.

Which is why the site is disappointing. It seems curiously contentless. There’s no there there.

Love and hate of Shakespearean proportion via the Unfair Vanity among amateurs and so called experts 😉
Never a dull moment in the land where toxic news is good for one and all readers

I can understand not explaining some things, flight data recorders for instance which numerous anchor-faces seem to think we need to have explained over and over again. But something like “debt ceiling” which is a phrase key to a particular kind of designed misunderstanding inherent in its usage both politically and ideologically, really does need explaining
phần mềm ghép ảnh cho android
phần mềm photoshop cho android

How, algorithmically, could one (say, go about ascertaining which areas/fields/subfields had “experts” that really were, i.e. were more likely to be right than the parachuters-in, versus which areas offered 538 a chance to provide real, new value? (Though perhaps Silver could create value in all fields, if he could mathematically figure out and apply this “distinguishing [fields with] real vs pseudo experts” approach. But I’m drawing a blank on how one could do it, so maybe it’s just a pie in the sky idea.)

Andy Donohue says:

It is way too early to judge 538. What matters is not what the product looks like now but how it adapts. So that’s what I’m excited to watch.

That kind of adaptation requires some humility. You have to acknowledge that what you thought would work wouldn’t and move on.

I’m incredibly excited about the project. My worry is whether the attitude they’ve given off early on will get in the way of their ability to change.

It starts at the fox logo, which says: We’re smarter than everyone else. Silver’s launch manifesto seemed to me to say “here’s why our approach is the best.”

Compare that to Bill Simmon’s launch manifesto at Grantland, a somewhat similar project.

“I would love to tell you that this website will work, that we’ll entertain you five days a week and blend sports and pop culture successfully. The truth is, I don’t know for sure. … You figure out what works, you figure out what doesn’t work, you keep moving. That’s the next nine months for us. Eventually, we will evolve into what we are. Whatever the hell that is.”

The 538 crew has said they’re listening and evolving since the launch, and that’s great. But Silver’s satirical take on Krugman didn’t seem to take real criticism to heart.

Hi Andy,

I appreciate the thoughtful comment.

We’ve published ~80 pieces of content at the new site so far. It’s entirely fair to criticize each and every article and blog post, and we couldn’t get better if not for that criticism.

I’m a baseball geek, so I’ve started to think about our content in terms of on-base percentage (OBP) and slugging average (SLG).

OBP is how often you’re getting on base by publishing decent content that meets your basic editorial standards and adds a little bit of value to the readers. SLG is how often you’re hitting home runs by doing some really exceptional analysis or reporting that advances the thesis of the site. In the long run, you hope to have both a high OBP and a high SLG, just as the best baseball players do.

In the near term we’re focused more on the OBP side of the equation. There are a few reasons for this. First, it’s easier to kill a mediocre story than to commission a great story. (You can increase your OBP by taking fewer swings.) Second, it’s clear that readers are holding us to a high standard. That’s terrific. But it also means we’ll be duly criticized for swings-and-misses. Third, the home run stories take more time. In our case, they’ll often involve building a statistical model, an interactive data visualization, a series of art elements, or all of the above, in addition to writing and reporting. The lead times on these projects are measured in weeks to months.

To torture the baseball metaphor:

I’d interpret Tyler Cowen’s criticism to mean “these guys don’t have a high SLG so far. They’re not publishing very much best-of-the-web type stuff that’s deeply interesting to me and people like me.” That’s totally fair, and probably largely right. But Tyler has also linked approvingly to a couple of our base hits, such as stories we’ve done on Venezuelan politics and the labor market. We appreciate that.

I’d interpret Jay’s gentle criticism to mean: “you know what, these guys aren’t hitting for all that high an OBP so far. But they’re rookies, so that’s about what you’d expect.” I appreciate Jay’s generosity.

Obviously, I feel a bit differently about one or two of the other critics. It’s one thing to say that our OBP or SLG is lower than it should be. It’s another thing to point toward our 4 or 5 worst at-bats and conclude that our OBP is zero.

I appreciate your readership and hope you’ll continue to have high hopes for the site.


Good to hear that Real Journalism Is Not Dead …
Conventional wisdom is that the Internet killed serious journalism, but Michael Kinsley begs to disagree.

” rich foundations and individuals appear downright eager to jump in and supply foreign or other prestige news if newspapers won’t. Former Times executive editor Bill Keller just quit the paper to help start a nonprofit to cover justice issues. Paul Steiger, formerly managing editor of The Wall Street Journal, founded ProPublica—a nonprofit that produces top-quality investigative journalism.”