How Statistician Nate Silver Threw A Wrench Into Traditional Election Metrics | Hacker News Reader

How Statistician Nate Silver Threw A Wrench Into Traditional Election Metrics

readwrite.com

8 pointstomio14 comments

Who is doing this guys PR? clearly a series of nearly "ghostwritten" "stories"... this is about the 10th "version" with the same headline. Ranging from US tech blogs to UK newspapers... The guy is writing a book and a democratic partisan -- says even...the New York Times.

For reference:

_____________

Why political journalists can't stand Nate Silver |

http://news.ycombinator.com/item?id=4736736

Nate Silver Took A Huge Shot Against Mainstream Political Pundits In His Latest Election Forecast

http://www.businessinsider.com/nate-silver-on-who-political-...

Pundits versus probabilities: The misguided backlash against Nate Silver

http://www.cjr.org/swing_states_project/pundits_versus_proba...

how-statistician-nate-silver-has-thrown-a-wrench-into-tradtional-election-metrics

http://readwrite.com/2012/11/05/how-statistician-nate-silver...

People Who Can't Do Math Are So Mad At Nate Silver

http://www.theatlanticwire.com/politics/2012/10/people-who-c...

________________

but then something different?

NY Times public editor: Silver has Dubious Judgement

http://publiceditor.blogs.nytimes.com/2012/11/01/under-attac...

He has been out there promoting his book...which has increased his visibility further....In short, he’s everywhere...

UK Telegraph: Nate Silver is partisan and wrong.

http://blogs.telegraph.co.uk/news/timstanley/100186850/romne...

We’ve witnessed the evolution of polling from an objective gauge of the public mood to a propaganda tool: partisan and inaccurate.

It's hard to create a dramatic image of a close race when someone crunches numbers, says "nah," and convinces a lot of people. Attempts to discredit him were were bound to happen as the election got closer, along with inevitable defenses.

I think you're so eager to unearth a PR conspiracy that you ignore the more plausible explanation: lazy media.

edit: Check 001sky's submission history.

No, this is not an exercise in statistics (which I understand, trust me). The NY times public editor article clearly articulates he's promoting his book.

Articles don't get written by people who wake up and all write the same headline, even lazy journalists do better than that, unless they are being fed information from PRs.

So what? HN is full of obvious PR. PG even has an essay on it (http://www.paulgraham.com/submarine.html). That doesn't discredit him. You say you know statistics. Take a crack at his methodology if you think it's flawed: http://fivethirtyeight.blogs.nytimes.com/methodology/

I'm sure it would make a great blog post.

I don't need to argue with him. Its the amusing narrative that using statistics somehow original or new that's interesting. I've even said here on HN some of the data visualizations they do are great.

But the storylines or headlines are clearly cynical, PR driven linkbait. Just look at them. They all say the same thing, just slightly disquised like an undergrad trying to hide his plagiarism.

Its just self-awareness that is missing. That's why i posted links: so people can see the patterns.

No PR person drove me to write this article. To tell you the truth, it was respect for Nate Silver. I have been following his work since he started PECOTA at Baseball Prospectus. I have barely read any other articles other than a few passing headlines about the hubbub with Scarborough but I figured it would be a decent time to touch on the subject, with his model being critiqued seemingly from all sides. For his baseball work, I am fond of the approach that Silver takes, and his record at politics using similar methods has proven efficient, worth comment. No PR person reached out to me, I was not influenced by other articles outside of what I could see myself on FiveThirtyEight and I have no political agenda. I can speak for no other author, but I wrote this because it was topical and timely and it struck me as an interesting subject.

I can accept that, and yet agree with the earlier comment that Silver must have some sort of PR thing going on. I first learned of Silver in early 2008 when he was blogging at DailyKos.com under a pseudonym. I was curious to know who he was, and spent some time to find out that it was Nate Silver - friends who knew on the inside told me. So I went to Google his name, and found his amazing bio at Wikipedia. If you know anything about Wikipedia, you know you're asked to NOT create your own page, but create pages of others. It's not intended to be a site for self-promotion. But someone named "Mack2" had created Silver's page (I just went back to confirm it) and here's what I found curious.

One: the page had an inordinate amount of detail, including Silver's high school awards and favorite foods. His page read more like the resume of a recent college graduate - which he was - than a typical Wikipedia article.

Two: the information was incredibly detailed and included a comprehensive set of personal data, the sort of over-the-top detail you would expect from a statistician. And very promotional in tone - go find it and look for the "view history" link, see for yourself, it's still there.

Three: at the time, the only other pages edited by "Mack2" were pages about baseball statistics, Soviet history (Silver is reportedly Russian), and Silver's company and co-workers. Nothing else. Does that sounds like a disinterested journalist documenting an interesting topic? Hardly. I suspect it was Nate Silver himself, writing his own PR, using a fake name - just like he eventually acknowledged doing at DailyKos.com.

I had the sense that Silver was promoting himself there, and I thought he did a fantastic job at it. I'm guessing that once the NY Times picked him up, he may have hired a professional PR agent at that time, or the NY Times got him one - it's a win-win for them, after all. PR agents and journalists go hand-in-hand, any there are individuals who do both, or change jobs from PR to journalism and back - all the time. Not unusual at all. PR is a big aspect of newspaper work and journalism.

It seems like you're trying to discredit Nate Silver by how bad his PR people are. I assume everything is PR so I can focus on merits. Are you interested in merits or taking shots at something you've (apparently) given no thought to?

It's not like 2012 is the first time he's used this methodology. I've been following 538 since it launched in 2008 and generally find it to be insightful and informative. And more right than wrong. And when he's wrong, he usually has a good post to talk about how he got it wrong, and how the miss will factor into future projections.

He wasn't trying to discredit anyone. The point is that there's something of a media frenzy about Silver right now, but I think he's saying there's something that doesn't feel right about it, and I agree.

Silver does good work, but let's be real - all he does is aggregate existing poll data, and give more weight to pollsters who call cell phones as well as land lines. It's the ground-level pollsters who are geniuses, Silver is just aggregating their work. Smart, but hardly rocket science.

If there's any real genius involved, I'd say it's in the PR campaign that's promoting Silver right now. It's clearly effective and well done.

Whatever. Ad hominem attacks are actually useful when people try to dress up their politics as "science" or "objectivity". Try reading Nietzche. Nothing is new here. The purpose of his analytics are propoganda, plain and simple. The notion that an election will be an "independent test" of a pollster's predictions the goal they are all <trying to be avoid>. That is why campaigns employ them, to test their messages, react, change message (as appropriate) and influence the public mood, etc.

What you're saying is you don't care. You can just say that. "I don't care to look further to see if this person's work has merit." That's ok. But don't try to dress it up as some kind of cynical enlightenment.

No, i don't care to argue with you.

This string of posts says otherwise. I think it annoys you that I won't go along with your desired narrative. Your submission history combined with the rapid rise in karma makes me think you're the one doing thinly-veiled PR.

edit: One of your first submissions is from Andrew Breitbart's site. And you're calling others partisan. Who are you working for?

Nah. You're not adding any information or insight with your back and forth. That's what you miss.

You might also be more interested in the PhD thesis I submitted on multi-touch user interfaces, adjacent.

or: http://news.ycombinator.com/item?id=4707742

http://flowingdata.com/2009/02/25/googles-chief-economist-ha...

But nice try at the smear.

Orthogonal Datapoint:

S&P guilty of misleading investors

Landmark ruling by Australian court could pave way for flood of cases against ratings agencies

http://www.guardian.co.uk/business/2012/nov/05/standard-poor...

http://www.cnn.com/2012/11/05/business/sp-guilt-misleading-i...

Dressing up ideas with statistics is not new, original, or scientific, etc.

Its just an informed, iterative marketing strategy.

In business or politics.

The level (AAA!) and quality of the "analysis" notwithstanding.[1]

_______

[1] Its worth noting the ~epistemological assymetry of ratings. They are so "objective" and "reliable" that their use is legally mandated, in the US and the EU. But the providers ratings themselves issue them as "non-expert" "public opinions". Thus, they cannot be sued if you rely on them. We'll see how far the Australian precedent goes, internationally. Until then, Caveat Emptor.

Reply on news.ycombinator.com