home studios equipment staff/friends booking/rates for sale forum contact

Dude: Nate Silver

Vote and debate.

Moderators: kerble, Electrical-Staff

Dude: Nate Silver

Person who renders all pundits, talking heads, and self-proclaimed experts useless
32
71%
NEEEEEEEEEEERDD!!!!
13
29%
 
Total votes : 45

Re: Dude: Nate Silver

Postby Big John on Thu Nov 17, 2016 4:36 pm

I saw a lot of polls that seem to be below the bar of what would have been considered in the past as being a big enough sampling group. These polls seemed to use under sampling and then a math model on top of that to correct the results.

The difficulty is that a "under-sampled" poll a traditional poll would have thousands of people usually between 5 and 10K or more giving you a pretty good sample of each state. All states and regions would be covered and you would not have to model above that if you had the phone numbers, population and voting districts worked out right. This was the standard practice after the Truman election where the straw polls had shown Dewey leading and caused a number of polling operations to go out of business. So there were fewer pools with bigger sampling numbers based more accurate US sampling models.

So a polls a lot of which seemed to be below a thousand in some cases a hundred (for who won the debates the next morning). The problem becomes what corrective math you use on top of the under sampling. It is apparent that the polling operations corrective math skewed the underlying data that was too small of a sampling group to self correct. Thus 40% leads for Hillary two weeks before the election.

Silver does not create his own data but compiles others polls so his special math sauce could not correct for the English put on the under sampled polls by the pollsters corrective math.
User avatar
Big John
Humankind's Greatest Musical Genius
Humankind's Greatest Musical Genius
 
Posts: 4399
Joined: Thu Nov 10, 2005 10:16 am

Re: Dude: Nate Silver

Postby joelb on Mon Nov 21, 2016 9:25 pm

Gramsci wrote:
joelb wrote:
Bottom line 28 percent is a lot. A lot of things had to break Trump's way and they did.


Here's a discussion about Taleb vs Silver:

https://www.quora.com/Whats-Nassim-Tale ... at-he-does

Which gets rather granular here:

https://mishtalk.com/2016/08/07/nassim- ... of-tweets/

At the root Taleb's problem is some making money as a Probability Sage. I listened to all of the 538 podcasts during the cycle. Silver was consistently way off during the Republican primary regarding Trump's chances, as we almost everyone. The actual race between HRC and Trump was almost entirely focused on how much she would win by, even when it was getting 50-50%.

That said Silver and team covered their asses with the occasion mention that Trump could win, but it was highly unlikely. In the end his model was way off compared to his near perfect predictions on 2008. Silver has rarely had the same level of success predicting since that election making it seem that was more a fluke than anything else. Remember that Silver was once lauded as a predictor, not wobbly probably wonk. To his credit since his accuracy decline since 2008 he has rebranded 538 as more about probability than prediction. Maybe it's just the media that haven't caught up.

What good is predictive analytics if they don't actually predict anything?


I forgot about this thread. Thanks for those links, they are really helpful as I did not follow all of Silver's podcasts or the debate among statisticians. As I understand Taleb's point, it is that the outcome of the election is so unpredictable outside of a couple weeks before as to be effectively random. If that's what he's saying, I don't agree. I do agree that there is a chance of an event close to the election that swings turnout or voter choice in a way that can change the result. But I don't know what the odds are of such an event occurring.

Isn't this claim, though, in direct conflict with the black swan concept? If a black swan is by definition not predictable, then a forecast should simply be able to posit that a chance of a random event exists, and that this cannot be incorporated into the probability. If you start saying you expect black swans so the election outcome is not predictable, then the black swan is no longer black.

I am NOT a statistician, obviously.
DrAwkward wrote:If SKID ROW likes them enough to take them on tour, they must have something going on, right?
joelb
tired of birds
tired of birds
 
Posts: 1121
Joined: Mon Dec 19, 2005 12:16 pm
Location: City by the sea

Re: Dude: Nate Silver

Postby matthew on Thu Mar 02, 2017 9:05 am

I think I've got Nate Silver's Trump blunder figured out!

He didn't heed the wisdom of a folksy Catholic bishop who, among other things, alluded to how and why the Literary Digest totally missed the mark on the results of the presidential election of 1936.

This was in the 1950's. How prophetic. Nothing new under the sun, and what a short memory so many people have.
...diesque nostros in tua pace disponas...
User avatar
matthew
Master of Lit., Arts, & Architecture
Master of Lit., Arts, & Architecture
 
Posts: 3503
Joined: Sat Jan 29, 2005 3:38 pm
Location: Gens non sancta.

Re: Dude: Nate Silver

Postby Mason on Fri Mar 03, 2017 2:17 pm

matthew wrote:I think I've got Nate Silver's Trump blunder figured out!

He didn't heed the wisdom of a folksy Catholic bishop—


I can't stop thinking about how good this is. Incredible idea, expressed perfectly. Diction, timing, holy shit.
Marsupialized wrote:'It's very important for men in bands together to French kiss, every day, even if they don't feel like it'
User avatar
Mason
Man with Encyclopedic Knowledge
Man with Encyclopedic Knowledge
 
Posts: 5102
Joined: Sat Apr 26, 2008 10:42 pm

Re: Dude: Nate Silver

Postby matthew on Sat Mar 04, 2017 7:48 pm

A_Man_Who_Tries wrote:Answer Boombats, Matthew.


Mason wrote:...Diction...


I'm quite good with dictioning.

Wait for it. Get it? Got it? Good.

Have a happy Lenten season.
...diesque nostros in tua pace disponas...
User avatar
matthew
Master of Lit., Arts, & Architecture
Master of Lit., Arts, & Architecture
 
Posts: 3503
Joined: Sat Jan 29, 2005 3:38 pm
Location: Gens non sancta.

Re: Dude: Nate Silver

Postby Clyde on Tue Feb 19, 2019 11:29 am

Nate Silver has fully transitioned* from a stats guy to a insufferable centrist pundit guy with Beltway approved "but how will this play in Peoria?" takes. He's barely distinguishable from Chris Cillizza.




*Or was he always this way?
NewDarkAge wrote:Dialectics, mate. No easy answers.
Clyde
Lode Star of the Twenty-First Century
Lode Star of the Twenty-First Century
 
Posts: 2462
Joined: Thu Aug 18, 2011 4:12 pm
Location: Life

Re: Dude: Nate Silver

Postby Andrew. on Tue Feb 19, 2019 11:22 pm

Thought I remembered posting in this poll and randomly clicked page 4 and found it.



I've always thought he was a status-quo centrist creep at heart but some people love them some horse race data analysis. I assume there's crossover fans with traders and numbers guys who loved The West Wing. But these are only biases and impressions of mine. No data to back them up.
User avatar
Andrew.
Power Incarnate with Endless Creativity
Power Incarnate with Endless Creativity
 
Posts: 7101
Joined: Mon May 07, 2007 11:02 pm

Re: Dude: Nate Silver

Postby Clyde on Wed Feb 20, 2019 10:36 am

I'm not a data guy so I can't say whether his methodology is sound. I remember that he got the 2008 and 2012 elections results almost exactly but I also don't know if that was remarkable in itself or if it was in line with other pollsters. Similarly, I don't recall if his predictions were better or worse than other people's in 2016. (Other than the master, Carl Diggler.)
But whatever the issues with his methods are, his whole spiel was that punditry was bad and people should look at the numbers themselves. Which is fine, as far as things go. One could argue that doing things that way stops privileging bullshit narratives and beltway consensus. But now 10 years later compare this

In 2020, Sanders won't benefit from those low expectations. He may not be the frontrunner in the race, but he absolutely -- if polling is to be believed -- starts in the top tier. Sanders has never been a frontrunner sort of guy, so this will all be new to him. He will no longer get credit for leaping over low expectations. In fact, he is much more likely to struggle with high expectations. What happens, for instance, if polling this fall shows Sanders in sixth place in Iowa? Or third place in New Hampshire? How does he and his campaign react to that?... There is zero question that Sanders has already had a profound impact on the Democratic Party -- even if he's not technically been a Democrat all that time. In 2016, Clinton laughed at Sanders' "Medicare for all" proposal. Now it is the default position of the 2020 frontrunners -- with only Biden resistant to endorsing it.
But Sanders won't have the liberal lane to himself in this race like he did in 2016. In fact, the liberal lane is stuffed full of candidates -- all of whom sound a hell of a lot like Sanders on policy. (This is not an accident.) Can Sanders win on a well-yeah-but-I-was-here-first argument? Or does he need something more, something beyond the ideas that energized his 2016 campaign? In interviews setting up this second candidacy, Sanders sounds exactly like he did in 2016: "Our campaign is about transforming our country and creating a government based on the principles of economic, social, racial and environmental justice," he said on Tuesday. Is that enough for voters with a lot more choices this time around?
]

and this

One year-ish ago, I thought Sanders was the frontrunner in a crowded Democratic field, but since then I've grown more pessimistic about his chances. I don't want to overcorrect too much though and think he still has a viable path. IMO the viable path probably involves the field remaining relatively divided. In that scenario, his high and loyal floor of support, his $$$ (powered by small donors), his experience (may be only 2020 candidate to have run before) with delegate rules etc, could be big advantages. What I've grown pessimistic about is his ability to unite the different factions/coalitions of the Democratic Party. He doesn't seem that interested in doing so, and polling shows he's a tough sell for some Ds. That's how nominations are *usually* won, through coalition-building. In some ways, the question is: if Bernie just did what he does 4 years ago, will it work? The optimist's case would be pretty Occam's Razor-y: He was 2nd place last time—and the 1st-place candidate isn't running. And the party is moving to the left. So…why not? But there are a lot of things that are more difficult than 4 years ago, by far the most obvious of which is that there's *much* more competition, including specifically on the left. At 77, his age is a bigger concern. And there are fewer caucuses than in 2016. Bernie's coalition is likely to be a bit different geographically in 2020, with the #NeverHillary vote divided between several different candidates. In particular, #NeverHillary voters helped him in Appalachia, and in the Western caucus states. Those aren't hugely delegate-rich areas. But it does mean that Bernie will have to hit his marks in his other strong areas, e.g. New England and the Upper Midwest. He and Klobuchar, although very different ideologically, are fighting for some of the same geographic territory.


The first one is CNN's political analyst and grade-A dum-dum Chris Cillizza and the second one is numbers guy Nate Silver. Okay, so Silver's writing is not as boneheaded as Cillizza's and he seems more thoughtful overall but what they are saying is essentially the same; they're the same kind of banal observations that anyone could make and which serve no other purpose than to gab on about the "horse race."
NewDarkAge wrote:Dialectics, mate. No easy answers.
Clyde
Lode Star of the Twenty-First Century
Lode Star of the Twenty-First Century
 
Posts: 2462
Joined: Thu Aug 18, 2011 4:12 pm
Location: Life

Re: Dude: Nate Silver

Postby Wood Goblin on Wed Feb 20, 2019 11:01 am

Punditry has never been Silver’s strong suit, and his Sanders piece is pretty banal—stuff to fill column space during a period of little statistics activity. He has a point about caucuses, but that’s about the only item of value in that excerpt.

By and large, people who ignore him fall victim to their own selectivity. As in, I’m not particularly interested in the opinions of people who (a) think Silver is useless but (b) tout recent polls as evidence of their preferred candidate’s popularity.

(I’m not accusing Andrew of this. I don’t think he does this.)
User avatar
Wood Goblin
Heaven-Sent Hero
Heaven-Sent Hero
 
Posts: 6108
Joined: Tue Jun 06, 2006 12:43 pm
Location: South Loop, Chicago

Re: Dude: Nate Silver

Postby Clyde on Wed Feb 20, 2019 12:22 pm

Wood Goblin wrote:Punditry has never been Silver’s strong suit, and his Sanders piece is pretty banal—stuff to fill column space during a period of little statistics activity. He has a point about caucuses, but that’s about the only item of value in that excerpt.


He does it more than occasionally, though. Just one example-he wrote a big long thing about Amy Klobuchar about a week ago. (https://fivethirtyeight.com/features/am ... n-kickoff/). The phrases "retail approach to politics" is used and he has a whole section on "beer track vs wine track" that was straight-up embarrassing. He threw some stats in there too but the final analysis was no different than the kind of dull op-ed The Washington Post puts out twice a week. He's either cynically or sincerely using his bona fides to draw conclusions that any hack could make, and neither scenario fills me with confidence about his project.
NewDarkAge wrote:Dialectics, mate. No easy answers.
Clyde
Lode Star of the Twenty-First Century
Lode Star of the Twenty-First Century
 
Posts: 2462
Joined: Thu Aug 18, 2011 4:12 pm
Location: Life

Re: Dude: Nate Silver

Postby Andrew. on Thu Mar 07, 2019 4:02 pm

If normative status-quo horse-race data diddling is your hobby, I guess that's fine. But let's be clear that's this guy's gig. And if he's gotta throw some totally unscientific polls up on Twitter and generate some graphs from them along with some snappy self-aware commentary over 1,500 words, that's just daily content at fivethirtyeight dot com. You gotta have regular content after all. Sling that 'tent, as Scott Auckerman says.

Some dudes play video games and some watch UFC and some keep up with Nate Silver. It's just kind of useless shit that dudes do imho.

https://fivethirtyeight.com/features/is ... nderrated/
User avatar
Andrew.
Power Incarnate with Endless Creativity
Power Incarnate with Endless Creativity
 
Posts: 7101
Joined: Mon May 07, 2007 11:02 pm

Re: Dude: Nate Silver

Postby Andrew. on Wed Aug 14, 2019 10:12 am

Zero percent surprised that this shill uses his platform to hate on Bernie.

https://twitter.com/NateSilver538/statu ... 4000327680

https://twitter.com/sexyfacts4u/status/ ... 3605550081
User avatar
Andrew.
Power Incarnate with Endless Creativity
Power Incarnate with Endless Creativity
 
Posts: 7101
Joined: Mon May 07, 2007 11:02 pm

Re: Dude: Nate Silver

Postby andteater on Wed Aug 14, 2019 10:54 am

The way that Bernie twitter reacts to any commentary on him blows me away. The worst Napoleon Complex defensive weirdos crying about everything. If you're all so secure that Bernie is, in fact, the front runner, the most popular candidate, etc. then why the fuck do you care about what Nate Silver says?

pathetic.
User avatar
andteater
Lode Star of the Twenty-First Century
Lode Star of the Twenty-First Century
 
Posts: 2302
Joined: Fri Jun 06, 2003 1:29 pm

Re: Dude: Nate Silver

Postby Andrew. on Wed Aug 14, 2019 12:50 pm

andteater wrote:The way that Bernie twitter reacts to any commentary on him blows me away. The worst Napoleon Complex defensive weirdos crying about everything. If you're all so secure that Bernie is, in fact, the front runner, the most popular candidate, etc. then why the fuck do you care about what Nate Silver says?

pathetic.


Haha. How's your Wednesday going my dude? Get any funny texts from friends or loved ones? I just got in from a run and texted my girlfriend a pic of three deer (mother and two fawns) I saw down by the river. For part of my run along the Assiniboine I listened to Vish Khanna's previously unreleased 2008 interview with David Berman. It's really great.

This is the Nate Silver thread where people post opinions about Nate Silver. I've made a couple posts here critical of Silver going back to 2016 (w no mention of Bernie). I just posted a link to a Warren supporter saying that even they are annoyed by Silver's hate-on for Bernie.

I hope you're less of an angry, reactive crank in real life than you are on this old internet forum.
User avatar
Andrew.
Power Incarnate with Endless Creativity
Power Incarnate with Endless Creativity
 
Posts: 7101
Joined: Mon May 07, 2007 11:02 pm

Re: Dude: Nate Silver

Postby Andrew. on Wed Aug 14, 2019 1:35 pm

Several people point out that Silver owns himself by referencing "Bernie's Chomskyian attempts at diagnosing the sociology of the media" without understanding that Chomsky's critique of media refers precisely to people like him. Lol.
User avatar
Andrew.
Power Incarnate with Endless Creativity
Power Incarnate with Endless Creativity
 
Posts: 7101
Joined: Mon May 07, 2007 11:02 pm

Re: Dude: Nate Silver

Postby askii on Wed Aug 14, 2019 4:17 pm

Andrew. wrote:Several people point out that Silver owns himself by referencing "Bernie's Chomskyian attempts at diagnosing the sociology of the media" without understanding that Chomsky's critique of media refers precisely to people like him. Lol.

Bingo. Kate Aronoff recently made pretty much this point, not about Silver specifically but polling and horse-race commentary in general.

On the whole, Nate Silver is NC, but the further he strays from modeling and into what you could only call punditry (notwithstanding his supposed contempt for the form), the more his biases show. I read The Signal and the Noise when it came out, in a reading group with a bunch of quantitative ecologists. We generally agreed it offered some useful insights about how different fields approach forecasting; I even picked up a technical idea or two.

But there were a few chapters that irritated the fuck out of me, and they were all of a piece. At the time, the worst offender was the climate change one (which prompted a now-classic rebuttal from Michael Mann titled FiveThirtyEight: The Number of Things Nate Silver Gets Wrong About Climate Change). More recently, though, I revisited parts of the book and was irritated afresh by the stuff on the 2008 financial collapse. That eventually led me to an excellent post by Cathy O'Neil that gets right to the heart of the problem:

Silver ignores politics and loves experts

Silver chooses to focus on individuals working in a tight competition and their motives and individual biases, which he understands and explains well. For him, modeling is a man versus wild type thing, working with your wits in a finite universe to win the chess game.

He spends very little time on the question of how people act inside larger systems, where a given modeler might be more interested in keeping their job or getting a big bonus than in making their model as accurate as possible.

In other words, Silver crafts an argument which ignores politics. This is Silver’s blind spot: in the real world politics often trump accuracy, and accurate mathematical models don’t matter as much as he hopes they would.

[. . .]

To be fair, there have been moments in his past when Silver delves into politics directly, like this post from the beginning of Obama’s first administration, where he starts with this (emphasis mine):

"To suggest that Obama or Geithner are tools of Wall Street and are looking out for something other than the country’s best interest is freaking asinine."

and he ends with:

"This is neither the time nor the place for mass movements — this is the time for expert opinion. Once the experts (and I’m not one of them) have reached some kind of a consensus about what the best course of action is (and they haven’t yet), then figure out who is impeding that action for political or other disingenuous reasons and tackle them — do whatever you can to remove them from the playing field. But we’re not at that stage yet."

My conclusion: Nate Silver is a man who deeply believes in experts, even when the evidence is not good that they have aligned incentives with the public.

Distrust the experts

Call me “asinine”, but I have less faith in the experts than Nate Silver: I don’t want to trust the very people who got us into this mess, while benefiting from it, to also be in charge of cleaning it up. And, being part of the Occupy movement, I obviously think that this is the time for mass movements.
User avatar
askii
julie from mod squad
julie from mod squad
 
Posts: 302
Joined: Sat Oct 31, 2015 9:53 pm
Location: Amazonia

Re: Dude: Nate Silver

Postby Get dog costumes on Wed Aug 14, 2019 4:37 pm

Andrew. wrote:Haha. How's your Wednesday going my dude? Get any funny texts from friends or loved ones? I just got in from a run and texted my girlfriend a pic of three deer (mother and two fawns) I saw down by the river. For part of my run along the Assiniboine I listened to Vish Khanna's previously unreleased 2008 interview with David Berman. It's really great.

This is the Nate Silver thread where people post opinions about Nate Silver. I've made a couple posts here critical of Silver going back to 2016 (w no mention of Bernie). I just posted a link to a Warren supporter saying that even they are annoyed by Silver's hate-on for Bernie.

I hope you're less of an angry, reactive crank in real life than you are on this old internet forum.

Canada's got talent.
User avatar
Get dog costumes
Saint Who Rules w/ Extensive Magnanimity
Saint Who Rules w/ Extensive Magnanimity
 
Posts: 2183
Joined: Wed Oct 05, 2005 10:57 pm
Location: Austin, TX

Re: Dude: Nate Silver

Postby Andrew. on Wed Aug 14, 2019 5:00 pm

askii wrote: gets right to the heart of the problem:

Silver ignores politics and loves experts
[...]
To be fair, there have been moments in his past when Silver delves into politics directly, like this post from the beginning of Obama’s first administration, where he starts with this (emphasis mine):

"To suggest that Obama or Geithner are tools of Wall Street and are looking out for something other than the country’s best interest is freaking asinine."

and he ends with:

"This is neither the time nor the place for mass movements — this is the time for expert opinion. Once the experts (and I’m not one of them) have reached some kind of a consensus about what the best course of action is (and they haven’t yet), then figure out who is impeding that action for political or other disingenuous reasons and tackle them — do whatever you can to remove them from the playing field. But we’re not at that stage yet."


Jesus Christ, these quotes are so damning they seem made up. The guy is a caricature of an elitist, status-quo technocrat.
User avatar
Andrew.
Power Incarnate with Endless Creativity
Power Incarnate with Endless Creativity
 
Posts: 7101
Joined: Mon May 07, 2007 11:02 pm

Re: Dude: Nate Silver

Postby warmowski on Thu Aug 15, 2019 9:55 am

It will be impossible to do the customary thing and ignore or triangulate to admit commercial media bias this time around; there is an actually viable candidate that is running on a platform that openly, directly attacks the business interests of the commercial media's advertisers and in the case of the WaPo, attacks ownership itself. You'd have to be thick to not expect these platforms to have their knives out for Bernie, and this bad-at-politics abacus named Nate and his 3.2M followers are no fucking exception. That's fine, and well-expected by adults, but fuck pretending it isn't happening. Fire back, absolutely, always and forever no matter how much it blows Andy's mind.

I mean, here's seven months out from the caucus, from ol' Nate's former employers:

Image

-r
User avatar
warmowski
Perfect Picture of Wisdom and Boldness
Perfect Picture of Wisdom and Boldness
 
Posts: 3072
Joined: Mon Feb 23, 2004 5:34 pm
Location: Chicago

Re: Dude: Nate Silver

Postby askii on Thu Aug 15, 2019 3:11 pm

Andrew. wrote:Jesus Christ, these quotes are so damning they seem made up. The guy is a caricature of an elitist, status-quo technocrat.

And this was years before the FiveThirtyEight relaunch and the Freakonomics-style clickbait cycle that accompanied it. Cathy "mathbabe" O'Neil's post is worth reading in full; she offers a more nuanced critique of the substance of the book, from the POV of an actual ex-Wall Street analyst.

Again, I think Silver is a good modeler, and a good pop-translator of modeling, in the domains he knows something about. And if we're going to do electoral forecasting, I would much rather it be based on rigorous, transparent* methodology than finger-in-the-wind bloviating.

Even beyond neoliberal ideology, though, the dude is prone to a kind of one-size-fits-all methodologism that leads him to speak with unearned authority about the analytical shortcomings of fields where he genuinely seems not to understand the content (the aforementioned climate change debacle) or the institutional context (O'Neil's finance and pharma examples). It's a strange sort of meritocratic hubris that dismisses or ignores the perspectives of people with actual, demonstrable expertise. It also seems more Hedgehog than Fox, by Silver's own corny taxonomy. I don't think this attitude is typical of folks with Silver's academic background, either. While some methodological issues really are universal (P-hacking, I'm looking at you), most applied statisticians recognize that domain knowledge is kind of a big deal.

*Well, kind of transparent. AFAIK Silver has never published his secret-sauce algorithm or his code, which is a significant caveat to my generally sympathetic assessment of him as an analyst.
User avatar
askii
julie from mod squad
julie from mod squad
 
Posts: 302
Joined: Sat Oct 31, 2015 9:53 pm
Location: Amazonia

PreviousNext

Return to Crap / Not Crap

Who is online

Users browsing this forum: tallchris and 17 guests