Nate Silver built his career and reputation on being a savvy understander of math and statistics. and...uhh...is this the best he can do? he says that 70% of the US is in this broadly defined...
Exemplary
That’s because the Indigo Blob is large enough to be the default. It represents about 70 percent of the country.
...
And Twitter’s users during this period skewed heavily left too, with liberals outnumbering conservatives roughly 3:1, according to Wojcieszak.
Nate Silver built his career and reputation on being a savvy understander of math and statistics. and...uhh...is this the best he can do?
he says that 70% of the US is in this broadly defined "blob" of being either left-of-center or moderate
(I disagree with this framing altogether, because it erases the differences between the left-wing, moderate liberals, and moderates, but that's a topic for another rant, and because all the statistics Silver presents are in terms of this framing, I'll stick with it for now)
then, a few paragraphs later, he says Twitter's userbase was 3:1 liberals vs. conservatives. with a tone that this is somehow undesirable and indicative of bias.
but 3:1 is equivalent to a 75% / 25% split. isn't that more or less exactly what you'd expect, given the underlying demographics? 70% vs 75% is probably within the margin of error of the studies he's relying on, and to the extent there's a small over-representation of liberals on Twitter, it seems like it could easily be explained by conservatives self-selecting for sites like Gab and Truth Social that are conservative echo-chambers where they won't get banned for egregious hate speech.
And of course, all of this was before Trump was banned from Twitter January 2021. Notwithstanding the inherent futility of trying to “deplatform” the current president, who had many other ways to communicate with his base, I get that this was a tough decision for Twitter after Trump had used the platform to help trigger the Jan. 6 insurrection.
he glosses over here that when Trump was banned from Twitter, he was only going to remain the current President for about 2 more weeks. this strikes me as Silver just wanting to complain that he thinks deplatforming is ineffective rather than anything connected to his central argument.
This is sometimes why relatively picayune arguments, like who is allowed publish an op-ed in The New York Times, blow up on Twitter.
he doesn't mention any specifics here, but the example that comes to my mind is Tom Cotton's "Send in the troops" op-ed advocating for the use of the military against protesters in the days after George Floyd's murder.
I'd be very curious to know if that's what Silver is calling a "relatively picayune" argument. looking at this wikipedia page there are plenty of other controversies but none of the other recent ones match his description of "who is allowed [to] publish an op-ed".
They are proxies for larger discussions about whether these institutions ought to be openly progressive. For instance, some public health types sought to use COVID policy to advance left/progressive political aims — and argued that the politicization of public health was a good thing.
unsurprisingly, he also glosses over attempts to push right-wing political aims as part of covid response, such as the Great Barrington Declaration, or Trump's initial downplaying of covid out of worries that it would hurt the stock market and the economy (and thus his chances of re-election)
As will not surprise you, I am strongly on the nonpartisan side on this question, although maybe a better way to put it is that I favor pluralistic institutions over a progressive monoculture.
here, Silver demonstrates what I think is one of the best examples of this sort of "centrism as a virtue unto itself" type of political thinking.
when he says "nonpartisan" or "pluralistic", what does he mean exactly? does he mean he wants to see a 50/50 split between liberals and conservatives? (as is reflected in the split between liberal and conservative politicians in the graph he shows)
if the country as a whole is 30% conservative, then wanting conservative representation to be 50% is putting a thumb on the scale in favor of conservatives. political centrism is not a neutral ideology.
I have not at all been impressed by Nate Silver post-538, especially recently. IDK if he had better editors before or what. Sort of trending toward Greenwald territory for me.
I have not at all been impressed by Nate Silver post-538, especially recently. IDK if he had better editors before or what. Sort of trending toward Greenwald territory for me.
I'm not sure what's happening with him, but on his last few appearances on the 538 podcast he was very flushed and tended to go on extended rambling tangents rather than talking about whatever the...
I'm not sure what's happening with him, but on his last few appearances on the 538 podcast he was very flushed and tended to go on extended rambling tangents rather than talking about whatever the topic of the moment was. He didn't look entirely well.
Pre-2022 my pet theory was that he saw the writing on the wall in terms of Republicans taking over the electoral system and making psephology irrelevant, so was positioning himself as an...
Pre-2022 my pet theory was that he saw the writing on the wall in terms of Republicans taking over the electoral system and making psephology irrelevant, so was positioning himself as an enlightened centrist talking head. But more and more it just looks like a terminal case of Twitter brain. Dude literally cannot stop getting into contrarian spats.
I think you're quick to accuse when you haven't really grappled with the data that this blog post is based on. I haven't gone into depth either, but here's a somewhat deeper dive: The statistics...
I think you're quick to accuse when you haven't really grappled with the data that this blog post is based on. I haven't gone into depth either, but here's a somewhat deeper dive:
The statistics about Twitter come from this paper and here's the abstract:
We find that the majority of users (60%) do not follow any political elites. Those who do follow in-group elite accounts at much higher rates than out-group accounts (90 versus 10%), share information from in-group elites 13 times more frequently than from out-group elites, and often add negative comments to the shared out-group information. Conservatives are twice as likely as liberals to share in-group versus out-group content.
I've only skimmed it, but digging into the methods:
In particular, we use 20,731,455 shares of elite accounts classified as liberal or conservative (moderates excluded) that were shared by 151,063 politically active users classified as liberal or conservative (moderates excluded as well). We included any quote tweet, even a quote tweet of someone that a user does not follow her/himself.
So it seems that "liberal" and "conservative" from this paper excludes moderates and anyone who doesn't tweet about politics at all.
(I also wonder how many were non-Americans.)
Silver's "Indigo Blob" is supposed to include centrists, so that's rather different from how the paper did it. A 3:1 ratio among partisans is quite different from a 3:1 ratio where one "side" includes centrists. I wouldn't expect those numbers to be comparable.
I sure am! one reason for that is that most of my criticisms of Silver's post (the talk about deplatforming Trump; the lack of specifics in his complaint about NYT op-ed controversies when I'm...
Exemplary
I think you're quick to accuse when you haven't really grappled with the data that this blog post is based on.
I sure am!
one reason for that is that most of my criticisms of Silver's post (the talk about deplatforming Trump; the lack of specifics in his complaint about NYT op-ed controversies when I'm pretty sure giving specifics would expose his argument as bullshit; him ignoring right-wing covid proposals while highlighting left-wing ones; and I didn't even go into the bigger problems with his section about "wokeism") have nothing to do with the specific paper he's citing.
but another reason is that I think both Silver and the paper suffer from the same fundamental flaw in their political analysis.
take a look at figure 1 in the paper, also featured in Silver's post. ignore the data for a moment, and just look at the background.
there's a line down the center labeled "moderate". where does that line come from? how did they define it? it's the axis of the graph. only in a political science paper can you get away with defining your axes this sloppily and hand-wavily.
this is the main problem I have with Silver, and with other self-identified "centrists" in the same mold (Matt Yglesias comes to mind), and including apparently the authors of this paper. they treat "the center" as some sort of Platonic ideal, something that they can define objectively, and draw as the axis of their graph, and then plot their data on top of.
I think they view the left-right spectrum as basically like the pH scale. it's a one-dimensional axis, it goes from 0 to 14, with 7 defined as being neutral. the two acid and base sides are mirror images of each other, equal in size, and are able to balance each other out. if you want to neutralize an acid with a pH of 3, you need a base with a pH of 11 (assuming equal molar volumes of each) because they're both 4 pH away from 7.
viewed through this lens, the argument Silver seems to be making is that pre-Elon Twitter was slightly acidic, maybe around 5 or 5.5 pH. and that Musk, by unbanning a lot of conservatives, is bringing it closer to a nice neutral 7.
I think the only meaningful definition of "the center" in politics is the median person. when Silver talks about how 70 percent of the country seems to be left-of-center / not-conservative, I view that as a fundamental error. 50 percent of the country is left-of-center, 50 percent is right-of-center. any definition of "the center" that yields a different result should get heavy scrutiny.
in the worldview I'm advocating, "the center" is a statistical average you take after plotting your datapoints along other axes, rather than as an artificially-defined axis that the datapoints are plotted within reference to. that's why I'm calling it a fundamental error - it's something that invalidates the entire graph you're drawing, not just some of the datapoints.
So it seems that "liberal" and "conservative" from this paper excludes moderates
wait
waitwaitwaitwaitwait
Silver's core argument is that there's a blob of "liberals and moderates" that is distinct from the blob of conservatives
and he's citing this paper to back up his argument
and the paper excludes moderates from their dataset and only looks at liberals and conservatives?
what are we even doing here. what is reality. why is this essay?
oh, but they don't always exclude moderates. sometimes they split them into liberals and conservatives (based on the arbitrary line that I think is bullshit):
Section S13 shows that our findings hold when including and classifying moderates into liberals and conservatives (using the vertical line in Fig. 1 as a cutoff point).
isn't classifying half of the moderates as conservatives...doing the exact thing that Silver is arguing against? in the paper that he's citing to support his position?
this is a general dislike I have of Silver's writing, both here and in other things I've read from him. he adopts this authoritative, "data-driven" tone. here's some scientific data. here's what it means. I'm just a numbers guy, I'm telling you what the numbers mean, I'm not putting my own bias into it, because I'm a perfectly neutral 7 on the pH scale.
but if you dig into the data he's referencing, it turns out not all of it supports the conclusions he's drawing. and he gets very sloppy in differentiating between what parts of his argument are data-based, and what parts are vibes-based.
like, what he labels as point #5 of his core hypothesis...seems to be entirely detached from this paper. it's based solely on the vibes he gets from twitter (and how many people criticize him on it):
The Indigo Blob is not an undifferentiated mass. If you look closely, it contains multitudes. However, it’s to some people’s advantage to maintain the Blob’s ambiguity. Trying to disambiguate the Blob will often make you the subject of intense criticism on Twitter, and Twitter’s architecture has tended to make such dissent painful.
Political views are not genetic traits and do not necessarily fall along standard deviations.
50 percent of the country is left-of-center, 50 percent is right-of-center. any definition of "the center" that yields a different result should get heavy scrutiny.
Political views are not genetic traits and do not necessarily fall along standard deviations.
I could be misunderstanding, but I think their point was that "the center" is an ambiguous term unless you literally mean the median person. Suppose a paper calls out Democrats as centrists,...
I could be misunderstanding, but I think their point was that "the center" is an ambiguous term unless you literally mean the median person.
Suppose a paper calls out Democrats as centrists, regardless of whether you agree with that statement, that's a judgement call about where the political "center" is. It's important to ask why the author defined center as something other than the literal median. To fully understand the paper, we should be clear-eyed about the author's definitions and any biases they might have.
Yeah, someone can be very liberal but hold certain conservative beliefs about certain things. No one is really centrist but, based on my achievements in Disco Elysium, I play one in video games....
Yeah, someone can be very liberal but hold certain conservative beliefs about certain things. No one is really centrist but, based on my achievements in Disco Elysium, I play one in video games. The idea of a centrist is essentially a broken one because we try to put it on a single axis when it's really more like one of those stat blobs with an infinite number of variables, each on a scale of one to ten. You can hold conflicting beliefs and you can hold different views on how to attain those goals - it's truly terrible.
Besides, left in America is right in Europe and right in Europe is somewhere between left in southeast Asia and full on fascists. That damn Overton window.
The blog post is only partially based on data. Here’s the opening paragraph again: It’s not a scientific paper, it’s an opinion piece based on his experience on Twitter with some data to back it...
The blog post is only partially based on data. Here’s the opening paragraph again:
This post was cooked on a griddle that has the grease stains of 1,000 arguments on and about the Internet. I’ve aimed to keep it to a reasonable length, and a lot of the arguments could be developed further. However, they are not casually tossed-off ideas. They reflect many years of thinking about American politics and many years of experience I’ve had working in the mainstream media (for The New York Times, ESPN and ABC News).
It’s not a scientific paper, it’s an opinion piece based on his experience on Twitter with some data to back it up. Writers are allowed to say what they think. It can just be based on “vibes.” That includes writers who sometimes do analysis.
You’re holding Silver to an unfair standard because you dislike him. You’ve invented a bogus standard where he’s not allowed to write about his opinions because he’s a numbers guy. Sometimes supplying some evidence that partially supports his opinions is somehow more discrediting than supplying no evidence at all (which is also allowed).
I’m going to push back on this one, skybrian. In your original response you said that @spit-evil-olive-tips was “quick to accuse” when he hadn’t “really grappled with the data”. Then when he...
I’m going to push back on this one, skybrian. In your original response you said that @spit-evil-olive-tips was “quick to accuse” when he hadn’t “really grappled with the data”. Then when he grappled with the data in response, you fell back to saying “it’s not a scientific paper, it’s an opinion piece”.
This feels a bit like Lucy and the football to me.
Yes, I probably should have made both arguments up front. I believe both that this is primarily an opinion piece and also, the data that Silver did bring up was misunderstood and being...
Yes, I probably should have made both arguments up front. I believe both that this is primarily an opinion piece and also, the data that Silver did bring up was misunderstood and being misrepresented.
I don’t think there’s compelling evidence in the blog post that the “Indigo blob” idea is true, though I can see how it might seem true, particularly to someone who got attacked on Twitter a lot. It’s a fairly vague metaphor that would be hard to even define well enough to test.
But it seems related to the observation that many institutions that aren’t explicitly conservative tend to become increasingly liberal. Left-of-center activism often does work in institutional settings?
you've got cause and effect reversed. I dislike Silver because I read his writing with a critical eye, not the other way around. and to be clear, I used to like him - I read The Signal and The...
You’re holding Silver to an unfair standard because you dislike him.
you've got cause and effect reversed.
I dislike Silver because I read his writing with a critical eye, not the other way around.
and to be clear, I used to like him - I read The Signal and The Noise when it first came out, and recommended it to friends & family. I used to be a regular listener of the 538 Politics podcast.
I've followed his career arc, including the shift he's made away from straightforward analysis of polls, and into this sort of general political punditry from a self-identified "nonpartisan" position.
and I don't think he's made that transition well. I think he's been coasting on his reputation from election modelling and forecasting, and tried to pivot into doing, as we see here, a sort of "punditry, but unlike some other pundits, mine is backed up with data because as a Numbers Guy that's what I'm good at".
the standard I'm advocating here is I think a fair one, and one that I think should be applied universally, to writers whose opinions I agree with as well as ones I disagree with.
specifically, if a pundit writes an opinion piece, and cites a scientific paper in that piece, the onus is on them to make sure the data they cite actually supports their argument rather than contradicting it (or, if the data partially supports them and partially contradicts them, to be up-front about that). and if the pundit has a mix of claims, some of which they believe are backed up by the data and some of which are just their opinion without supporting data, the onus is on them to make clear which is which.
You’ve invented a bogus standard where he’s not allowed to write about his opinions
you're putting words in my mouth here, in a rather uncharitable way.
nowhere did I say Silver shouldn't be allowed to express his opinions.
he's expressing his opinions in this Substack post, I'm expressing my opinions about the post in Tildes comments. isn't that...exactly how it's supposed to work?
this is a really unfortunately common anti-pattern that tends to come up with this sort of political argument.
I'm criticizing Silver's opinions. my opinion is that Silver's opinions are wrong.
and in defending Silver's post, you're not criticizing my opinions directly, instead you're bringing up a sort of "cancel culture" argument out of nowhere, where you assume that because I disagree with Silver, I want to silence him and not allow him to express those opinions at all.
I know this is getting a bit off topic, but I'm a bit confused by Matt Yglesias being labelled a centrist. He would be better off if he avoided twitter, but he's a staunch social democrat. At...
I know this is getting a bit off topic, but I'm a bit confused by Matt Yglesias being labelled a centrist. He would be better off if he avoided twitter, but he's a staunch social democrat. At least on his substack, Slow Boring, he advocates for policies like pricing externalities into production (carbon tax) and increasing taxes to boost social welfare. Politically, he's more aligned with Joe Biden than Joe Manchin.
Nah, you lost me here Nate. The reason why we have public health officials that dictate health policy is because they are experts in health policy and the general public is not. I'll refrain from...
For example, if public health officials would strongly prioritize “equity” when coming up with a plan to distribute vaccines but the general public would not, they should defer to the public.
Nah, you lost me here Nate. The reason why we have public health officials that dictate health policy is because they are experts in health policy and the general public is not. I'll refrain from making slippery slope arguments but you can imagine what a stupid idea this is if you apply it to any other field in which leaders with education and experience make decisions on behalf of the public and imagine for yourself the consequences of pursuing this kind of pseudoegalitarian nonsense.
Individuals are a mixed bag, some are intelligent and some are not. Groups are dumber. The general public, as the largest group, is the dumbest of the lot. The general public possesses no special insight or reasoning ability or perspective that would make overriding the considered judgment of health policy officials to make even a little bit of sense.
I already have to live in a society where the general public dictates political leadership. That's an awfully mixed bag even when things operate as they should, without thousands of thumbs and billions of dollars pressed on the scales by people pursuing their own agendas, but it's the way our country works so we do the best we can. That said, I do not want to live in a society where the general public (and all the stupidity and partisanship and conspiratorial thinking and dark money that goes along with that) dictates health policy.
Heath experts aren't experts in everything, though, and the wisdom of their plans varies. They might not be experts in distribution or understand all the consequences of their policies. They might...
Heath experts aren't experts in everything, though, and the wisdom of their plans varies. They might not be experts in distribution or understand all the consequences of their policies. They might not understand what happens when people try to carry out their policies.
In the SF bay area, there was a brief time when whether you got the vaccine depended on which zip code you were in. That was nuts!
Here's a long article by Patrick McKenzie who got to see what happened when the plans of well-meaning officials got implemented.
I've spent years working for and with Bay-area startups, and while I agree that in this specific instance -- the logistics of early vaccine rollout -- government dropped the ball, I'm not sure I...
I've spent years working for and with Bay-area startups, and while I agree that in this specific instance -- the logistics of early vaccine rollout -- government dropped the ball, I'm not sure I follow you from "plucky tech startups save the day by moving fast and breaking things" to "public health policy should be dictated by plucky tech startups".
For every success story and well-meaning tech entrepreneur in this area there's another Elizabeth Holmes or Martin Shkreli just waiting for an opportunity to get rich by convincing investors that public health in America is ripe for disruption. Thanks, but your "general public > health policy officials" argument was bad enough before you shifted to "techbros > health policy officials".
I didn’t say that and I don’t think that’s what the article was about. The question is really why the hodgepodge of government organizations responding to a crisis couldn’t put together some...
I didn’t say that and I don’t think that’s what the article was about. The question is really why the hodgepodge of government organizations responding to a crisis couldn’t put together some better solutions on their own. Why was there such a big gap for an outsider to fill?
It seems that in part, it was because they were more concerned about “equity” than logistics. The asked for things that made the problem a lot harder than it needed to be. There were multi-page forms you had to fill out online to see if you were eligible. They could have done it primarily by age and also given the staff making appointments and giving out the vaccine some discretion, and not worried too much about some people getting the vaccine who shouldn’t have, as long as it’s approximately right.
More generally, I don’t actually think that “deferring to the public” is always the wisest policy, but it’s a vague suggestion that covers a lot of ground, and I can imagine good ways to do it. Just because you run a focus group doesn’t mean you always do what they say or let them do your thinking for you, but it does give you a better idea about how people think, and that’s helpful.
We don’t need to imagine a bad way of implementing a vague suggestion and assume that’s what Silver meant.
We're getting off-topic from the original post, but I'd recommend rereading skybrian's comment that you're responding to. The context is San Francisco and ultra-progressive cities having overly...
We're getting off-topic from the original post, but I'd recommend rereading skybrian's comment that you're responding to. The context is San Francisco and ultra-progressive cities having overly complicated criteria to meet nebulous "equity" goals. Trump was absolutely not the reason that San Fransisco used arduous forms and measures like zip code to dole out vaccines when they first became available.
Early in the vaccination campaign, there were targeted efforts to increase vaccination rates amongst racial and ethnic minorities in the US. The public health campaigns worked well for those communities. As you point out, White Americans are still lagging in vaccination rates largely due to conservative politicians and conspiracy theorists.
The core unique insight VaccinateCA had was that America has access to a reliable technology for getting information from the healthcare system. It is called a telephone.
This was our mantra on accuracy, and we repeated it to healthcare providers, the government, funders, and publishers. We felt like imposters, but when we needed it, I used every trick I knew to present as the CEO of an officially recognized healthcare nonprofit and elide the fact that my sole prior leadership experience had been running a World of Warcraft guild.
Never really was a Twitter user but I’d always hear about reactions to things trending on Twitter pre-Elon and think, “this seems like an amplified leftist echo-chamber.” I’m relatively...
Never really was a Twitter user but I’d always hear about reactions to things trending on Twitter pre-Elon and think, “this seems like an amplified leftist echo-chamber.” I’m relatively left-leaning but this did not seem like a realistic representation of the viewpoints of the masses. I don’t know if there can ever be a place like that on the internet.
One issue I have with the way the media talks about popular political opinion (and Nate Silver is a big perpetrator of this too) is that they always talk as if the people at large are forming...
Exemplary
One issue I have with the way the media talks about popular political opinion (and Nate Silver is a big perpetrator of this too) is that they always talk as if the people at large are forming their political positions and perspectives based on some exogenous source of information and narratives that is outside the political media, and then the political media is there to do this anthropological exercise where they impartially come in and figure out what the people are really thinking "outside" the bubble.
But this is nonsense. The media creates the narratives the masses are thinking. The things people prioritize are consequences of what the media chooses to spend its time talking about and how it chooses to talk about them. It's not a perfect transmission and a lot of processing happens along the way, but political journalists in general are completely unable to understand the reflexivity baked into the relationship here. People don't just have immutable opinions embedded in their souls that lead them to gravitate to one set of media outlets and not another. The media outlets are creating the spectrum of opinions. They are inculcating sense of who is good and who is bad and what right and wrong is in their listeners.
When Silver divides up the "Indigo Blob" into left and center, the divisions of what's blue + gray and what's red is arbitrary. He talks as if the blob has a liberal bias and the other 30% of America is naturally gravitating to the other red mode. But he has the causal relationship here exactly backwards. The other 30% of America is deep red and unable to engage with the stuff in the blob because they spend their time in a specific media echo chamber that is designed to radicalize people and close them off from other points of view. The Indigo blob doesn't engage with the red stuff because that mode explicitly goes out of its way to characterize the blob as liberal and out of touch, and engages in eliminationist rhetoric about them.
The real story here is that we have a parallel media sphere that operates in a realm that is agnostic to facts or evidence and is completely in the tank for a specific political agenda. They have eschewed general standards of journalistic practice and adopted a position as propagandists to advance a that agenda under the guise of journalism. By doing so, they have basically starved mainstream media out of any useful source of intelligent or humane conservative opinions. Everyone has grown more and more unhinged, more dependent on extremist language, and focused on political ends rather than good journalistic practice. This makes them unable to engage productively with anyone outside the red bubble. If you're not a wingnut, it becomes harder and harder to actually have a productive conversation with someone who has marinated too long in the red bubble because their view of reality is insane and distorted.
The "Indigo Blob," on the other hand, has a wide tent and while the leftist side of it can punch above their weight, there is plenty of dissenting opinion from the grays to at least keep the discussion as a whole balanced and at least tries to align itself towards figuring out what objective reality looks like, even if it's skewed. There are the histrionic nutters, but once the general consensus figured out how to avoid the more obvious rhetorical traps they set, people got comfortable making fun of them, telling them to log off and touch grass. Nobody wants to put those people in real positions of power outside of special little carve outs to placate some political faction.
"The media creates the narratives" is somewhat true, but that treats it as one-way. As you say later, it's reflexive. The media is often amplifying things already out there. Their sources include...
"The media creates the narratives" is somewhat true, but that treats it as one-way. As you say later, it's reflexive. The media is often amplifying things already out there. Their sources include social media and politicians. This makes cause and effect difficult to untangle.
Many writers (most of us) have little influence. I think it's reasonable for us to assume that our own writing has a negligible effect on the opinions of millions and billions of people out there in a nation or the world. It's reasonable to try to understand what's going on in the world while thinking of ourselves as observers.
For celebrities, it will be somewhat less true, and for people who are often the "main character" like Musk, it's definitely not true. But even CEO's and politicians are often reacting to public sentiment. Some of the things they say get a lot of attention and other things get ignored. Some people have substantial power and influence, but their zones of control are limited and their plans often fail or result in unintended consequences.
They, too, are affected by the Algorithm. I'm making up a non-standard definition that includes the behaviors of newspapers and television and bloggers and the masses and bots. There are many algorithms that add up to one Algorithm, much like the difference between internets and the Internet. Nobody controls the Algorithm, not even social networks. Something you tossed off will go viral and articles you spent weeks or months researching will be ignored. What you say may be repeated and distorted until it's unrecognizable, but you will be "held responsible."
I expect nearly all writers think reflexively. We are curious about the effects of the Algorithm on how their messages reach an audience. The Algorithm can be studied, but it can't be entirely predicted. That's why surveys are useful. Writers can only partially predict their effects on readers and are often surprised. They still want to know what other people think, even if they're fully aware that the masses got much of their thinking from the media and they control a tiny bit of the media.
Writers do hope to have some effect on their audience. It's a reason for writing. Nate Silver's analysis has probably had some effect on elections because others use it to gauge how well candidates are doing. Campaigns have probably used 538 to decide where to spend effort improving their numbers. More generally, the media is partially aware of their own effects, which is why they don't report results on election day until polls have closed. But I think it would be arrogant for Silver to assume he personally affects the US and the world all that much. There are other poll aggregators who do much the same thing and would continue to do so without him.
Also, regarding being "unable to engage with red stuff," I think that's a choice, because a lot of it is out there on the public Internet and we could investigate it if we wanted to. I personally choose not to, but I am honor-bound to support curiosity, so I think people who are genuinely curious about how other people think should investigate and report back. Journalists can even talk to people and do reporting in person, which is likely to result in a better picture than looking around online.
It would be helpful if people reporting back would say more about what they saw than how they felt about it, and if we didn't dismiss their reporting if it seemed strange. It would also be helpful if they weren't drowned out by us wanting to share our opinions even though we weren't there and we didn't do any investigation at all. I don't think we have a good handle on how to moderate for that yet.
I don't know of any good sources of "intelligent or humane" conservative opinions, but I do know a few sources of interesting libertarian opinions. I've learned to be selective about sharing those links, though.
This was a very well thought-out response, and I appreciate it. Again, since I wasn’t really a Twitter user I can see how the narrative about the site could get distorted to fit someone’s agenda...
This was a very well thought-out response, and I appreciate it. Again, since I wasn’t really a Twitter user I can see how the narrative about the site could get distorted to fit someone’s agenda as well.
It is because it recommends what it thinks you want to see. It does the same thing for right wingers. Now that Musk has taken over it just seems to be shoving right wing content at everybody. I...
I’d always hear about reactions to things trending on Twitter pre-Elon and think, “this seems like an amplified leftist echo-chamber.” I’m relatively left-leaning
It is because it recommends what it thinks you want to see. It does the same thing for right wingers. Now that Musk has taken over it just seems to be shoving right wing content at everybody. I have Twitter on my phone only for notifications from one particular Twitter account, and every time I go on it tries to shove alt right bullshit (and Musk's posts, despite him being blocked) in my face.
Twitter is a piece of infrastructure that many rely on for emergency information - deleting it unfortunately isn't a possibility for some until that infrastructure is replaced.
Twitter is a piece of infrastructure that many rely on for emergency information - deleting it unfortunately isn't a possibility for some until that infrastructure is replaced.
This extension is a must for anyone still using Twitter these days: Control Panel for Twitter for Firefox and for Chrome. it removes all the algorithmic crap and makes Twitter still usable.
I have Twitter on my phone only for notifications from one particular Twitter account, and every time I go on it tries to shove alt right bullshit (and Musk's posts, despite him being blocked) in my face.
This extension is a must for anyone still using Twitter these days: Control Panel for Twitterfor Firefox and for Chrome. it removes all the algorithmic crap and makes Twitter still usable.
Unfortunately I only use it on my phone because notifications, and most of the time I don't even open it when I get those. I don't actively use Twitter at all or have any plans to, I never visit...
Unfortunately I only use it on my phone because notifications, and most of the time I don't even open it when I get those.
I don't actively use Twitter at all or have any plans to, I never visit it on my desktop. Unlike Reddit, which does have a wealth of useful info and I find it harder to get away from, Twitter holds almost nothing of value for me.
I can't really believe that anyone actually believes any "trending" sections on the web though. Do people believe the "trending" tab on Netflix is actually shows people are watching and not just...
I can't really believe that anyone actually believes any "trending" sections on the web though. Do people believe the "trending" tab on Netflix is actually shows people are watching and not just shows Netflix wants them to watch?
Twitter wasn't ever trying to be a source for "unbiased" reporting. It's just a way to get news. Poker news or animation news or political news. I always felt like it was kinda up to me to pick by choosing who I followed.
I dunno, it is the application that created the concept of a hashtag. Its entire purpose is virality, not to mention any algorithm-centric feed-based application is intended to take control out of...
I dunno, it is the application that created the concept of a hashtag. Its entire purpose is virality, not to mention any algorithm-centric feed-based application is intended to take control out of the users’ hands for better or for worse. It is only up to you to a certain degree. I don’t necessarily want to validate this article or argue about its merits but I do not think you are not some kind of pleb if you fall victim to an algorithm.
I’m basing on the idea that he’s a statistician and that I believe he probably knows the framework of his thesis is skewed. - see all the other very valid and pretty obvious holes pointed out on...
I’m basing on the idea that he’s a statistician and that I believe he probably knows the framework of his thesis is skewed. - see all the other very valid and pretty obvious holes pointed out on this thread for evidence of this - and then consider the source (Twitter) and the self-selection, manipulation, and exploitation that’s happened over years to warp the conversation. The branch he’s hanging his thesis on IMO has little to no real-world underpinning.
Edit: I don’t think he’s insincere. I think he’s framing a disingenuous argument by declaring “this is the middle!”. It’s not. It’s a veiled blue “land grab” to try and ‘both sides’ a 90-10 conversation.
I think you’re right that the foundations for making this sort of argument are shaky. First of all, how is anyone supposed to know what “the middle” is? It’s mostly vibes-based based on what we...
I think you’re right that the foundations for making this sort of argument are shaky.
First of all, how is anyone supposed to know what “the middle” is? It’s mostly vibes-based based on what we see in our particular bubbles.
That is, with the exception of opinion polls, which are tricky to interpret since it depends on the exact wording of the question. We often don’t know what people mean when they answer a survey a certain way.
For elections, there is a ground truth. How do people vote in the end? This is why statistics works at all. Even that is subject to misinterpretation about “mandates” or whatever. We often don’t really know why people voted a certain way. Did they misunderstand something? Maybe they didn’t like the vibes? But there is a voting algorithm and returned an answer, and with a fair bit of work, some useful predictions can be made.
The thing is, I think the fact that this is all kind of vague and shaky means we should have more charity when reading other people’s opinions? It’s fine to point out that the argument might not be all that solid, and quite another to make the sort of accusations I see in this topic, which aren’t just about disagreement but about how he’s doing something terrible (“malicious”) by writing what he thinks.
The argument that he’s doing something terrible requires a lot more confidence in both your own understanding of the US (many millions of people you haven’t met in many places you haven’t visited) and what he’s privately thinking than makes sense to me.
All valid. I was pretty harsh yesterday. Part of my condemnation of Silver was coming from some half-remembered info about his work from a podcast (I think it was The Daily Zeitgeist) I listen to....
All valid. I was pretty harsh yesterday. Part of my condemnation of Silver was coming from some half-remembered info about his work from a podcast (I think it was The Daily Zeitgeist) I listen to. I do think as other have said - he's dabbling in punditry more these days and I suspect he's more conservative than he lets on. Sometimes my frustration at the state of affairs in 2023 gets the better of me.
Among politicians on Twitter, we see a traditional bimodal distribution with liberal and conservative peaks corresponding to the Democratic and Republican parties. But everything else is left-skewed. There’s not a clear distinction between left-wing and centrist news organizations — they’re all part of the Indigo Blob — and the right-wing media isn’t terribly well-represented. The same holds for journalists, although the right flank is even more attenuated. And Twitter’s users during this period skewed heavily left too, with liberals outnumbering conservatives roughly 3:1, according to Wojcieszak. Anyone who was active on the platform during this period knew that content satisfying to progressives and Democrats was more likely to be rewarded with likes and retweets.
I believe this contributed to the misestimation of the political preferences of the American public by journalists, Twitter-savvy politicians and others who were active on the platform. Pre-Elon Twitter contained a comprehensive enough set of viewpoints that you could feel like everyone was there, but in fact it differed from the overall U.S. electorate in many important ways. For instance, a New York Times/Upshot study in 2019 found that Democrats on Twitter were far more progressive, far more white and far more likely to be college-educated than those in the Democratic electorate overall. That’s why Elizabeth Warren won the Twitter primary — while Joe Biden became the actual Democratic nominee.
It seems pretty clear to me that this work from Nate Silver is disingenuous at best, but really more likely straight up malicious. I’ve not seen anyone even mention that the body of Twitter isn’t...
It seems pretty clear to me that this work from Nate Silver is disingenuous at best, but really more likely straight up malicious. I’ve not seen anyone even mention that the body of Twitter isn’t by it’s very nature ‘neutral’ - it was a naturally formed social body built on virality and popularity - on follows and likes.
Point being- as others alluded - there is a valid argument that Nate’s “middle ground” actually sits deep in a sea of blue. Forget indigo. Nate’s “work” here seems like smoke screen propaganda trying to cover up a reality where there are like 15 very confused moderate people in the country left, a mob of very loud hurting and often hateful GOP shills misinformed and manipulated into being used by bad actors… and then the rest of us- who know the problem is we don’t live in a representative democracy anymore. It’s been hijacked by dark and big money and there’s little to nothing we can do about it except vote at elections and watch things continue to diverge from the reality we live day-to-day.
Silver had a time when he kicked the right into action with some better use of stats. Now he’s just a propagandist.
Nate Silver built his career and reputation on being a savvy understander of math and statistics. and...uhh...is this the best he can do?
he says that 70% of the US is in this broadly defined "blob" of being either left-of-center or moderate
(I disagree with this framing altogether, because it erases the differences between the left-wing, moderate liberals, and moderates, but that's a topic for another rant, and because all the statistics Silver presents are in terms of this framing, I'll stick with it for now)
then, a few paragraphs later, he says Twitter's userbase was 3:1 liberals vs. conservatives. with a tone that this is somehow undesirable and indicative of bias.
but 3:1 is equivalent to a 75% / 25% split. isn't that more or less exactly what you'd expect, given the underlying demographics? 70% vs 75% is probably within the margin of error of the studies he's relying on, and to the extent there's a small over-representation of liberals on Twitter, it seems like it could easily be explained by conservatives self-selecting for sites like Gab and Truth Social that are conservative echo-chambers where they won't get banned for egregious hate speech.
he glosses over here that when Trump was banned from Twitter, he was only going to remain the current President for about 2 more weeks. this strikes me as Silver just wanting to complain that he thinks deplatforming is ineffective rather than anything connected to his central argument.
he doesn't mention any specifics here, but the example that comes to my mind is Tom Cotton's "Send in the troops" op-ed advocating for the use of the military against protesters in the days after George Floyd's murder.
I'd be very curious to know if that's what Silver is calling a "relatively picayune" argument. looking at this wikipedia page there are plenty of other controversies but none of the other recent ones match his description of "who is allowed [to] publish an op-ed".
unsurprisingly, he also glosses over attempts to push right-wing political aims as part of covid response, such as the Great Barrington Declaration, or Trump's initial downplaying of covid out of worries that it would hurt the stock market and the economy (and thus his chances of re-election)
here, Silver demonstrates what I think is one of the best examples of this sort of "centrism as a virtue unto itself" type of political thinking.
when he says "nonpartisan" or "pluralistic", what does he mean exactly? does he mean he wants to see a 50/50 split between liberals and conservatives? (as is reflected in the split between liberal and conservative politicians in the graph he shows)
if the country as a whole is 30% conservative, then wanting conservative representation to be 50% is putting a thumb on the scale in favor of conservatives. political centrism is not a neutral ideology.
I have not at all been impressed by Nate Silver post-538, especially recently. IDK if he had better editors before or what. Sort of trending toward Greenwald territory for me.
I'm not sure what's happening with him, but on his last few appearances on the 538 podcast he was very flushed and tended to go on extended rambling tangents rather than talking about whatever the topic of the moment was. He didn't look entirely well.
Pre-2022 my pet theory was that he saw the writing on the wall in terms of Republicans taking over the electoral system and making psephology irrelevant, so was positioning himself as an enlightened centrist talking head. But more and more it just looks like a terminal case of Twitter brain. Dude literally cannot stop getting into contrarian spats.
I think you're quick to accuse when you haven't really grappled with the data that this blog post is based on. I haven't gone into depth either, but here's a somewhat deeper dive:
The statistics about Twitter come from this paper and here's the abstract:
I've only skimmed it, but digging into the methods:
So it seems that "liberal" and "conservative" from this paper excludes moderates and anyone who doesn't tweet about politics at all.
(I also wonder how many were non-Americans.)
Silver's "Indigo Blob" is supposed to include centrists, so that's rather different from how the paper did it. A 3:1 ratio among partisans is quite different from a 3:1 ratio where one "side" includes centrists. I wouldn't expect those numbers to be comparable.
I sure am!
one reason for that is that most of my criticisms of Silver's post (the talk about deplatforming Trump; the lack of specifics in his complaint about NYT op-ed controversies when I'm pretty sure giving specifics would expose his argument as bullshit; him ignoring right-wing covid proposals while highlighting left-wing ones; and I didn't even go into the bigger problems with his section about "wokeism") have nothing to do with the specific paper he's citing.
but another reason is that I think both Silver and the paper suffer from the same fundamental flaw in their political analysis.
take a look at figure 1 in the paper, also featured in Silver's post. ignore the data for a moment, and just look at the background.
there's a line down the center labeled "moderate". where does that line come from? how did they define it? it's the axis of the graph. only in a political science paper can you get away with defining your axes this sloppily and hand-wavily.
this is the main problem I have with Silver, and with other self-identified "centrists" in the same mold (Matt Yglesias comes to mind), and including apparently the authors of this paper. they treat "the center" as some sort of Platonic ideal, something that they can define objectively, and draw as the axis of their graph, and then plot their data on top of.
I think they view the left-right spectrum as basically like the pH scale. it's a one-dimensional axis, it goes from 0 to 14, with 7 defined as being neutral. the two acid and base sides are mirror images of each other, equal in size, and are able to balance each other out. if you want to neutralize an acid with a pH of 3, you need a base with a pH of 11 (assuming equal molar volumes of each) because they're both 4 pH away from 7.
viewed through this lens, the argument Silver seems to be making is that pre-Elon Twitter was slightly acidic, maybe around 5 or 5.5 pH. and that Musk, by unbanning a lot of conservatives, is bringing it closer to a nice neutral 7.
I think the only meaningful definition of "the center" in politics is the median person. when Silver talks about how 70 percent of the country seems to be left-of-center / not-conservative, I view that as a fundamental error. 50 percent of the country is left-of-center, 50 percent is right-of-center. any definition of "the center" that yields a different result should get heavy scrutiny.
in the worldview I'm advocating, "the center" is a statistical average you take after plotting your datapoints along other axes, rather than as an artificially-defined axis that the datapoints are plotted within reference to. that's why I'm calling it a fundamental error - it's something that invalidates the entire graph you're drawing, not just some of the datapoints.
wait
waitwaitwaitwaitwait
Silver's core argument is that there's a blob of "liberals and moderates" that is distinct from the blob of conservatives
and he's citing this paper to back up his argument
and the paper excludes moderates from their dataset and only looks at liberals and conservatives?
what are we even doing here. what is reality. why is this essay?
oh, but they don't always exclude moderates. sometimes they split them into liberals and conservatives (based on the arbitrary line that I think is bullshit):
isn't classifying half of the moderates as conservatives...doing the exact thing that Silver is arguing against? in the paper that he's citing to support his position?
this is a general dislike I have of Silver's writing, both here and in other things I've read from him. he adopts this authoritative, "data-driven" tone. here's some scientific data. here's what it means. I'm just a numbers guy, I'm telling you what the numbers mean, I'm not putting my own bias into it, because I'm a perfectly neutral 7 on the pH scale.
but if you dig into the data he's referencing, it turns out not all of it supports the conclusions he's drawing. and he gets very sloppy in differentiating between what parts of his argument are data-based, and what parts are vibes-based.
like, what he labels as point #5 of his core hypothesis...seems to be entirely detached from this paper. it's based solely on the vibes he gets from twitter (and how many people criticize him on it):
Political views are not genetic traits and do not necessarily fall along standard deviations.
I could be misunderstanding, but I think their point was that "the center" is an ambiguous term unless you literally mean the median person.
Suppose a paper calls out Democrats as centrists, regardless of whether you agree with that statement, that's a judgement call about where the political "center" is. It's important to ask why the author defined center as something other than the literal median. To fully understand the paper, we should be clear-eyed about the author's definitions and any biases they might have.
Yeah, someone can be very liberal but hold certain conservative beliefs about certain things. No one is really centrist but, based on my achievements in Disco Elysium, I play one in video games. The idea of a centrist is essentially a broken one because we try to put it on a single axis when it's really more like one of those stat blobs with an infinite number of variables, each on a scale of one to ten. You can hold conflicting beliefs and you can hold different views on how to attain those goals - it's truly terrible.
Besides, left in America is right in Europe and right in Europe is somewhere between left in southeast Asia and full on fascists. That damn Overton window.
The blog post is only partially based on data. Here’s the opening paragraph again:
It’s not a scientific paper, it’s an opinion piece based on his experience on Twitter with some data to back it up. Writers are allowed to say what they think. It can just be based on “vibes.” That includes writers who sometimes do analysis.
You’re holding Silver to an unfair standard because you dislike him. You’ve invented a bogus standard where he’s not allowed to write about his opinions because he’s a numbers guy. Sometimes supplying some evidence that partially supports his opinions is somehow more discrediting than supplying no evidence at all (which is also allowed).
I’m going to push back on this one, skybrian. In your original response you said that @spit-evil-olive-tips was “quick to accuse” when he hadn’t “really grappled with the data”. Then when he grappled with the data in response, you fell back to saying “it’s not a scientific paper, it’s an opinion piece”.
This feels a bit like Lucy and the football to me.
Yes, I probably should have made both arguments up front. I believe both that this is primarily an opinion piece and also, the data that Silver did bring up was misunderstood and being misrepresented.
I don’t think there’s compelling evidence in the blog post that the “Indigo blob” idea is true, though I can see how it might seem true, particularly to someone who got attacked on Twitter a lot. It’s a fairly vague metaphor that would be hard to even define well enough to test.
But it seems related to the observation that many institutions that aren’t explicitly conservative tend to become increasingly liberal. Left-of-center activism often does work in institutional settings?
That’s just my vibe-based opinion, though.
you've got cause and effect reversed.
I dislike Silver because I read his writing with a critical eye, not the other way around.
and to be clear, I used to like him - I read The Signal and The Noise when it first came out, and recommended it to friends & family. I used to be a regular listener of the 538 Politics podcast.
I've followed his career arc, including the shift he's made away from straightforward analysis of polls, and into this sort of general political punditry from a self-identified "nonpartisan" position.
and I don't think he's made that transition well. I think he's been coasting on his reputation from election modelling and forecasting, and tried to pivot into doing, as we see here, a sort of "punditry, but unlike some other pundits, mine is backed up with data because as a Numbers Guy that's what I'm good at".
the standard I'm advocating here is I think a fair one, and one that I think should be applied universally, to writers whose opinions I agree with as well as ones I disagree with.
specifically, if a pundit writes an opinion piece, and cites a scientific paper in that piece, the onus is on them to make sure the data they cite actually supports their argument rather than contradicting it (or, if the data partially supports them and partially contradicts them, to be up-front about that). and if the pundit has a mix of claims, some of which they believe are backed up by the data and some of which are just their opinion without supporting data, the onus is on them to make clear which is which.
you're putting words in my mouth here, in a rather uncharitable way.
nowhere did I say Silver shouldn't be allowed to express his opinions.
he's expressing his opinions in this Substack post, I'm expressing my opinions about the post in Tildes comments. isn't that...exactly how it's supposed to work?
this is a really unfortunately common anti-pattern that tends to come up with this sort of political argument.
I'm criticizing Silver's opinions. my opinion is that Silver's opinions are wrong.
and in defending Silver's post, you're not criticizing my opinions directly, instead you're bringing up a sort of "cancel culture" argument out of nowhere, where you assume that because I disagree with Silver, I want to silence him and not allow him to express those opinions at all.
I know this is getting a bit off topic, but I'm a bit confused by Matt Yglesias being labelled a centrist. He would be better off if he avoided twitter, but he's a staunch social democrat. At least on his substack, Slow Boring, he advocates for policies like pricing externalities into production (carbon tax) and increasing taxes to boost social welfare. Politically, he's more aligned with Joe Biden than Joe Manchin.
Nah, you lost me here Nate. The reason why we have public health officials that dictate health policy is because they are experts in health policy and the general public is not. I'll refrain from making slippery slope arguments but you can imagine what a stupid idea this is if you apply it to any other field in which leaders with education and experience make decisions on behalf of the public and imagine for yourself the consequences of pursuing this kind of pseudoegalitarian nonsense.
Individuals are a mixed bag, some are intelligent and some are not. Groups are dumber. The general public, as the largest group, is the dumbest of the lot. The general public possesses no special insight or reasoning ability or perspective that would make overriding the considered judgment of health policy officials to make even a little bit of sense.
I already have to live in a society where the general public dictates political leadership. That's an awfully mixed bag even when things operate as they should, without thousands of thumbs and billions of dollars pressed on the scales by people pursuing their own agendas, but it's the way our country works so we do the best we can. That said, I do not want to live in a society where the general public (and all the stupidity and partisanship and conspiratorial thinking and dark money that goes along with that) dictates health policy.
Heath experts aren't experts in everything, though, and the wisdom of their plans varies. They might not be experts in distribution or understand all the consequences of their policies. They might not understand what happens when people try to carry out their policies.
In the SF bay area, there was a brief time when whether you got the vaccine depended on which zip code you were in. That was nuts!
Here's a long article by Patrick McKenzie who got to see what happened when the plans of well-meaning officials got implemented.
I've spent years working for and with Bay-area startups, and while I agree that in this specific instance -- the logistics of early vaccine rollout -- government dropped the ball, I'm not sure I follow you from "plucky tech startups save the day by moving fast and breaking things" to "public health policy should be dictated by plucky tech startups".
For every success story and well-meaning tech entrepreneur in this area there's another Elizabeth Holmes or Martin Shkreli just waiting for an opportunity to get rich by convincing investors that public health in America is ripe for disruption. Thanks, but your "general public > health policy officials" argument was bad enough before you shifted to "techbros > health policy officials".
I didn’t say that and I don’t think that’s what the article was about. The question is really why the hodgepodge of government organizations responding to a crisis couldn’t put together some better solutions on their own. Why was there such a big gap for an outsider to fill?
It seems that in part, it was because they were more concerned about “equity” than logistics. The asked for things that made the problem a lot harder than it needed to be. There were multi-page forms you had to fill out online to see if you were eligible. They could have done it primarily by age and also given the staff making appointments and giving out the vaccine some discretion, and not worried too much about some people getting the vaccine who shouldn’t have, as long as it’s approximately right.
More generally, I don’t actually think that “deferring to the public” is always the wisest policy, but it’s a vague suggestion that covers a lot of ground, and I can imagine good ways to do it. Just because you run a focus group doesn’t mean you always do what they say or let them do your thinking for you, but it does give you a better idea about how people think, and that’s helpful.
We don’t need to imagine a bad way of implementing a vague suggestion and assume that’s what Silver meant.
It was because of trump. Conservative states followed him and is followers rejected any efforts for help for conspiracy theories.
We're getting off-topic from the original post, but I'd recommend rereading skybrian's comment that you're responding to. The context is San Francisco and ultra-progressive cities having overly complicated criteria to meet nebulous "equity" goals. Trump was absolutely not the reason that San Fransisco used arduous forms and measures like zip code to dole out vaccines when they first became available.
Early in the vaccination campaign, there were targeted efforts to increase vaccination rates amongst racial and ethnic minorities in the US. The public health campaigns worked well for those communities. As you point out, White Americans are still lagging in vaccination rates largely due to conservative politicians and conspiracy theorists.
This article is great.
Never really was a Twitter user but I’d always hear about reactions to things trending on Twitter pre-Elon and think, “this seems like an amplified leftist echo-chamber.” I’m relatively left-leaning but this did not seem like a realistic representation of the viewpoints of the masses. I don’t know if there can ever be a place like that on the internet.
One issue I have with the way the media talks about popular political opinion (and Nate Silver is a big perpetrator of this too) is that they always talk as if the people at large are forming their political positions and perspectives based on some exogenous source of information and narratives that is outside the political media, and then the political media is there to do this anthropological exercise where they impartially come in and figure out what the people are really thinking "outside" the bubble.
But this is nonsense. The media creates the narratives the masses are thinking. The things people prioritize are consequences of what the media chooses to spend its time talking about and how it chooses to talk about them. It's not a perfect transmission and a lot of processing happens along the way, but political journalists in general are completely unable to understand the reflexivity baked into the relationship here. People don't just have immutable opinions embedded in their souls that lead them to gravitate to one set of media outlets and not another. The media outlets are creating the spectrum of opinions. They are inculcating sense of who is good and who is bad and what right and wrong is in their listeners.
When Silver divides up the "Indigo Blob" into left and center, the divisions of what's blue + gray and what's red is arbitrary. He talks as if the blob has a liberal bias and the other 30% of America is naturally gravitating to the other red mode. But he has the causal relationship here exactly backwards. The other 30% of America is deep red and unable to engage with the stuff in the blob because they spend their time in a specific media echo chamber that is designed to radicalize people and close them off from other points of view. The Indigo blob doesn't engage with the red stuff because that mode explicitly goes out of its way to characterize the blob as liberal and out of touch, and engages in eliminationist rhetoric about them.
The real story here is that we have a parallel media sphere that operates in a realm that is agnostic to facts or evidence and is completely in the tank for a specific political agenda. They have eschewed general standards of journalistic practice and adopted a position as propagandists to advance a that agenda under the guise of journalism. By doing so, they have basically starved mainstream media out of any useful source of intelligent or humane conservative opinions. Everyone has grown more and more unhinged, more dependent on extremist language, and focused on political ends rather than good journalistic practice. This makes them unable to engage productively with anyone outside the red bubble. If you're not a wingnut, it becomes harder and harder to actually have a productive conversation with someone who has marinated too long in the red bubble because their view of reality is insane and distorted.
The "Indigo Blob," on the other hand, has a wide tent and while the leftist side of it can punch above their weight, there is plenty of dissenting opinion from the grays to at least keep the discussion as a whole balanced and at least tries to align itself towards figuring out what objective reality looks like, even if it's skewed. There are the histrionic nutters, but once the general consensus figured out how to avoid the more obvious rhetorical traps they set, people got comfortable making fun of them, telling them to log off and touch grass. Nobody wants to put those people in real positions of power outside of special little carve outs to placate some political faction.
"The media creates the narratives" is somewhat true, but that treats it as one-way. As you say later, it's reflexive. The media is often amplifying things already out there. Their sources include social media and politicians. This makes cause and effect difficult to untangle.
Many writers (most of us) have little influence. I think it's reasonable for us to assume that our own writing has a negligible effect on the opinions of millions and billions of people out there in a nation or the world. It's reasonable to try to understand what's going on in the world while thinking of ourselves as observers.
For celebrities, it will be somewhat less true, and for people who are often the "main character" like Musk, it's definitely not true. But even CEO's and politicians are often reacting to public sentiment. Some of the things they say get a lot of attention and other things get ignored. Some people have substantial power and influence, but their zones of control are limited and their plans often fail or result in unintended consequences.
They, too, are affected by the Algorithm. I'm making up a non-standard definition that includes the behaviors of newspapers and television and bloggers and the masses and bots. There are many algorithms that add up to one Algorithm, much like the difference between internets and the Internet. Nobody controls the Algorithm, not even social networks. Something you tossed off will go viral and articles you spent weeks or months researching will be ignored. What you say may be repeated and distorted until it's unrecognizable, but you will be "held responsible."
I expect nearly all writers think reflexively. We are curious about the effects of the Algorithm on how their messages reach an audience. The Algorithm can be studied, but it can't be entirely predicted. That's why surveys are useful. Writers can only partially predict their effects on readers and are often surprised. They still want to know what other people think, even if they're fully aware that the masses got much of their thinking from the media and they control a tiny bit of the media.
Writers do hope to have some effect on their audience. It's a reason for writing. Nate Silver's analysis has probably had some effect on elections because others use it to gauge how well candidates are doing. Campaigns have probably used 538 to decide where to spend effort improving their numbers. More generally, the media is partially aware of their own effects, which is why they don't report results on election day until polls have closed. But I think it would be arrogant for Silver to assume he personally affects the US and the world all that much. There are other poll aggregators who do much the same thing and would continue to do so without him.
Also, regarding being "unable to engage with red stuff," I think that's a choice, because a lot of it is out there on the public Internet and we could investigate it if we wanted to. I personally choose not to, but I am honor-bound to support curiosity, so I think people who are genuinely curious about how other people think should investigate and report back. Journalists can even talk to people and do reporting in person, which is likely to result in a better picture than looking around online.
It would be helpful if people reporting back would say more about what they saw than how they felt about it, and if we didn't dismiss their reporting if it seemed strange. It would also be helpful if they weren't drowned out by us wanting to share our opinions even though we weren't there and we didn't do any investigation at all. I don't think we have a good handle on how to moderate for that yet.
I don't know of any good sources of "intelligent or humane" conservative opinions, but I do know a few sources of interesting libertarian opinions. I've learned to be selective about sharing those links, though.
This was a very well thought-out response, and I appreciate it. Again, since I wasn’t really a Twitter user I can see how the narrative about the site could get distorted to fit someone’s agenda as well.
It is because it recommends what it thinks you want to see. It does the same thing for right wingers. Now that Musk has taken over it just seems to be shoving right wing content at everybody. I have Twitter on my phone only for notifications from one particular Twitter account, and every time I go on it tries to shove alt right bullshit (and Musk's posts, despite him being blocked) in my face.
Sounds like it's time to delete your account and not give them any more revenue.
Twitter is a piece of infrastructure that many rely on for emergency information - deleting it unfortunately isn't a possibility for some until that infrastructure is replaced.
This extension is a must for anyone still using Twitter these days: Control Panel for Twitter for Firefox and for Chrome. it removes all the algorithmic crap and makes Twitter still usable.
Unfortunately I only use it on my phone because notifications, and most of the time I don't even open it when I get those.
I don't actively use Twitter at all or have any plans to, I never visit it on my desktop. Unlike Reddit, which does have a wealth of useful info and I find it harder to get away from, Twitter holds almost nothing of value for me.
I can't really believe that anyone actually believes any "trending" sections on the web though. Do people believe the "trending" tab on Netflix is actually shows people are watching and not just shows Netflix wants them to watch?
Twitter wasn't ever trying to be a source for "unbiased" reporting. It's just a way to get news. Poker news or animation news or political news. I always felt like it was kinda up to me to pick by choosing who I followed.
I dunno, it is the application that created the concept of a hashtag. Its entire purpose is virality, not to mention any algorithm-centric feed-based application is intended to take control out of the users’ hands for better or for worse. It is only up to you to a certain degree. I don’t necessarily want to validate this article or argue about its merits but I do not think you are not some kind of pleb if you fall victim to an algorithm.
I’m basing on the idea that he’s a statistician and that I believe he probably knows the framework of his thesis is skewed. - see all the other very valid and pretty obvious holes pointed out on this thread for evidence of this - and then consider the source (Twitter) and the self-selection, manipulation, and exploitation that’s happened over years to warp the conversation. The branch he’s hanging his thesis on IMO has little to no real-world underpinning.
Edit: I don’t think he’s insincere. I think he’s framing a disingenuous argument by declaring “this is the middle!”. It’s not. It’s a veiled blue “land grab” to try and ‘both sides’ a 90-10 conversation.
I think you’re right that the foundations for making this sort of argument are shaky.
First of all, how is anyone supposed to know what “the middle” is? It’s mostly vibes-based based on what we see in our particular bubbles.
That is, with the exception of opinion polls, which are tricky to interpret since it depends on the exact wording of the question. We often don’t know what people mean when they answer a survey a certain way.
For elections, there is a ground truth. How do people vote in the end? This is why statistics works at all. Even that is subject to misinterpretation about “mandates” or whatever. We often don’t really know why people voted a certain way. Did they misunderstand something? Maybe they didn’t like the vibes? But there is a voting algorithm and returned an answer, and with a fair bit of work, some useful predictions can be made.
The thing is, I think the fact that this is all kind of vague and shaky means we should have more charity when reading other people’s opinions? It’s fine to point out that the argument might not be all that solid, and quite another to make the sort of accusations I see in this topic, which aren’t just about disagreement but about how he’s doing something terrible (“malicious”) by writing what he thinks.
The argument that he’s doing something terrible requires a lot more confidence in both your own understanding of the US (many millions of people you haven’t met in many places you haven’t visited) and what he’s privately thinking than makes sense to me.
All valid. I was pretty harsh yesterday. Part of my condemnation of Silver was coming from some half-remembered info about his work from a podcast (I think it was The Daily Zeitgeist) I listen to. I do think as other have said - he's dabbling in punditry more these days and I suspect he's more conservative than he lets on. Sometimes my frustration at the state of affairs in 2023 gets the better of me.
From Nate Silver’s blog post:
It seems pretty clear to me that this work from Nate Silver is disingenuous at best, but really more likely straight up malicious. I’ve not seen anyone even mention that the body of Twitter isn’t by it’s very nature ‘neutral’ - it was a naturally formed social body built on virality and popularity - on follows and likes.
Point being- as others alluded - there is a valid argument that Nate’s “middle ground” actually sits deep in a sea of blue. Forget indigo. Nate’s “work” here seems like smoke screen propaganda trying to cover up a reality where there are like 15 very confused moderate people in the country left, a mob of very loud hurting and often hateful GOP shills misinformed and manipulated into being used by bad actors… and then the rest of us- who know the problem is we don’t live in a representative democracy anymore. It’s been hijacked by dark and big money and there’s little to nothing we can do about it except vote at elections and watch things continue to diverge from the reality we live day-to-day.
Silver had a time when he kicked the right into action with some better use of stats. Now he’s just a propagandist.
I don’t see any evidence that Silver isn’t sincere in his opinions. Where did you get that from?
sorry - new to the surfboard app, and my reply went out-of-thread...