site banner
PaperclipPerfector  1yr ago (text post)   30843 thread views

Culture War Roundup for the week of July 3, 2023

This weekly roundup thread is intended for all culture war posts. 'Culture war' is vaguely defined, but it basically means controversial issues that fall along set tribal lines. Arguments over culture war issues generate a lot of heat and little light, and few deeply entrenched people ever change their minds. This thread is for voicing opinions and analyzing the state of the discussion while trying to optimize for light over heat.

Optimistically, we think that engaging with people you disagree with is worth your time, and so is being nice! Pessimistically, there are many dynamics that can lead discussions on Culture War topics to become unproductive. There's a human tendency to divide along tribal lines, praising your ingroup and vilifying your outgroup - and if you think you find it easy to criticize your ingroup, then it may be that your outgroup is not who you think it is. Extremists with opposing positions can feed off each other, highlighting each other's worst points to justify their own angry rhetoric, which becomes in turn a new example of bad behavior for the other side to highlight.

We would like to avoid these negative dynamics. Accordingly, we ask that you do not use this thread for waging the Culture War. Examples of waging the Culture War:

  • Shaming.

  • Attempting to 'build consensus' or enforce ideological conformity.

  • Making sweeping generalizations to vilify a group you dislike.

  • Recruiting for a cause.

  • Posting links that could be summarized as 'Boo outgroup!' Basically, if your content is 'Can you believe what Those People did this week?' then you should either refrain from posting, or do some very patient work to contextualize and/or steel-man the relevant viewpoint.

In general, you should argue to understand, not to win. This thread is not territory to be claimed by one group or another; indeed, the aim is to have many different viewpoints represented here. Thus, we also ask that you follow some guidelines:

  • Speak plainly. Avoid sarcasm and mockery. When disagreeing with someone, state your objections explicitly.

  • Be as precise and charitable as you can. Don't paraphrase unflatteringly.

  • Don't imply that someone said something they did not say, even if you think it follows from what they said.

  • Write like everyone is reading and you want them to be included in the discussion.

On an ad hoc basis, the mods will try to compile a list of the best posts/comments from the previous week, posted in Quality Contribution threads and archived at /r/TheThread. You may nominate a comment for this list by clicking on 'report' at the bottom of the post and typing 'Actually a quality contribution' as the report reason.

6
Top Bottom Old Controversial Comments
Jump in the discussion.

No email address required.

Nate Silver left FiveThirtyEight amid layoffs and Elliott Morris, ABC's new hire immediately set about ruining it. A threat he sent to conservative polling company Rasmussen Reports:

Rasmussen must explain the nature of its relationship with several right-leaning blogs and online media outlets, which have given us reason to doubt the ethical operation of the polling firm... Failure to reply, or failure to notify us of an intent to speedily reply, by the end of the day on Friday, June 30th, 2023 will be taken as a final concession of our grounds for a ban. The ban would take effect imminently thereafter.

As Nate Silver puts it, Why, unless you’re a dyed-in-the-wool left-leaning partisan, would having a “relationship with several right-leaning blogs and online media outlets” lead one to “doubt the ethical operation of the polling firm”?. I agree with Silver's overall attitude on the new direction of his company: hope ABC will stop use of 538 brand so it isn't associated with me.

Some people are probably mad at ABC for being partisan hacks but frankly that's business as usual. I'm mad because FiveThirtyEight was one of the only good analysis sites out there and these vandals are going to turn it into another factory pumping out generic progressive sludge. God damnit! 538 was the best in the business, where am I supposed to go for election forecasts now?

Amazing how consistent the pattern is these days.

Iowahawkblog said it best in 2015

  1. Identify a respected institution.

  2. kill it.

  3. gut it.

  4. wear its carcass as a skin suit, while demanding respect.

Some small news/analysis outlet will find some audience and gain traction for producing quality, mostly unbiased, and interesting/unique content, which forms the reputation on which its' appeal rests.

The outlet hits a critical mass of audience/attention, then some known lefty/prog investors buy out the brand, and in short order remake it into left-leaning opinion mouthpiece #418210, BUT they try to demand everyone treat it as just as reliable and quality as before, and maybe they even make a vague attempt to retain what made it unique in the first place.

In many cases, it dies not long thereafter.

Happened to Axios most recently, also happened to Vice. Also the Onion but I can't count them as a 'reliable' media brand.

FiveThirtyEight ALWAYS had a detectable liberal bias and yet the analysis they did, the actual numbers, at least seemed to reflect underlying reality and they weren't afraid to report conclusions that were disfavorable for Dems. My 'problem' with them was usually their intentional selection of issues to analyze that were pretty much only relevant to left-leaning readers, and the framing of everything as "we all know that [progressive opinion] is the best one, but the polls show that support for it is weaker than we'd like..."

Unless I'm misremembering, Rasmussen was one of the most accurate predictors of the 2020 election results (as reported by The Washington Post no less):

Almost all of the remaining polls — except the Rasmussen poll released Nov. 1 — overestimated support for Biden. Taken as a group, the average bias in the 2020 polls overall is -0.085, which is not statistically significant. However, these five polls’ pro-Democratic bias is statistically significant: Economist/YouGov, CNBC/Change Research, NBC News/Wall Street Journal, USC Dornslife, and Quinnipiac.

Even FiveThirtyEight itself finds them overall rather accurate. Indeed, seems like their tilt towards the GOP often counteracts whatever factor seems to make certain conservative opinions appear underrepresented.

So it seems absurd to select THAT ONE of all the options to question their reliability and literally threaten with expulsion if they don't explain themselves.

The only thing that could make this situation more farcical as a culture war issue is if the whole point of this move was to drive the value of the Rasmussen brand down so that it can be purchased by lefty/prog investors and they can pull the exact same game by converting Rasmussen itself into a prog mouthpiece.

I actually 100% believe that is at least part of the intent, if they thought they could acquire it and use it to their ends, they already would have.

Anyhow, Conquest's Second Law remains an excellent heuristic.

Most of the utility from 538 was plausible deniability for 'free thinkers' to be on the "neutral" side. They had the numbers and the data. Nate Silver is a nerdy numbers guy. I mean, just look at him. ABC are taking a mechanism to be 'safely' "neutral" and ruining it. And Nate recognizes it.

That being said, it's all a sham. Nate Silver, just like so many of his 'followers' are always looking for plausible deniability to be on the "neutral" side because not being on the "neutral" side is way too much work. And much more importantly, you can't be anything other than "neutral" if you expect to participate in 'civilized' urban society, let alone show up on TV and be respected for your work. The guy was not slightly wrong or missing a few steps in 2016, he was off on the numbers and way off with his attitude.

It's not that "conservatives" accidentally got this one right against all odds against all polling against all reality. It's that there was a manufacture process going on with an extremely biased media complex doing everything it could to push one candidate forward whilst pulling another back. If you recognized this you would have no issue predicting, at the least, a very tight race.

If you didn't recognize it you were either stupid or willfully complicit. The 2016 election cycle was genuinely insane in its bias and one should not take people who missed the insanity part of it either too seriously or too trustingly.

The goal is the pluck out your lying eyes. There can be no feedback in the system to cast any doubt what so ever on whoever the regime picks to rule next.

Yawn. Pointlessly lazy and uncharitable. If the 'regime' plans to manufacture consent or whatever for the next 'ruler' by excluding a single slightly R-biased pollster from 538 forecasts, thereby increasing Biden's predicted chances of victory by 0.1% on a site no ordinary voter reads, I suggest the regime tries a bit harder.

If there are no sites predicting an R win or even a close win when someone R leaning googles, they may be less likely to care about voting. Seems like a no-brainer to use your control over the media to make this happen by any means necessary.

This is pretty uncertain. Just as plausibly it could discourage Democratic voters by making them complacent.

Maybe, that's true. Unsure if there have been any psychological studies on this effect, would be interesting to know for sure. However, clearly elite dems subscribe to my prediction, otherwise why buy 538 and libbify it?

Insofar as there may be some discrepancy in the treatment of Republican pollsters, I think the more plausible explanation is some kind of motivated reasoning against/semi-conscious suspicion of right-wing organisations on the part of the 538 team rather than some attempt to fix forecasts. You see the converse happen on RCP sometimes.

why buy 538 and libbify it?

This is just silly. ABC bought 538 to change forecasts by 0.1% and therefore somehow deliver elections to Democrats? No-one thinks that conspiratorially in real life.

I do, seems like common sense. Why do you think they wouldn't take an action that helps them? The action not instantly guaranteeing them a win doesn't mean its not worth taking. It's one action among many

It's hardly common sense; the business of business is business, and it's seems a reasonable assumption to suppose that, in the absence of convincing evidence to the contrary, ABC's use of 538 is ultimately driven by business incentives.

This might become my go to example of cancel culture:

Besides, the proof is in the fucking pudding! Unlike most people in politics or journalism, pollsters have a track record that can be tested against actual results. They can be held accountable for political bias that might render their polls less accurate. Rasmussen has indeed had strongly Republican-leaning results relative to the consensus for many years. Despite that strong Republican house effect, however, they’ve had roughly average accuracy overall because polls have considerably understated Republican performance in several recent elections (2014, 2016, 2020).

Even when you can track and mitigate the "bias", the wrong thinkers must be purged.

The nutty thing is that, per 538's own data, their bias isn't even that egregious! While they definitely have a Republican tilt, that tilt comes out of them having misses on BOTH sides of the column but just having MORE misses in the GOP's favor.

https://projects.fivethirtyeight.com/pollster-ratings/rasmussen-reports/

They otherwise have exactly the same kind of shape you would expect a decent pollster to have: Some outliers in both directions with a big mass of near misses in the middle.

Bias would best be measured against ground truth if you have it, not the "consensus", so Silvers's statement above means that Rasmussen is only about as Republican-biased as the typical pollster is Democratic-biased, and thus singling out Rasmussen is even more partisan.

If you read the substack, what is funny is some people are claiming the Republican leaning pollsters are deliberately trying to change the outcome due to polling whereas the democrat leaning pollsters just make honest mistakes.

It is funny how each partisan group genuinely believes that all of their guys are just stupid but all of the other guys are just evil.

It's another victory for conflict theory. Politics is every bit as tribal as your standard Basketball game. Everything the other team does is a foul. Every foul that gets called on your team is a rigged conspiracy against you. And then there's the trappings of the mundane reality somewhere, hanging around.

Although in politics you could make the argument that conspiracies abound. I think William Blum had a good political heuristic:

... Once is an accident; twice is a coincidence; three times is a conspiracy...

... I believe in conspiracies. So do all of you. American and world history are full of conspiracies. Watergate was a conspiracy. The cover-up of Watergate was a conspiracy. So was Enron. And Iran–Contra. The October Surprise really took place. For a full year, George W. Bush and Dick Cheney conspired to invade Iraq while continually denying that they had made any such decision. The Japanese conspired to attack Pearl Harbor while negotiating with Washington to find peaceful solutions to the issues separating the two governments. There are numerous people sitting in prison at this very moment in the United States for having been convicted of ‘conspiracy’ to commit this or that crime...

"I make accidental errors, they make stupid mistakes, you try to influence outcomes."

Crikey, my existing view of 538 commentary (outside of their polls) is that they possibly could be more progressive if they tried, but that they would have to try very hard.

If the new regime is "You are conservative? To the gulag!" then wow. Nate got out at the right time (I think he's a lefty himself, but at least he can produce good data even if he then has to write a post about why it's terrible the Bad People are slightly getting an advantage in Woollybully, ND).

... he can produce good data even if he then has to write a post about why it's terrible the Bad People are slightly getting an advantage in Woollybully, ND

You can only hide reality so long before you're condemned by reason to insanity to admit what the data indicates over time. I'll give this one to 538, but this doesn't score any integrity points for them on their side of the ledger, IMO.

"You are conservative? To the gulag!"

What are you havering on about. The controversy here is not that Elliot Morris is threatening to exclude Rasmussen because conservatism is bad, rather the exclusion is being considered on the grounds that Rasmussen is a partisan agency that produces Republican-friendly polling deliberately. This is still a silly thing to do for the reasons Nate points out, but the reason their close affiliation with conservative organisations is somewhat incriminating, or at least suspicious, in the eyes of Morris, is that it calls into question the motivation behind their polling.

And who decides Rasmussen is partisan while the left-leaning, Democrat-supporting polling groups are fair and impartial? That's the problem here. Maybe Rasmussen is partisan, but this makes it look like "if you produce any results that are Republican-friendly, we will consider you partisan". That isn't much use if the results on particular questions really are friendly to Republican views because the general public is switching to a more conservative view of a topic. It further means if you want not to be tagged as "unclean", you better massage your data not to get Republican-friendly results.

How do you think that is going to help with finding out the true attitudes on the ground? "Why won't anyone vote for our 'teach three year olds to be furries' initiative, it polled so well! Truly a mystery!"

You are wildly overestimating the effect of pollster 'house effects' in both directions. Even PPP, which is explicitly affiliated with the Democratic party, has overrated Democratic candidates by an average of just 0.9% - and to be fair Rasmussen is not much higher than that in the opposite direction. Of course such fine margins do matter in elections, but when you say something like this you're giving a bit of a false impression;

That isn't much use if the results on particular questions really are friendly to Republican views because the general public is switching to a more conservative view of a topic

Polling 'bias' probably isn't turning thumping conservative majorities in issue polling into close runs or liberal majorities, where they exist they are very slightly shifting the scale of the majority or in a very, very close cases they might tip the balance. In any case, we can rest easy because the big-name 'establishment' pollsters publish issue polls favourable to Republicans all the time, so clearly they aren't afraid to do so.

Upthread, people are saying that Rasmussen is as partisan as the average polling company, but it's the only one that's partisan for Republicans.

If you accept that claim, then you'd expect to see a story like:

The controversy here is not that Elliot Morris is threatening to exclude Nessumsar because progressivism is bad, rather the exclusion is being considered on the grounds that Nessumsar is a partisan agency that produces Democrat-friendly polling deliberately. This is still a silly thing to do for the reasons Nate points out, but the reason their close affiliation with progressive organisations is somewhat incriminating, or at least suspicious, in the eyes of Morris, is that it calls into question the motivation behind their polling.

several times over, for the companies with the same amount (but opposite valence) of partisanship.

The partisan lean per se was not the cause of the letter. After all, there are other R-biased pollsters he didn't go after. Harris for instance has a similar average overestimate of Rs to Rasmussen. The crucial point was their close relationship to explicitly Republican/conservative outlets and institution.

The crucial point was their close relationship to explicitly Republican/conservative outlets and institution.

I'll repeat myself: would a close relationship to explicitly Democrat/progressive outlets and institutions trigger the same response? I don't think so, and neither does Nate Silver:

If you wanted to argue that pollsters and other people working in public opinion should be more austere about disclosing their political opinions, I could maybe get on board with that. But if that were the basis for banning pollsters, you’d have to wipe out dozens of pollsters from the FiveThirtyEight ledger. (And frankly, it’s not the sort of philosophy that Morris tends to espouse.) I guess I can’t know for sure, but I strongly doubt that Morris would view being friendly with left-wing political figures as something that might “cloud their judgement in the operation of a public poll”.

https://natesilver.substack.com/p/polling-averages-shouldnt-be-political

I think this is a quite different claim. What I suspect is the case here is that the undoubtedly Democratic views of the 538 team lead them, consciously or perhaps more likely not, to be more suspicious of right-leaning pollsters. Which is to say that they genuinely do think Rasmussen is guilty of uniquely bad practice, but perhaps yes, their ideological dispositions brought them to that conclusion. So I don't think Morris considers it targeting of Conservatives per se, he just sincerely thinks that Rasmussen are unusually unscrupulous pollsters.

Stuff like this is why I'm longterm bullish on platforms like Substack. People ultimately don't care about 538 as an institution. They mostly just cared about Nate Silver's takes. I suspect this is true across many other media orgs. There's typically a few voices who actually matter and the rest are background noise. Paying people directly for their insights is ultimately a better model than subsidising a whole team indirectly, whom you are mostly not interested in.

It's a bit like why cable started to get a lot less interesting to people.

The thing about Substack is the majority of its successful writers are people who became well known (mostly through legacy media) before they joined substack - Greenwald, Sullivan etc.

The ‘discovery mechanism’ for Substack is essentially Twitter (and, to some extent, the blogroll of other Substack users, which still usually comes down to Twitter). Very different to a publication with many writers where you browse by article title or genre/section.

It’s almost the difference between coming here and every regular having their own blog. This way is more efficient.

That could be visibility that legacy media gave them but I’d make a different argument. That people needed the training of doing good reporter and learning how to work a beat and research. Seeing how the grey haired guy works sources and the process of research helps to develop talent. Substack isn’t going to develop guys but provides the opportunity for people to avoid editor bias and the suits business plan. Substack probably won’t ever being any good at training writers.

But neither are major newspapers. They found it is nore advantageous from a pecuniary perspective to just hire a bunch of 20 or early 30 somethings to write from Brooklyn. Doing hard hitting investigative reporting is hard; doing a report on the latest thing the president said while heavily mixing in your opinion is cheap and easy.

Substack created it's own Twitter-like feed a while back, and had links to them banned in response. You can use it to discover new writers, writers can interact with each other thus boosting each other's visibility etc. Things are still pretty slow, but if they reach critical mass, that's all they need for discoverability. I can also tell you that my response to Twitter login wall wasn't "I guess I'll install Twitter then", it was "I guess I'll install Substack then".

They used to let you see posts people liked by looking at their profiles, and I kept finding new substacks that way. It's a good bit more effort for me to discover good ones now.

A lot of what people cared about 538 for is the use of data and models far beyond what you see anywhere else (maybe not explicitly, but at least that people might have some idea of what's going on beyond idle speculation), and at least currently, 538 is still the best place for any of that that I know of.

Edit: just read further down that Nate Silver is the one with the rights to the models—guess 538 being the best may no longer be the case.

Nate says in his footnote that Morris wants to move 538 in a more explicitly progressive direction, so I expect more woke advocates like Clare Malone to be the norm. Without Nate's models or his push for data-driven rigor, the site will likely devolve into having an occasional statistical chart here and there, but otherwise being functionally indistinguishable from Vox. A sad day indeed.

Other than Nate they were pretty much all Vox-wannabees anyways -- so yeah, pour one out.

Harry Enten and Nathanial Rakich were and are pretty good respectively, so not all of them were woke activists. But they'll probably be the last of their kind at the site.

If that was my goal then writing this letter makes sense knowing that it would be leaked. Helps them boost their profile in progressive spaces.

That being said that seems like the most crowded market position in media and one of the worst business plans you could come up with to make money. 538 had a brand of quality. I feel like their the type of place modern neolibs like that’s data oriented etc. But market positioning yourself to be progressive media and out progressiving the progressives seems like an incredible crowded spot.

Seems fucking stupid to me, a god damn socialist.

Nate was, of course, a LIBERAL (spits, makes sign of the cross) and had some takes I thought were pretty dumb, but he is the guy at 538. It's just him. The whole point was that his models beat random chance and beat the average at doing something purely empirical: using poles to predict outcomes. Why bother even having the outfit at that point?

Also, why bother sending a letter at all? Pointlessly demonstrative. If I wanted to punish Rasmussen from Morris's position, I'd just make consistent noise about "Rasmussen is so biased we actually give them negative weight in out model" or some shit.

Nate was, of course, a LIBERAL (spits, makes sign of the cross)

I laughed thanks.

How the hell are you supposed to produce good results if you're deliberately cutting yourself off from data about the views, beliefs, actions and behaviour of Those People Over There? You can think they're all rich white cis het Christian old guys who want to eat the poor, burn gay trans non-binary furries at the stake, and chain women up in breeding camps to pump out babies, and still need to incorporate that data into your forecast about a national election.

What is gained by becoming Site Number 9658742 on the Innertubes that rants about the Rethuglicans?

Also, why bother sending a letter at all? Pointlessly demonstrative.

I wonder. Maybe the message isn't for Rasmussen as such, it's to demonstrate to the people on the right side of history that no, we got with the programme, we're not even going to give the Fascists that much toleration by letting them contribute to our site. We are purging the wrongthinkers and demonstrating our purity.

How the hell are you supposed to produce good results if you're deliberately cutting yourself off from data about the views, beliefs, actions and behaviour of Those People Over There?

Parsing polls to learn how to win elections for the team is at most a secondary goal among most people: to win a purity spiral among your peers is far more immediate and practical. And why shouldn't it, to be honest? If you somehow knew the exact outcome of an election a year in advance, you're not going to be able to do anything to change it. Even skilled politicians with top pollsters usually just flail against the tide of shifting poll numbers.

Rasmussen does lean right but 538 has applied 'House effects' to various pollsters to deal with right/left bias for ages. The posture of 'banning' biased pollsters rather adjusting for bias is a bad omen.

I think this is an also an instance where actual expertise is expensive and replacement level biased writers are cheap. Part of the reason Nate Silver is out is because he had a big contract and 538 itself didn't make money it was purchased as a prestige booster for ABC news.

Or a good omen. People doing this kind of stupid things are usually worried.

I mean I think what gets me about it is Nate silver and his team were liberals, they didn’t try to pretend they weren’t, but they felt like liberals who were trying to get at an accurate reflection of reality. Sure, they had biases. But for all their faults they came reasonably close to an accurate reflection of what was actually happening and they seemed like some people I could sit down and have a reasonable discussion with, despite our many disagreements.

Then ABC gets rid of him and turns his brand into partisan hacks because it’s not enough to be reasonable liberals seeking the truth.

Was it ever really a “team” though? A browse through Wikipedia makes it sound like Fivethirtyeight was Nate’s personal blog at first. It was only after he signed with the New York Times that “writers” started showing up. It was comical just how uninterested Nate seemed in the partisan hackery, even when it constituted the bulk of “his” website.

It was originally a series of diaries on the orange site. No, not that one. DailyKos.

538 was the best in the business, where am I supposed to go for election forecasts now?

Whatever new brand Nate Silver comes up with? He's got the rights to the actual valuable part of 538.

ABC News is scrambling to find a replacement for Nate Silver’s election forecasting model, which the 538 founder is taking with him when he leaves the news organization at the end of his contract. According to two sources with knowledge of the situation, ABC executives were slow to realize that the company only owned some of the models that 538 used to forecast major elections, and that many were on rent from Silver as part of the initial deal to bring the data journalism site to ESPN ten years ago.

“They have put very very very little bandwidth into managing 538, and they seem pretty clueless about who-owns-what IP questions,” one person familiar told Semafor.

If true, Silver is smarter than I thought. Well played Nate.

Nate Silver has been one of my favorite commentators for a long time, since well before he was primarily in the politics business due to his clear writing on baseball analytics. Every time he comes up in a controversy of some sort, I'm reminded of why I'm a fan - whether I agree with him or not, I think he really, truly does his best to get things right via careful, non-partisan analysis. Silver has been incredibly consistent on how to use pollsters predictively, part of which is including pollsters that have known house effects and simply correcting for that when incorporating them into the analysis.

As a result of his insistence on avoiding partisan hackery, Silver takes a ton of shit from people on Twitter and other elements of the commentariat that are just less competent at actually analyzing things than him. Amusingly, this used to come largely from the right-wing, who kept making fun of his model for giving Trump a roughly 30% chance to win the 2016 election, because apparently grasping that 2:1 underdogs win pretty often is basically impossible for some people. That Silver is now more controversial on the left than the right is another example of what I view as American progressives dissociating themselves from ground truth, with that phenomenon accelerating aggressively with Covid, Summer of Floyd, and gender ideology.

"Amusingly, this used to come largely from the right-wing, who kept making fun of his model for giving Trump a roughly 30% chance to win the 2016 election, because apparently grasping that 2:1 underdogs win pretty often is basically impossible for some people."

I'm going to push back on this as a mis-recollection of the actual facts.

Trump's rise badly damaged Silver's credibility, but its wasn't Trump's general election win, it was his GOP nomination.

Examples that aged poorly:

https://fivethirtyeight.com/features/donald-trump-is-winning-the-polls-and-losing-the-nomination/

https://fivethirtyeight.com/features/donald-trumps-six-stages-of-doom/

To his credit, Silver has largely fessed up to screwing this up:

https://fivethirtyeight.com/features/how-i-acted-like-a-pundit-and-screwed-up-on-donald-trump/

Another article delving into the details of this:

https://slate.com/news-and-politics/2016/01/nate-silver-said-donald-trump-had-no-shot-where-did-he-go-wrong.html

All in all, I'm a Silver fan, in the grand scheme of things, I think he does a pretty good job, but the Trump nomination screw up showed that he's not immune to certain biases.

Amusingly, this used to come largely from the right-wing, who kept making fun of his model for giving Trump a roughly 30% chance to win the 2016 election, because apparently grasping that 2:1 underdogs win pretty often is basically impossible for some people.

Oh, not this again. The 30% chance was from polls right before the election, the thing people had with 538 was how polls changed completely from day-to-day. On October 17th 2016 the 538 predicted Hillary with 88.1% after it was basically 50:50 on July 30th, then two weeks later on November 4th only it dropped 24 percentage points down to 64.5% for Hillary. The point is that Silver does not have the prediction, he has series of dozens of predictions that swing wildly so it is hard to pinpoint if he was "right" as there is many possible definitions of that.

I stand by my claim that polls in general are more then useless, it is akin to using aggregator to predict weather on a particular day 2 years from now somewhere. You either use absolutely obvious and thus useless take (it will be warm because it will be in the middle of summer - California will remain blue), but you will not know the specifics of events close to the day, such as specific front forming in Arctic or whatnot and thus your predictions are useless until the very day. Silver's polls are useless to watch prior to election because they are for sure to be subject to wild swings so they are not really predictions in that sense, and they are useless to watch on election day because you will have results anyway very soon without needing to then put forth lame defense of how often 1:2 or 1:10 or 1:100 event happens.

Of course it is perfect for Silver's business, we have elections only every few years so it can take decades of data to prove that he is actually full of shit. By that time he will be multimillionaire, in that sense I say well played.

538 is pretty well calibrated even accounting for the number of updates they make.

Let's see your calibration plot if you think this is what "full of shit" looks like.

Look at their politics calibration, it does not look nearly as well as sports one. It is also because of this:

When calculating the calibration and skill scores for forecasts that we updated over time, such as election forecasts that we updated every day, we weighted each update by the inverse of the number of updates issued. That way, we counted each forecasted event equally, regardless of how many updates we issued to the forecast.

Nevertheless at minimum stop spreading stupidity like "538 predicted 30% chance of Trump winning". They made 150+ predictions with wildly different number assigned. Having articles that "Today 538 predicts politician X winning with 90% probability" means nothing, in a week there may be a new prediction reassigning it to 50%. At best you may gauge who is favorite and who is not, but one does not have to be an expert for that.

The House predictions (which have the most data points) are also well calibrated.

Again, if it's bullshit, it's trivial to do better. Post chart.

I don't give a shit about "good calibration" in this context. This is endless debate also with Scott's prediction. What I am interested in is usefulness of the info 538 provides. Look at the daily graph of predictions for 2016 elections and please tell me what it is good for. You are somebody who finds 538 useful, what is it that you are getting from them?

I basically never check 538 because I don't really care who wins federal office. My point is merely that their predictions are better than "bullshit" by virtue of their calibration. If you don't give a shit about calibration then what's the basis of your opinion? Again, recall that the daily swings are factored into the calibration.

Of course it is perfect for Silver's business, we have elections only every few years so it can take decades of data to prove that he is actually full of shit

538 doesn't just do Presidential races, it does all Congressional races too, and other bits here and there like primaries. They did a short post about the historical performance of their forecasts (sports and politics) a while ago and the overall picture is that they do reasonably well.

https://projects.fivethirtyeight.com/checking-our-work/

I like the idea of what he was trying to do with polling, though I wish it were done for more things along the same lines. I’ve never seen anyone attempt to predict the outcome of business or technology or social events using polling data, and it seems like it would be interesting to try. Even stuff like whether a theory is likely to prove true, I think the idea of simply using the data published in the mainstream to try to publicly predict trends before they happen on a blog site would probably be an interesting read even beyond just politics.

I'm not sure how much it makes sense to use polling data to predict other kinds of events. My understanding is that using polling to predict elections makes sense because you're essentially running the election early on a sample, so going from there to estimating the election results is what statistics is good at. And there are a lot of polls on elections people care about so you have enough data to do something with.

The other really useful polling is economic indicators, because so many rely on the confidence of large groups of people. If small business owners collectively think bad times are coming and stop investing, bad times will come. If consumers are collectively optimistic, they will spend money, and times will be good.

Except for those situations you have other actors who can change because of the collective belief. If business owners think bad times are coming then the federal reserve sees the polling sentiment and prints a bunch of money to prevent bad times. The economy is now an area where that doesn’t play out before policy makers play a hand too. You need unexpected events to cause bad times now.

But does the Federal Reserve engaging in rate-adjusting and QE/T policies act by economic force, or does it act by publicly stating that its actions will have such-and-such effect which will then impact expectations? A lot more of the latter. Interest rates rise, smart people "know" that means a recession is coming, money managers look at each other and say "you can't fight the Fed," investment pulls back, and bingo a recession.

That hasn’t happened in 16 years. The game has changed. The fed pumps more money into the system then people have more money and hence spend more. They’ve canceled the thing you speak of.

Business confidence Index is already a thing. Don't know how reliable it is as a predictor, however.

I think this and basically the stuff like whether or not people are really optimistic about new technology or whether experts in afield think an event is likely or not.

They’ve been decaying into a progressive mouthpiece for years now. Clare Malone has been a frothing leftist scold from the jump, and once they brought on Perry Bacon Jr. as their “here’s why anti-black racism explains every single political trend in America” guy, their analysis stopped being even remotely useful. After the 2016 election it was clear to me that everyone on that staff other than Silver was determined to atone for why they didn’t do more to stop Trump. I haven’t cared about them or paid attention to them in ages, and I’m not remotely sad to see them crumble to dust.

Yeah, there's a solid wokist tilt to a lot of their writers, although some like Harry Enten and now Nathanial Rakich were pretty solid. Still obviously left-leaning, but more Silver-esque in trying to use data to arrive at good conclusions.

I concur. I used to go there regularly, but I dropped off some time after 2016 because it went from somewhat left to extremely left. That's also when Leah Libresco stopped writing for them.

After the 2016 election it was clear to me that everyone on that staff other than Silver was determined to atone for why they didn’t do more to stop Trump.

We really should keep a tally of how many organizations Trump ruined.

He might end up owning the libs purely via an immune system overreaction.

Obviously this was a while in coming, but most of their non-opinion journalism still seemed aimed at getting to an accurate depiction of reality.

I don't know if "most" is a good enough standard. Try that with any other profession:

  • "Most surgeons cut into people in an attempt to heal them (but some do it for the opposite reason)"

  • "Most engineers build bridges in an attempt to safely convey traffic (but some do it for the opposite reason)"

  • "Most accountants manage accounts in an attempt to help their clients' finances (but some do it for the opposite reason)"

  • "Most journalists write articles in an attempt to inform the public (but some do it for the opposite reason)"

A few bad apples literally do spoil the bunch, because professional fields live and die on their reputations. Laypeople simply don't have the expertise to judge the choices that a professional makes (at the object level, before the fact), so all that's left is the industry's reputation.

What is this place?

This website is a place for people who want to move past shady thinking and test their ideas in a court of people who don't all share the same biases. Our goal is to optimize for light, not heat; this is a group effort, and all commentators are asked to do their part.

The weekly Culture War threads host the most controversial topics and are the most visible aspect of The Motte. However, many other topics are appropriate here. We encourage people to post anything related to science, politics, or philosophy; if in doubt, post!

Check out The Vault for an archive of old quality posts. You are encouraged to crosspost these elsewhere.


Why are you called The Motte?

A motte is a stone keep on a raised earthwork common in early medieval fortifications. More pertinently, it's an element in a rhetorical move called a "Motte-and-Bailey", originally identified by philosopher Nicholas Shackel. It describes the tendency in discourse for people to move from a controversial but high value claim to a defensible but less exciting one upon any resistance to the former. He likens this to the medieval fortification, where a desirable land (the bailey) is abandoned when in danger for the more easily defended motte. In Shackel's words, "The Motte represents the defensible but undesired propositions to which one retreats when hard pressed."

On The Motte, always attempt to remain inside your defensible territory, even if you are not being pressed.


New post guidelines

If you're posting something that isn't related to the culture war, we encourage you to post a thread for it. A submission statement is highly appreciated, but isn't necessary for text posts or links to largely-text posts such as blogs or news articles; if we're unsure of the value of your post, we might remove it until you add a submission statement. A submission statement is required for non-text sources (videos, podcasts, images).

Culture war posts go in the culture war thread; all links must either include a submission statement or significant commentary. Bare links without those will be removed.

If in doubt, please post it!


Rules


Recommended Posts And Communities

Recommended Realtime Chats