Nate Silver on the Demise of FiveThirtyEight (natesilver.net) 85
FiveThirtyEight founder Nate Silver, on the site's demise: Last night, as President Trump delivered his State of the Union address, the Wall Street Journal reported that ABC News would lay off the remaining staff at 538 as part of broader cuts within corporate parent Disney. Having been through several rounds of this before, including two years ago when the staff was cut by more than half and my tenure expired too, I know it's a brutal process for everyone involved. It's also tough being in a business while having a constant anvil over your head, as we had in pretty much every odd-numbered (non-election) year from 2017 onward at 538/FiveThirtyEight. I don't know all of the staffers from the most recent iteration of the site, but the ones I have met or who I overlapped with are all extremely conscientious and hard-working people and were often forced to work double-duty as jobs were cut but frequently not replaced. My heart goes out to them, and I'm happy to provide recommendations for people I worked with there.
[...] The basic issue is that Disney was never particularly interested in running FiveThirtyEight as a business, even though I think it could have been a good business. Although they were generous in maintaining the site for so long and almost never interfered in our editorial process, the sort of muscle memory a media property builds early in its tenure tends to stick. We had an incredibly talented editorial staff, but we never had enough "product" people or strategy people to help the business grow and sustain itself. It's always an uphill battle under those conditions, particularly when it comes to recruiting and retaining staff, who were constantly being poached by outlets like the New York Times and the Washington Post.
[...] The basic issue is that Disney was never particularly interested in running FiveThirtyEight as a business, even though I think it could have been a good business. Although they were generous in maintaining the site for so long and almost never interfered in our editorial process, the sort of muscle memory a media property builds early in its tenure tends to stick. We had an incredibly talented editorial staff, but we never had enough "product" people or strategy people to help the business grow and sustain itself. It's always an uphill battle under those conditions, particularly when it comes to recruiting and retaining staff, who were constantly being poached by outlets like the New York Times and the Washington Post.
Polling Averages (Score:5, Interesting)
Re: (Score:3, Interesting)
Congratulations. You are an idiot.
Re: (Score:2)
You guys have learned nothing I see. Never change!!!
Re: (Score:2, Insightful)
I seriously doubt we'll see anything as overt as the US military occupying blue and swing states.
It's more likely we'll see a continuation of what happens now: gerrymandering and voter suppression.
Whatever side you're on, go ahead and work for your side to win. But above all, work for democracy. Respect the truth and the law.
Re: (Score:2)
That's not necessary, and certainly not desirable. Insurrection sounds great, and plays into the hands of those that will dominate you. There is violence, death, and not much improves.
You simply need a way to take the money out of politics.
The 538 erred many times, and Nate was part of the problem. His bad forecasting changed the face of the US. Voters need to arrive, cast their votes, despite obstacles placed in their path. The margins of success for bad outcomes is very small because of apathy. The apathy
Re: (Score:2)
You simply need a way to take the money out of politics.
You can't. The organizations from whom you must purchase political visibility for your ideas (including that one) will never give you the coverage you need to promote it.
Re: (Score:2)
If you remove money from politics, you also allow the debate to shift to what's actually important-- away from a shopping center of ideas. Consensus building provided by democracy can easily work without bribery and corruption incentives from campaign contributions, political party endowments, and more.
Think outside the current box. We've become a Let's Make A Deal society where money is the incentive for success. Imagine a world where outcomes are less dependent on money, and more on sustainable decision m
Re: (Score:2)
If you remove money from politics
If you removed the cartel leaders, drug use would vanish. Yeah. Right.
No. People are trained to receive their ideologies from major media outlets [wikipedia.org]. Remove one and others will step in to fill the vacuum. In the short term, people will wander aimlessly through the political landscape feeling lost. Witness the late decision by the Washington Post not to endorse a candidate. The liberals were left without their guidance and they were feeling like lost lambs as a result.
Re: (Score:2)
They're actually-- in reality-- shrunken into social media silos. They have very common needs, virtues, aspirations, and hopes.
All of that is grouped by how the other silos are doing deceitful things that prevent aspirations and hope, through non-virtuous behavior.
Media is unable to provide objective, yet compelling information to make decisions upon, and so the conscious and unconscious bias-side vortexes inside each silo both contains it, and feeds on its energy.
Some of that energy is fueled by fear and g
Re: (Score:2)
That won't turn the blue states red. They'll double down on their constitutional right to run elections as they see fit. You'll need an "insurrection" and then you can march the troops in and occupy the halls of power. Believe me, Trump's minions are pouring through the Reconstruction era to prepare for just such an eventuality.
You may have a point. I stress may. Trump has surprised us over and over again. We think he's kidding about something, and then he does it.
I'm a Canadian ex-pat permanent resident. I wonder at the back of my mind whether Trump is serious about trying to make Canada the 51st state. And I have nightmares about what might become of it.
Re: Polling Averages (Score:2)
Re: (Score:2)
While I am also concerned by the 51st state rhetoric, that seems less likely than a much more problematic possibility. Recently Trump has been telling Canadian officials that previous treaties defining the US/Canada border are invalid. My concern is that through executive action he will redefine the border to annex portions of Canada, similarly to what Russia has done in Ukraine. After the stupidity with renaming the Gulf of Mexico and various mountains, it seems shockingly possible.
Annexing part of Canada would (still) be an act of war. It would trigger Article 5 of the NATO treaty, and surely result in the USA being expelled from NATO, if it hadn't left on its own already. Suddenly the USA would be at war with most of the Western world. Maybe Trump would wager that Europe would be busy pushing Russia out of Ukraine?
I'm certainly hoping that times don't get more interesting than they are already.
Re: (Score:2)
While I am also concerned by the 51st state rhetoric, that seems less likely than a much more problematic possibility. Recently Trump has been telling Canadian officials that previous treaties defining the US/Canada border are invalid. My concern is that through executive action he will redefine the border to annex portions of Canada, similarly to what Russia has done in Ukraine. After the stupidity with renaming the Gulf of Mexico and various mountains, it seems shockingly possible.
Annexing part of Canada would (still) be an act of war. It would trigger Article 5 of the NATO treaty, and surely result in the USA being expelled from NATO, if it hadn't left on its own already.
Congress passed legislation during Biden's term that prevents the president from withdrawing from NATO without Congressional action, which seems very unlikely to succeed. Maybe this is plan B to get out of NATO...
In a sane world, of course, a US president doing anything of this sort would trigger instant impeachment and conviction, but the GOP has abandoned any pretense of holding any principled positions and just goes along with whatever Trump wants, so I wouldn't hold my breath.
Re: (Score:2)
In a sane world, of course, a US president doing anything of this sort would trigger instant impeachment and conviction, but the GOP has abandoned any pretense of holding any principled positions and just goes along with whatever Trump wants
It's worse than that. When he started talking about this expansionist bullshit during his joint address, they stood and clapped for it.
They're not only not going to exercise their Constitutional powers of oversight and sole responsibility to declare war. They're actively complicit in it due to having their spines surgically removed as part of the primary election process. They have abdicated their responsibilities to the Constitutional order, in service to a lying shitbag grifter who is intent on giving
Re: (Score:2)
Re: (Score:2)
The way it usually works is the great powers get together and partition the world. Russia gets eastern Europe, which will keep Europe too busy to help us. China gets Taiwan and control of SE Asia, America gets the Americas.
Even as things are, we're getting very little support from the rest of the world. Even our natural allies, the Commonwealth aren't speaking up but rather bowing down to Trump.
Re: (Score:2)
Suddenly the USA would be at war with most of the Western world
Don't threaten his boss in Moscow with exactly what he wants!
Re: (Score:2)
No chance of Canada becoming the 51st State as we would then have the vote. We'd be a territory like Puerto Rico. A lot of people up here don't understand this, we might not even get citizenship as a territory, just American residency and being a territory, the American Federal government would own us.
Re: (Score:2)
I don't know why anyone would take something that a sitting head-of-state says as "oh he's only kidding."
He is telling you exactly who he is, and what he wants. Believe him.
This bullshit is how he got elected to begin with. The whole "oh he's just trolling, he's not really going to fire hundreds of thousands of employees or put tariffs on Canada and Mexico" - how's that working out?
Believe them when they tell you who they are, and what they want to do.
Re: Polling Averages (Score:2)
Re: (Score:2)
Ah, see there's the real problem. If they did that, "their side" would never be in government again. You can see how that present a problem for them.
Re: (Score:2)
Re: (Score:2)
It's astonishing that anyone doubts that Trump's team is reviewing the Insurrection Act and Reconstruction-era Federal policies as we speak.
Re: (Score:2)
I don't think polling is going to mean very much in the United States going forward. The Insurrection Act is just waiting in the wings for the blue states, and swing states will get the message; choice is irrelevant, you will vote Republican or face the US military occupying your state to restore "Republican" government.
Congratulations on the censor moderation. You hit the right nerves of the nerveless cowards.
But what's your concern? You're already safely out of the blast range.
Peripheral thoughts about dead heroes versus live cowards. When a REAL test of courage happens, most of the survivors are the cowards. The real heroes usually wind up dead in real life tests, not to be confused with the movies. Bravery != bravado. (And yes, I know there are outliers. An extremely lucky hero might survive and some unlucky cowards ma
Re: (Score:1)
I'm in Canada. We're in Trump's sights. It will be harder to engineer a crisis up here, and perhaps Canada's only salvation is that he will need to use the now largely unconstrained might of the Executive branch to crush Blue states, which will probably lead to significant disorder. But make no mistake, he and his MAGA cronies are trying to restart the Monroe Doctrine. I'll wager we'll get a few Fenian raids north of the border by the more deranged of his supporters.
Re: (Score:2)
I would rather have my King, universal health care and a somewhat diminished quality of life than the vile political system that a pack of 18th centuries idealists with too much affinity for Aristotle and Montesquieu thrust upon the world.
Your country sucks, and is sucking worse every day. And we're no major entry point for criminals. The US has no problem producing criminals all on its own, heck they can even be elected President. Most of our gun crime comes from your country.
Now do fuck off.
Re: (Score:2)
AC sock puppets, I presume? But when you feed them they do NOT go away and they are already fscked off in the head.
Minor item on the list of improvements for Slashdot would be an option to turn off AC comments. I don't think that is too much to ask for in the "private" space of a journal, though I can see it as more difficult to implement fairly for the more public areas. Slashdot already offers the option to make them invisible in most cases (and it is actually not a bad compromise that the accident of AC
Re: (Score:2)
"And we're no major entry point for criminals"
Wasn't an Isis member recently found living in Canada with Canadian citizenship? Not to mention the vast criminal car theft rings...
Re: (Score:2)
> It will be harder to engineer a crisis up here,
There are already at least 2 Canadian provinces that have a very real possibility of seceding from the "post-national" state in the near future: Quebec and Alberta.
Ironically much more than say California leaving the US.
But somehow it's "harder" to engineer a crisis there?
Re: Polling Averages (Score:2)
Re: (Score:3)
Their "Atlas of Redistricting" [fivethirtyeight.com] is a really nice visual tool as well that makes an important but pretty dry topic interesting to wrap your head around the alternatives.
Re: (Score:3)
The site has really gone downhill since Nate Silver left. However, I am going to miss their polling averages. [fivethirtyeight.com] Outlier polls make news, but the average tells the real story.
I'll be sorry to see it go. It was one of the very few sites that tried to simply look at the data, without biasing what they see based on a political agenda telling them what they want to see.
Re: (Score:1)
Did you just say that data is irrelevant to capitalism?
Re: (Score:2)
Did you just say that data is irrelevant to capitalism?
No, did you?
Re: (Score:2)
>The site has really gone downhill since Nate Silver left.
I never found anything on there not written by Silver that was worth the effort to read.
I tried a few, but they were painful, and I don't think I completed any.
(ok, so I have a Ph.D. in Economics & Statistics, and know what they're talking about).
What's fascinating right now is the nitwits (one in the WSJ this morning, even) who think that he screwed up by calculating a 70% chance for Hillary in 2016. Uhm, that's roughly 2/3. When your chan
Re: (Score:2)
Agreed, I never understood why they took so much heat for that prediction. 70% does not equal 100% and it's not like it wasn't super close. Hilary even got the popular vote.
Re:Nate trying to call polling a "real business".. (Score:5, Informative)
Nate Silver was never a pollster. Nate Silver is a statistician. He started out with predictions about sport, especially league sports. The methods he learned there he used also for predictions of things like the Oscars or election results. He raised to fame when his predictions went against the common wisdom that Barack Obama was toast in his quest for re-election, because he noticed that election results in many states are correlated, that means, changes in polling results in one state are often mirrored in neighboring states. This has nothing to do with opinion, it's all about statistics.
His website is also the only one that in 2016 in the last update before the presidential election titled "Donald Trump is just a polling error away from the presidency". And that's how it went. The polling error for the average presidential election is about 3%, larger than the lead Hillary Clinton had in the last polling averages as published by 538, and so Nate Silver gave Donald Trump a 30% chance of winning the election, and a 30% chance is pretty good.
Whatever you think about Nate Silver, I know that your opinion is based on wrong information.
Margin of error [Re:Nate trying to call pollin...] (Score:4, Interesting)
Yes. If he had just stayed rigorous with the way statistics works in the sciences, their final prediction should have been "2016 election result is too close to call", because the margin of error said that the difference between the candidates was not statistically significant. But the drive to make a firm prediction overwhelmed his push to keep the statistics rigorous.
Fivethirtyeight's post-election analysis of why the predictions weren't right was very insightful, however.
Re: (Score:2)
But the drive to make a firm prediction overwhelmed his push to keep the statistics rigorous.
I wouldn't call giving Hillary a 70% chance of winning a "firm prediction". It's statistically on par (almost) with saying "if you flip a coin twice, you probably won't get heads both times".
Re: (Score:2)
how exactly was this ever supposed to be an actual business?
Sell ads?
Re: (Score:3)
I don't recall exactly what was said but he made it sound as though there was no uncertainty about Trump losing.
I'm going to shamelessly steal this link from another poster who disagreed with me: Final Election Update: There’s A Wide Range Of Outcomes, And Most Of Them Come Up Clinton [fivethirtyeight.com]
They prominently gave Trump a ~30% chance of winning.
Re: (Score:1)
But the drive to make a firm prediction overwhelmed his push to keep the statistics rigorous.
I wouldn't call giving Hillary a 70% chance of winning a "firm prediction".
The fivethirtyeight headline: "Final Election Update: There’s A Wide Range Of Outcomes, And Most Of Them Come Up Clinton."
The 27-28% chance of Trump winning was mentioned in there... but in the fine print. If the exact same article had the headline "Statistically too close to call" it would have been honest, and Silver would have been proclaimed a genius by being the only one to not say "it's Hillary".
https://fivethirtyeight.com/fe... [fivethirtyeight.com]
Re: (Score:3)
Literally the second paragraph says Hilary had a 71% chance of winning, that's hardly "fine print" as an articles main body is not fine print. The title reflects those odds as well, notice it doesn't say "...all of them come up Clinton". I also don't remember having any problems understanding this point during that election and 538 was my go to then.
A 71% chance is not 100%. I think a lot of people just don't understand how probability works.
Re: (Score:2)
The 27-28% chance of Trump winning was mentioned in there... but in the fine print.
I looked at the article you linked and the second paragraph says "Clinton is a 71 percent favorite to win the election". Trump's 29% chance is only in the "fine print" to the degree that you first have to do the mental math of 100 minus 71. Even setting that aside, the 28-29% figure is even prominently displayed in the graphic.
If the exact same article had the headline "Statistically too close to call" it would have been honest
Again, Silver didn't "call it" for Hillary. He gave her better than even odds of winning. You seem to be making the assumption that polling error would go in Trump's favor (which wit
Re: (Score:2)
The 27-28% chance of Trump winning was mentioned in there... but in the fine print.
I looked at the article you linked and the second paragraph says "Clinton is a 71 percent favorite to win the election".
Was I unclear? I am talking about the headline. Not the second paragraph. Not the first paragraph. The headline.
Trump's 29% chance is only in the "fine print" to the degree that you first have to do the mental math of 100 minus 71. Even setting that aside, the 28-29% figure is even prominently displayed in the graphic.
You apparently missed the line where I said
If the exact same article had the headline "Statistically too close to call" it would have been honest
...
You seem to be making the assumption that polling error would go in Trump's favor
No. I am saying that if the difference is less than the margin of error, the result is not significant.
Re: (Score:3)
Nate Silver ran simulations based upon the poll data and the model he had, and in (roughly) 70% of the simulations Clinton won, and (roughly) 30% of simulations Trump won. So that's ultimately where the number came from - it wasn't just something he picked arbitrarily so he could have prediction.
Assumptions had changed. (Score:1)
For a 2nd point of reference, I note that in the run up to the 2016 election his *reasoning* for his predictions was based on historical trends.
Before the party conventions he predicted Marco Rubio to be the candidate, because historically the position Marco held had been the candidate in many previous elections.
That's a basic fallacy. The "Baltimore Stockbroker" is more than a scam, it's a fallacy that we should watching for in our daily lives. It comes up in finance whenever you hear "this guy predicted t
Re: (Score:1)
A far simpler explanation is that candidates still matter and Hilary Clinton was not a good candidate, so less likely to win. Unless you argue that "the conglomerates fix" was in and that's why
Re: (Score:2)
For a 2nd point of reference, I note that in the run up to the 2016 election his *reasoning* for his predictions was based on historical trends.
Before the party conventions he predicted Marco Rubio to be the candidate, because historically the position Marco held had been the candidate in many previous elections.
That's a basic fallacy. The "Baltimore Stockbroker" is more than a scam, it's a fallacy that we should watching for in our daily lives. It comes up in finance whenever you hear "this guy predicted the past 5 stock market crashes, and for $10 you can get his book where he predicts the next one". (Compare with the number of people trying to hawk a book on the stock market, the number of published measurements of the stock market (around 200, AFAIK), and the probability that one of these correlates with the stock market crashes of the past.)
That's not how that fallacy/scam works.
The fallacy/scam is among a pool of many groups of random predictions at least a few of those groups of predictions will be very accurate. But, since they're random, they're highly unlikely to repeat that success in the future.
On the contrary, relying on historical trends is saying that the thing that happened many times before it likely to happen again, as long as you're relying on inputs that are highly related to the outcome, which Silver presumably did.
And so what
Re:Assumptions had changed. (Score:5, Insightful)
And you completely ignore his advice that you should not use the models to prematurely declare a "winner", because all the models are giving are probabilities. A 99% probability to win is not a win, and if you read a win there, you misread it. He did not predicted Marco Rubio to be the candidate. He just reported the chances of Marco Rubio to become the candidate. And at the begin of the election cycle, Marco Rubio had a pretty good lead. But as in sports, an early lead does not cement the final result. Otherwise we could stop each NFL game after the first touchdown.
It's a general problem with reporting on polling, and many people fall into this trap. Polling gives you probabilities, nothing more, and even they have their limits. Read a book about sampling, and sampling error, and how you have to account for that. Everyone who said that a polling "predicted a winner" ignores how polling works, and what polling actually provides.
Re: (Score:2)
That's a basic fallacy. The "Baltimore Stockbroker" is more than a scam, it's a fallacy that we should watching for in our daily lives
It's just classic survivorship bias. Every time somebody says something like "they just don't build things like they used to" falls right into it -- they're looking only at the things that are still around, and ignoring everything that isn't.
Re: (Score:2)
Nate trying to call polling a "real business" is hilarious.
Fivethirtyeight isn't a polling site. It's a news site. They report on polling, but the polling is done by (a wide variety of) other organizations.
With that said, it was a news site occupying a very narrow niche. It was by far the best occupant of that niche, but the niche audience was pretty narrow.
Re: (Score:2)
There were a relatively small number of outcomes and a large number of predictors being platformed by the corporate media. One of them was bound to be right. Nate has never demonstrated any skillset other than marketing. And marketing is worthless to consumers so why would he have an audience?
Your claim is testable. If Nate Silver is part of a "Baltimore Stockbroker Scam" as you claim, there should be another individual that has proven to be more accurate by now. However, Nate Silver famously was wrong about the 2016 election, [fivethirtyeight.com] and no one has claimed the prize as a superior prognosticator.
Re: (Score:2)
Asking me to "find someone better than Nate" when no one has actually demonstrated that Nate actually has any ability. His just a degenerate gambler. His background was in sports betting, not political science.
The entire argument for Nate's acumen as a p
Who needs polling (Score:2, Funny)
If we aren't going to have more elections?
Re: (Score:1)
Even if we do have elections, voter suppression is creating a growing gap between how people want to vote and how they can actually vote. Asking them about their intentions prior to the election is becoming increasingly irrelevant.
Huh! (Score:2)
Wow. Well, at least our Canadian version [338canada.com] still seems to be going strong.
Fuck nate silver (Score:1)
Nate made one decent model of a single election (Score:1)
And he did a good job of riding it as far as possible.
Who would have think that ABS does not care? (Score:2)
"The basic issue is that Disney was never particularly interested in running FiveThirtyEight as a business, even though I think it could have been a good business."
It does not take Superhero powers to figure out that small business like 538 do not typically thrive under the umbrella of big businesses like that. They are peanuts to them, and if typically does not even matter if they are good. I am sorry for the people who put their hearts into it, but maybe people will learn from this .... Nah, who am I kidd
Poaching should stop soon (Score:1)
I can't imagine the Washington Post poaching staff who tell any uncomfortable truths about President Trump or Co-President Musk in future, so that's a problem that Disney would not have had in future. The NY Times also isn't going to say anything bad about those two (having ceased to do so well before Trump won the election) so they're not a likely poacher either.
But Disney and ABC are not in the business of broadcasting or publishing anything that the Trump-Musk administration dislikes either, so I guess t
Re: (Score:1)
NPO this (Score:2)
we never had enough "product" people or strategy people to help the business grow
Hey, here's an idea, which doesn't require a marketing staff!
Start a non-profit organization, and crowdsource your money. I would donate for sure.
Otherwise, you reap what you sow.
If 538 started as a 501(c)3 or 501(c)4, or was one under Disney, I apologize for the assumption. My assumption is if ABC News was supporting it, they at least pressured them to be profit-oriented. The quoted text indicates to me that the corporate structure of 538 was the problem.
Re: (Score:2)
Nate himself is famous enough that he "doing just fine" (AFAIK; not like I know the dude). He maintained ownership of his model and methodology and he's still working on them and publishing on his Substack, where TFA is.
Re: (Score:2)
Originally the Disney buy wasn't in the name of ABC but rather ESPN: applying the statistical models to sports for predictions.
While they did continue sports reporting 'til the end, I'm guessing the accuracy rate wasn't quite as high as their reputation for political polling was, and they didn't grow. An educated audience doesn't go to back prognosticators when they don't show accuracy, and an uneducated audience only goes to those that keep telling them what they want to have happen and somehow bury their
Noncompetes? (Score:2)
I wonder of the 538 staff had to sign non-competes to prevent them from forming a new company after these layoffs. Assuming they're under a non-compete for two years (Usually the maximum permitted time), that puts us past the midterm election cycle, but maybe will allow a new company to be formed before the next presidential election.
Before they were bought out be Disney, they had wonderful in-depth reporting of the presidential election polling. It wasn't as good during the last election cycle, and I suspe
If you wanted control (Score:1)
If you wanted control, you shouldn't have sold out.
When you sold the business, you gave up the right to determine it's future. Now it is gone, and you are rich.
So sad. Too bad.
Re: (Score:2)
Re: (Score:2)