Why we should care about the Nate Silver vs. Nassim Taleb Twitter war • Hacker News Board

Re: Why we should care about the Nate Silver vs. Nassim Taleb Twitter war

#161

Post by scott_s » Tue, Dec 18, 2018, 3:23 PM UTC

hn_throwaway_99 wrote: post #149

Kurtz79 wrote: post #63

Earlier quoted context omitted.

"Saying "this is not something their readers follow suit on" - only if you're a reader who doesn't understand probability." Unfortunately that is the more common case, and I agree on the author on that that most people take a binary stance on polls. It should not be the case for regular readers of a website whose main topic is statistical analysis, although I imagine that just before a major election they would see a…

Fine, but Silver goes to extreme ends in my opinion to explain what a probability means. During the 2016 presidential election he was emphasizing "Hey, it's less likely than the alternative, but there is a real possibility Trump will win. There is a very plausible path for him to wrap up the electoral votes." And afterwards, I think the results showed how accurate Silver was - he gave a much higher probability to Tru…

I agree. I think Silver laid it out clearly in a piece that people called him a troll for writing: Election Update: Yes, Donald Trump Has A Path To Victory, https://fivethirtyeight.com/features/election-update-yes-don.... Another political writer at 538, Harry Enten, had a similar story a few days later, Trump Is Just A Normal Polling Error Behind Clinton, https://fivethirtyeight.com/features/trump-is-just-a-normal-...

I also listen to their politics podcast, and they constantly talk about how to prevent people from rounding up an 80% chance to a 100% chance. One change they made this cycle was to try say 4 out-of 5 chance (which is mathematically the same) to convey both the lack of precision and to make it more intuitive. I think it helps. Personally, when I talked to people about election outcomes, I phrased it as "Based on what I know, I will be surprised-but-not-shocked if x happens, and shocked if y happens" to differentiate between, say, a 20% chance and a 1% chance.

Re: Why we should care about the Nate Silver vs. Nassim Taleb Twitter war

#162

Post by pmart123 » Tue, Dec 18, 2018, 3:28 PM UTC

_imba_ wrote: post #52

pmart123 wrote: post #33

I liked the first chapter of "Fooled by Randomness," but I stopped reading the book when each chapter regurgitated the same points, each more nihilistic than the last. Perhaps Taleb should concentrate on the accuracy of his own predictions: https://www.businessinsider.com/taleb-every-single-human-bei... >>> "every single human being" should bet U.S. Treasury bonds will decline. >>> Short the S&P vs Long Gold, in a 5…

He gets predictions more than most. You should go read that book. There are many chapters on the incorrect weight people add to the outcome of correct predictions (or incorrect predictions, if that is what suits their agenda).

For me, it'a less about the accuracy of his predictions, and more about his intellectual smugness rubbing me the wrong way. He has no problem with the hypocrisy of attacking Nate Silver over some supposed weaknesses in his election prediction model, but does not seem to care about the inaccuracies of his own public predictions.

Nate Silver took political polls and hunches, which were far less accurate, and came up with something significantly better. Even in 2016, Fivethirteight's model showed the odds of Trump winning were not that remote versus major media outlets "guessing" the race was all but over.

Election prediction models will always be inherently unstable in my mind. It's pretty much impossible to predict the when, if, how, who of an Anthony Weiner type scandal.

Re: Why we should care about the Nate Silver vs. Nassim Taleb Twitter war

#163

Post by scott_s » Tue, Dec 18, 2018, 3:34 PM UTC

ThrowMeDown01 wrote: post #154

yorwba wrote: post #91

Earlier quoted context omitted.

> As the blog post rightly said, they present 538 present the forecast in a way that they are never wrong. The whole point of probabilistic modeling is to replace absolute decisions like "right or wrong" by continuous weights on the possibilities. If you absolutely need a definite decision, you can sample a prediction according to the probability assigned by the model. If the true outcome is x and the model assigned…

> The whole point of probabilistic modeling The whole point about the current topic as well as of my post: You missed by about a thousand miles. Please read it again. It's really pointless to argue about a strawman created by you. Your model is useless, that's the point! It makes no real (!) predictions - not usable for anything apart from blowing ever more hot air, and if it doesn't come to pass, you are never wrong…

Nate Silver analyzes their recent predictions against actual outcomes here: How FiveThirtyEight’s 2018 Midterm Forecasts Did, https://fivethirtyeight.com/features/how-fivethirtyeights-20...

Re: Why we should care about the Nate Silver vs. Nassim Taleb Twitter war

#164

Post by scott_s » Tue, Dec 18, 2018, 3:43 PM UTC

FactolSarin wrote: post #120

paganel wrote: post #62

Earlier quoted context omitted.

>The 2016 election was swingy and it was uncertain. I don't remember it that way, in fact I remember seeing a lot of very shocked faces in the Democratic camp when the results were beginning to take shape. I don't remember having seen similar confused reactions after any previous US presidential elections, not even after the Bush vs Gore one which was a lot closer in terms of electoral votes. Back to the article, I t…

However, FiveThirtyEight gave Trump a 30% chance of winning, and in the days before the election was stressing the fact that a Trump win was very possible. They didn't predict his win, but they did better than anyone else in the mainstream.

I agree. I think people's shock was a failure of imagination. They couldn't imagine a Trump win, so even though the polling data showed he could, they discounted it. That, I think, explains why Silver was criticized so heavily for saying what he did days before the election: they couldn't imagine a Trump win, therefore he must lose, which Silver must also think, therefore Silver is trolling.

Re: Why we should care about the Nate Silver vs. Nassim Taleb Twitter war

#165

Post by mschuster91 » Tue, Dec 18, 2018, 3:44 PM UTC

Of course the US elections are not as accurately predictable as e.g. German ones, or sports events. Not simply due to different pollster sources and their data quality (which Nate Silver tries to account for).

The biggest and most unpredictable factor in US (and British) elections is their winner-takes-all system, with the second layer of the Electoral College in the US which does not account for popular votes either.

Re: Why we should care about the Nate Silver vs. Nassim Taleb Twitter war

#166

Post by lozenge » Tue, Dec 18, 2018, 3:49 PM UTC

logifail wrote: post #76

ummonk wrote: post #72

Earlier quoted context omitted.

I normally like Taleb's insights but it was obvious when following this that he (and Dinesh D'Souza) were ridiculously misconstruing Nate Silver's comment. Two outcomes being "extremely possible" in no way means they have a 50/50 chance of happening. It merely means that neither outcome has a low enough chance that you should assume it won't happen.

>> Two outcomes being "extremely possible" What on earth is a professional like Silver doing using the word "extremely" in that context? If I'd have written that kind of statement in school, the teacher would have crossed it out in red pen and told me to rephrase. If I toss a coin, are the outcomes "heads" and "tails" both supposed to be "extremely possible", with "landing on its edge" being merely "unlikely"? (In ca…

There's nothing wrong with extremely likely if defined. Even without a strict definition people have some understanding of what it means. https://www.theglobeandmail.com/news/world/what-likely-and-e...

Re: Why we should care about the Nate Silver vs. Nassim Taleb Twitter war

#167

Post by mercutio2 » Tue, Dec 18, 2018, 3:56 PM UTC

CompelTechnic wrote: post #131

duckerude wrote: post #113

Earlier quoted context omitted.

If I'm understanding it correctly, aleatory uncertainty is uncertainty you understand, and epistemic uncertainty is uncertainty due to lack of knowledge. Uncertainty about the outcome of an unbiased die roll is aleatory. You don't know what's going to happen, but that's because of inherent uncertainty, not because you don't know enough about the die. So while you can't be sure what the outcome will be, you can know t…

To continue the dice analogy- I can imagine a world where gambling sharks arrive at the roulette wheel in casinos, wearing hidden google-glass-ish devices, and place bets while the ball is in motion. Unbeknownst to everyone else, their device performed real time computation based on the position and velocity of the ball and wheel, and gave an output of a tight probability distribution (10% chance 25 red, 35% chance 2…

You don’t have to just imagine such a world. This actually happened in the 1970s (although it was a shoe computer and a team of well trained physicists, not Google Glass)!

https://en.m.wikipedia.org/wiki/The_Eudaemonic_Pie

Re: Why we should care about the Nate Silver vs. Nassim Taleb Twitter war

#168

Post by kilburn » Tue, Dec 18, 2018, 4:04 PM UTC

0-_-0 wrote: post #151

kilburn wrote: post #128

Earlier quoted context omitted.

Using the phrase "extremely possible" as an expert is either a gross mistake or intentional misguidance. "Possibility" is like "optimality" in the sense that they are binary attributes and thus don't admin grades. Something either is or isn't possible. Qualifying something as "more possible" is a mistake, and qualifying it as "extremely possible" is just nonsense. A guy who lives from probabilistic analysis should no…

I suspect the main reason he was attacked by Taleb was this misuse of language, which can be misinterpreted as meaning a 50-50 chance. Like others have said, he could have said "far from impossible" or "entirely possible" to convey that he meant the estimate had a large variance. Then again, I think Taleb should have realized this and not attacked him for simply failing to convey variance correctly. This seems like a…

Apparently there's people who have hard feelings in this argument too.

I could understand people downvoting my comment (it may be interpreted as harsh even if that wasn't my intention) but... yours? Something weird is at play here.

Re: Why we should care about the Nate Silver vs. Nassim Taleb Twitter war

#169

Post by uncletaco » Tue, Dec 18, 2018, 4:04 PM UTC

zimablue wrote: post #82

cm2012 wrote: post #8

Earlier quoted context omitted.

Taleb is just mixing up the nowcast and the forecast, even after being corrected many times.

I think he would say that the "nowcast" is so useless and deceptive to be malpractice, since everyone will interpret it as a forecast, and even if they didn't it's not in any way useful. Which I agree with (unsurprising since I've semi-invented it as an opinion for another person).

Silver has said before, at least in the podcast, that everyone at FiveThirtyEight hates the nowcast and its various incarnations but I think there's some pressure from the business side of things (for views, likes, shares, etc) to have something volatile for people to eat see day in and out.

Re: Why we should care about the Nate Silver vs. Nassim Taleb Twitter war

#170

Post by mercutio2 » Tue, Dec 18, 2018, 4:12 PM UTC

huffmsa wrote: post #141

ubernostrum wrote: post #121

Earlier quoted context omitted.

I've always treated 538's forecasts as meaning nothing more or less than "We have a model, we fed the current data into it and ran N simulations of the event, and in X% of those simulations, this was the result". Which, as I understand it, is literally how they're generated. Any discussion of the accuracy of the model (which should be the only thing that admits debate) has to be retrospective. Just saying "well the f…

Yes but that's not how the proles understand the blog. They take it as "x has a y % chance of happening". Which is Taleb's beef with it. The commoner thinks its Silver putting his money on something, but when it doesn't happen he says "well I didn't say _that_". And it in the day and age of more robust machine learning models, that the 538 models are so volatile is a pretty weak excuse.

Humans are volatile. Politics even more so.

Our options aren’t “high confidence converging predictions, or 538”.

The options in the current era for understanding of the electorate’s mood are “overly confident individual polls, poorly analyzed with completely inadequate models”, vs “538-style epistemically humble models accompanied by discussions of their confidence, which can be scored in aggregate after each election”.

I’ll take 538 any day.