News for hackers. Stuff that matters.

A comment on Why we should care about the Nate Silver vs. Nassim Taleb Twitter war. Parent comment

Re:Why we should care about the Nate Silver vs. Nassim Taleb... by hn_throwaway_99 on Tuesday December 18, 2018 @09:58AM (#18706787)
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 Trump winning than anyone else IIRC, and the outcome was pretty much that: An unlikely probability (many of the deciding states were very much within any possible margin of error) that actually happened.
[ Parent | on HN ]
Re:Why we should care about the Nate Silver vs. Nassim Taleb... by Kurtz79 on Tuesday December 18, 2018 @12:16PM (#18708323)
I was following the website frequently at the time and I fully agree.

Bur I was just commenting that that the author is spot-on at least in that many readers do not take these explanations into account and they see just 28.6% of winning chance and they automatically convert it to 0% in their mind.

And the explanation is that they don't understand math and probability (or don't want to), whether the author likes it or not.

[ Parent | on HN ]
Re:Why we should care about the Nate Silver vs. Nassim Taleb... by coryfklein on Tuesday December 18, 2018 @01:06PM (#18708869)
I still don't get why the author thinks it's OK to assign blame to 538 for readers interpreting things incorrectly when Nate clearly spends a significant amount of time trying to educate the readers on how to interpret.

It makes no sense to me to hold 538 accountable for readers interpretations of the data. If I wander into some realm of science I'm unfamiliar with (say, quantum mechanics) and start making false inferences from it, does that mean the scientists were misleading me?

[ Parent | on HN ]
Re:Why we should care about the Nate Silver vs. Nassim Taleb... by scott_s on Tuesday December 18, 2018 @10:23AM (#18707078)
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.

[ Parent | on HN ]
Re:Why we should care about the Nate Silver vs. Nassim Taleb... by leereeves on Tuesday December 18, 2018 @01:48PM (#18709322)
That's not the whole picture. Before he learned to be more careful, Silver was calling Trump a troll, doomed to fail, who wasn't even trying to win.

https://fivethirtyeight.com/features/donald-trump-is-winning...

https://fivethirtyeight.com/features/donald-trump-is-the-wor...

[ Parent | on HN ]
Re:Why we should care about the Nate Silver vs. Nassim Taleb... by scott_s on Tuesday December 18, 2018 @01:51PM (#18709353)
Yup, and after that, he issued a complete mea culpa and made it a point to look at the actual data.
[ Parent | on HN ]
Re:Why we should care about the Nate Silver vs. Nassim Taleb... by leereeves on Tuesday December 18, 2018 @02:06PM (#18709510)
Yes, and he admitted to misleading with fake math:

> But it’s not how it worked for those skeptical forecasts about Trump’s chance of becoming the Republican nominee. Despite the lack of a model, we put his chances in percentage terms on a number of occasions.

So it would be more accurate to say that he was careful during the second half of the election, after making significant mistakes (both mathematical and ethical) during the first half.

[ Parent | on HN ]
Re:Why we should care about the Nate Silver vs. Nassim Taleb... by VintageCool on Wednesday December 19, 2018 @09:39PM (#18722066)
By "first half" you mean before Iowa, and by "second half" you mean after Iowa.
[ Parent | on HN ]
Re:Why we should care about the Nate Silver vs. Nassim Taleb... by afthonos on Tuesday December 18, 2018 @10:11AM (#18706920)
Even more confidence-inspiring: before the election, they had discussed what a path to victory for Trump would look like, and their most likely path is what ended up happening: the midwest went for Trump, in a correlated manner, which is exactly what other forecasters had failed to account for properly.

If your model can help you predict how it's going to fail, I'm fairly impressed.

[ Parent | on HN ]
Re:Why we should care about the Nate Silver vs. Nassim Taleb... by leereeves on Tuesday December 18, 2018 @01:43PM (#18709280)
Nate Silver didn't always hedge his bets; early in the election he was making emphatic (and wrong) predictions like these:

> The lesson, rather, is that Trump’s campaign will fail by one means or another.

> Donald Trump Is Winning The Polls — And Losing The Nomination

> Our emphatic prediction is simply that Trump will not win the nomination. It’s not even clear that he’s trying to do so.

[ Parent | on HN ]
Re:Why we should care about the Nate Silver vs. Nassim Taleb... by grzm on Tuesday December 18, 2018 @01:49PM (#18709333)
Do you have dates for those quotes? I suspect many (if not all of them) refer to pre-nomination model. Silver explicitly addressed how he messed up the primary model by not trusting the numbers in his 18 May 2016 blog post "How I Acted Like A Pundit And Screwed Up On Donald Trump: Trump’s nomination shows the need for a more rigorous approach."

https://fivethirtyeight.com/features/how-i-acted-like-a-pund...

[ Parent | on HN ]
Re:Why we should care about the Nate Silver vs. Nassim Taleb... by leereeves on Tuesday December 18, 2018 @02:20PM (#18709657)
Yes, they were pre-nomination articles. At that time Silver was acting in the way an earlier comment in this thread said he does not act:

> saying things with false-certainty and then basking in self-proclaimed genius when luck happened to run in his favor

Something seems to have gone wrong at 538 after their stellar performance in 2012. In 2016 even their Congressional predictions were only 90% accurate. Or perhaps 2016 was simply different from past elections.

[ Parent | on HN ]