Nate Silver just proved a second election in a row he doesn't know "how math works". (cont)https://twitter.com/NateSilver538/status/1059149034693316608 …
Show this thread
(cont)
which explain why @DineshDSouza is right: you don't change a forecast from 80% to 50% under uncertainty. Second time klueless Nate Silver makes a mistake.
https://arxiv.org/pdf/1703.06351.pdf …pic.twitter.com/CD5FEhVMVf
-
-
When someone says are event and its opposite are extremely possible I infer either 1) 50/50 or 2) the predictor is shamelessly hedging, in other words, BS.
Show this thread -
4- Things are actually worse: that
@NateSilver538 doesn't get that if BOTH X & Non-X are "extremely possible = Probability converging to 50-50 is EXACTLY my problem w/his misunderstanding of probability in forecasting, & point of my paper. Gabish? https://twitter.com/coachglove/status/1059217783886503937 …This Tweet is unavailable.Show this thread -
5- FOR THE RECORD I am not translating
@DineshDSouza's point that it was 50-50: I ALSO understood ~50-50 DIRECTLY from Silver's "extremely" in the linked "The Hill". Gabish?Show this thread -
Footnotes: 1) An interesting point: you can also use the maximum entropy approach to get the same result as my election paper/problem with Klueless Nate. 2) The interesting part of the ArXiv paper isn't elections, but binary martingales. This can be applied to other processes.pic.twitter.com/4pRDSrtLCV
Show this thread
End of conversation
New conversation -
-
-
He *didn't* change the forecast. In the piece, Silver said that Democrats had a 4 in 5 chance of taking the House. Which is ... 80%. D'Souza took The Hill's headline that both outcomes were "extremely possible" and assumed that meant they were equally possible, and thus 50/50.pic.twitter.com/msJu4TW9vi
-
Aren't these people supposed to be the intellectual wing of the GOP? dog help the rest of them is all I can say
-
Yes, they are the intellectual wing of the GOP. That's why Seattle, Boston, Silicon Valley, and pretty much every tech hub is liberal. Republicans don't to science, math, and engineering. Only 6% of scientists are Republican.
- 1 more reply
New conversation -
-
-
He didn’t change it. He literally just said that an eighty percent chance of a democrat victory in the house still leaves a 20% chance of a Republican win and 20% chances happen ... once every five times!
-
I wish people could understand that an 80% chance occurs only 80% of the time. The problem is that Nate spoiled everyone in 2008 where he got every single coin flip right. He said some state was 55-45 and it turned out the 55 came in. Idiots figured that 55% actually meant 100%.
-
Would you cross the street if you have a 20% chance of being hit by a car? Seems like it’s a pretty high chance when you look at it that way.
-
20% IS a big chance. 20% odds happen all the time. OK they happen 20% of the time but they happen a lot. I wouldn't cross the street if I had a 0.01% chance of getting hit by a car.
-
I don’t feel comfortable swimming in the ocean. My chances of being attacked by a shark are 1 in 11.5 million.
-
Are the odds really that high? I guess it depends on definitions. Lots of people go into the water a number of times each day. But it makes a lot of sense to avoid high risk activities even when high risk is 10,000-1.
-
So you're saying you dont ride a motorcycle ...
-
I don’t know what the stats on motorcycles are, but my cousin died on one and an aunt and uncle were hospitalized for months after an accident.
End of conversation
New conversation -
-
-
Well this is sad for Nassim Nicholas Taleb. This is a brand damaging tweet.
-
Seems to be right on brand. Frequently wrong and never in doubt (tm).
- 2 more replies
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.