What they are displaying is clearly not a "probability" in any sense, it is far too stochastic. So why present it as such?
The @538politics model looks terrible --if a probability is 70% now, it should never have been 90% 2 weeks ago when uncertainty was higher.
-
-
-
incentives are not aligned s.t. accurate prediction -> more reward. Forecasting tournaments (e.g.
@superforecaster) are aligned. - Show replies
New conversation -
-
-
The very fact that this could change so easily shows that uncertainty was in fact very high back then. It just wasn't accounted for.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
I think it’s safe to say this isn’t exactly a “normal” election. IIRC they mentioned in one of their posts…
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
…that in previous elections most people would have already made up their mind by this point.
-
Here is the article I was talking about: http://fivethirtyeight.com/features/how-much-do-october-surprises-move-the-polls/ …
End of conversation
New conversation -
-
-
You should read
@NateSilver538's book - he attempts to feign expertise in all sorts of fields.Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Prob of 2 tails is 25%. I flip one coin and it's a tails. Prob of 2 tails is now 50%. Change of 25%, but correct.
-
That said, they could have taken into account typical tightening toward the end of the race to hedge predictions.
End of conversation
New conversation -
-
-
If some future event is 90% probable, shouldn’t that decrease at least 10% of the time?
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
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.