Election 2016; Super Bowl 2017. Probability models are not like poll percentages. Probability models are not like poll percentages. Probabi—
-
-
A 60-40% election is a landslide. A 60-40% probability model is a near toss-up, especially if error bars are sizeable (as they often are).
-
That extra digit after the decimal? Scientism. Probability models shouldn't imply that kind of confident precision.https://twitter.com/WSJ/status/828419447698366465 …
- Show replies
New conversation -
-
-
there were intrastate correlated errors in election. Not a particularly applicable term here.
-
8.4% chances happen sometimes! About 8.4% of the time.
End of conversation
New conversation -
-
-
yes. Thank you. A million times yes.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Not just bad communication, but bad science/math, failure to allow for "black swans," etc.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Agree. Headline that said score and quarter would have been much more useful.
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.
