*adds to long-form writing list*
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
And we don’t get “20% erosion of democracy”
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Can you elaborate? Of course a probabilistic prediction of a one-off event is less useful than a daily weather forecast but you implied they were not useful at all and misleading.
-
It’s the tightly-coupled errors—not just polling errors—due to rare event in changing terrain plus the reflexivity. Me taking an umbrella along doesn’t interact with rain chances. And *admittedly hand-waving* both outcomes are probably within one-two sigma true error. Now what?
- Show replies
New conversation -
-
-
You didn't say they were "less useful than widely-assumed." You said they were "pointless or misleading" because an event forecast as an 8% probability occurred. If I (and others) misunderstood you, I apologize, but if so I still don't understand your point.
-
Misleading because true error isn’t reflected. Probably not possible. Too few events to model from. Pointless because of reflexivity. Will see if I can write longer.
- Show replies
New conversation -
-
-
They're fun to read, though.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Also... useful to whom?
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
How much does this extend, in your view, to Congressional races? On the one hand, a lot of the same problems and worse polling. On the other, a LOT more past data to use.
-
Districts are easier imo. US electoral college (a lot of coupled first-past the post events) is wicked.
- Show 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.