The @538politics model looks terrible --if a probability is 70% now, it should never have been 90% 2 weeks ago when uncertainty was higher.
What they are displaying is clearly not a "probability" in any sense, it is far too stochastic. So why present it as such?
-
-
incentives are not aligned s.t. accurate prediction -> more reward. Forecasting tournaments (e.g.
@superforecaster) are aligned. -
538's model and betting markets track pretty closely. Markets slightly more conservative.https://electionbettingodds.com/WIN_chart_maxim_lott_john_stossel.html …
End of conversation
New conversation -
-
-
I'll call it lousy Bayesian statistics with a biased prior.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
new data does not always lower uncertainty. Think black swan.
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.