Um, poll Twitter. State errors may be correlated. More interesting Q: can we model a rare event based only on 11 previous cases—as 538 does?
I buy some state errors may well be correlated. But multiple sources of error, so may not be as important overall.
-
-
All im saying is I don't see why assuming zero correlation would be better than best estimate based on available evidence.
-
Up 3 points in 5 heterogeneous swingers != up 3 points in 5 homogeneous states. Fail to account for this = bad model.
End of conversation
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.