The polls aren't skewed, Trump is losing to Clinton http://washex.am/2aOkfKi via @DCExaminer
I think in 14 it underestimated GOP performance in some senate races but got trend right. (You're right about 2012 Obama.)
-
-
…then the cause was same: they’re doing such heavy weighting based on such small sub-samples of young/PoC hard to get right
-
Fair and good points. ... Curious to see what we (me) learn from polling this cycle.
-
I suspect we’ll see a growing gap btwn media & campaign pollsters, bc of growing diff btwn RDD & samples from voter file
-
I'm assuming that data analytics allows a good campaign to know A LOT more than media or run-of-mill pollsters. Yes?
-
Yes, bc trends in oppo directions. Campaigns getting better, & RDD polls getting nearly impossible to consistently do well
-
What I figured. I figure that's a real good reason to pay attn to ad patterns and personnel.
-
SHHHH. Don’t give away a good idea that sets you apart from vast majority of campaign reporters & analysts!
- End of conversation
New conversation -
-
-
BTW, I think 2014 was different also in that battlegrounds mostly states w few minority voters. Played to pollsters’ strength
-
More landlines, easier demo mix to nail down - that's possible.
-
Someone—maybe
@davidshor ?—recently said he’d trust an all-landline poll using a file over an RDD poll w mix of land/cell
End of conversation
New conversation -
-
-
2014 was only time recently where it wasn’t overwhelmingly tilted R by 3-4 points (was mostly too Dem). But I think even…
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.