Free Republic
Browse · Search
News/Activism
Topics · Post Article

To: monkeyshine

If all the polls are way off, watch for the meltdown tomorrow.


57 posted on 11/06/2018 10:41:19 AM PST by goldstategop (In Memory Of A Dearly Beloved Friend Who Lives In My Heart Forever)
[ Post Reply | Private Reply | To 55 | View Replies ]


To: goldstategop

I’m looking forward to it. I’m in CA, so I don’t even have to stay up late to watch the lamentations start tonight (I hope)... and I have my own popcorn machine :-)


63 posted on 11/06/2018 10:46:03 AM PST by monkeyshine
[ Post Reply | Private Reply | To 57 | View Replies ]

To: goldstategop

BTW, I am convinced the polls are way off. Putting aside the internal biases of methodologies I think 2 factors the pollsters haven’t adjusted for are 1) GOP voters that won’t tell the truth (they will lie or decline to state) for reasons such as being called racist or general angst about being assaulted by crazy lefties and 2) the polls are self-selecting - it only polls people who are willing to talk to pollsters.

I think there may be something like 2%-3% GOP voters who wont tell the truth, and another 2% or so bias from self-selection. It’s just a hunch but I think there is a slight edge that GOP voters won’t even talk to pollsters.

If I am right, that alone erases most margins of errors.

Then there are all the other problems will the methodology, the biggest ones being that the pollsters wrongly presume that Is break more to the left than they actually vote, and that they get their demographic turnout predictions wrong. I saw a clip of Carl Rove saying that among 18-34 year olds this group is trailing by about 1.5% points lower in EV than they were in 2016.

If I am right about these the 3 stats, it is possible that every poll showing a D lead could be inverted in the final analysis. There is reason to be hopeful. Polls are only as good as the analysis of the data and the assumptions made. I know you know it but so many people don’t realize that pollsters censor some of the results they get to fit their demographic turnout model. If their model doesn’t hold true the whole exercise is useless. I am cautiously optimistic.


72 posted on 11/06/2018 10:57:07 AM PST by monkeyshine
[ Post Reply | Private Reply | To 57 | View Replies ]

Free Republic
Browse · Search
News/Activism
Topics · Post Article


FreeRepublic, LLC, PO BOX 9771, FRESNO, CA 93794
FreeRepublic.com is powered by software copyright 2000-2008 John Robinson