2016 Postmortem
Related: About this forumBasic problem: primary polling doesn't work very well.
Turnouts are small compared to generals. Local factors and voter motivation on the day of the vote is everything. Winds change quickly. It's just one state at a time and highly specialized demographics can swing results.
It seems Nate Silver's too far up the bunghole of his own successes to admit these obvious realities even to himself. Cleverly weighted averages of a bunch of junk are still junk.
When the data-driven people have everyone wired via brain chip - I'm sure some of them are looking forward to that - they might manage the feat of making primary polls as reliable as they are for general elections.
Meanwhile every other post on this site is like, "oooooh, 98 point lead for Hillary in State X! Stop the fight!"
We see what the effects of this voter suppression strategy have been: higher turnouts for the Republicans, who are eating up all the coverage with their shitshow. (It probably won't work out for them in the end, for obvious reasons, dick size, etc. We will learn if there is a bottom to underestimating the intelligence of the American public...)
CoffeeCat
(24,411 posts)(Posting here too)--
I hope others share my feeling about Real Clear Politics and how they are totally unreliable in primary predictions.
First off, when you have a state that has an impending primary--it is not helpful to factor in polls that are 2 or 3 weeks old. We have very little time between primaries. It is not helpful to rely on RCP averages that aggregate polls from weeks ago. When you've got a week between state votes--so much happens to the numbers in that week.
Look what happened to Michigan! Yesterday, RCP said that Hillary was ahead by 21 points. This was based on past polls and a couple of recent polls that did show the race tightening. However, they included two AWFUL polls that should have been relegated to the trash heap. One robo poll had Hillary ahead in Michigan by 30+ points. It was ridiculous.
When you have states in which there is only one or two weeks to convince people--the dynamics are constantly shifting. The campaigns hold rallies, events, their ground games get geared up and ads are running. The numbers can shift drastically day to day.
Going forward, I think it would be wise to just trash any outfit that aggregates these polls. What we really need is a very recent, reliable poll. Anything else, is shit. And a waste of time.
JackRiddler
(24,979 posts)The use of polls intentionally as manipulation and psychological warfare against whoever's allegedly behind (in this case, Sanders).
It's one thing for large state and national polling tracks to be used to build momentum, given all the other obstacles involved in campaigns with the media and the money and the political culture. But this daily drumbeat of posting the results from 100 for-profit polling outfits and celebrating as if they mean anything is utter noise and bullshit.
It has amounted to a voter suppression strategy. And from this moment, obviously, that has failed for the Clinton camp.
angrychair
(8,594 posts)Like some deviant perpetual motion machine. The outcome is polls as soft as mama's angel food cake and bullshit media talking points that feed off each other to create as deceptive a false narrative as possible.
I have never seen this much effort put into a single candidate like this. CNN has a panel of 8 and every single one of them is completely pro-Clinton.
Sucks.