In Georgia there are two polls on the last weekend that are 10 points apart? One has Warnock up six and other has Walker up four?
Overall... that would suggest that the internal polling on both sides (which are generally not released but seem to be more accurate than media polling) is more in line with the "red wave" polling and less in line with the outliers that suggest a more muted election where the GOP barely picks up the House and the Democrats keep the Senate (maybe even picking up a seat).
So how can you be that far off from each other? Very simple. One poll shows Walker winning only 84% of Republicans and 37% of Independents. The other shows Walker winning 92% of Republicans and 48% of Independents. That is a big difference.
The lack of clarity in these demographics (which used to be fairly consistent) has caused problems for my manner of prognosticating. Historically, the main differences between pollsters was not how people were going to vote, but how many of each group would turn out. Partisan breakdown was the major difference in most divergence. Adjusting to national averages could mitigate much of these differences. But now they cannot even agree with how people will vote.
_______
Are conservative pollsters "flooding the zone"?
I have been watching the major election players on Twitter (Nate Silver, Nate Cohn, Sean Trende, etc) and there has been a lot of debate as to whether or not the heavy recent movement towards the GOP has been nothing more than conservative pollsters "flooding the zone". In other words, making it look like a GOP landslide to encourage people to vote for the "winner" (who wants to vote for a loser) and otherwise discouraging Democrats from voting.
I am not sure how all of this started, but it has gotten to the point of Trende literally "mocking" the argument as pollster after pollster (right and left) keep showing the GOP doing well. I believe it was Silver who suggested the obvious. If liberal pollsters had different results, they would be releasing them. The lack of many of these liberal pollsters "releasing" their poll results is supposedly something in an of itself. The idea floated by Silver being that it is inconceivable that these pollsters are not polling. But rather they are polling but not releasing the results because they do not like them.
_______
Should we be looking at the actions and buzz from each Party?
There has been no lack of Republican insiders, Republican pollsters, and Republicans strategists suggesting that they are actually doing better than much of the polling suggests. I have read several such prediction that the GOP will sweep the close Senate races, winning Georgia, Pennsylvania, Nevada by bigger margins than expected, and even winning in New Hampshire and Arizona. The dude who predicted Trump in 2016 and Brexit is suggesting this will be a total wipeout. We'll see.
Meanwhile the buzz is that Democrats are expecting to lose the House and in spite of public statements saying otherwise, they do not like their chances in the Senate either. While the President is out on the stage telling everyone that he believes Democrats are poised for pick ups, those insiders, pollsters, and strategists are far less excited about their chances.
Overall... that would suggest that the internal polling on both sides (which are generally not released but seem to be more accurate than media polling) is more in line with the "red wave" polling and less in line with the outliers that suggest a more muted election where the GOP barely picks up the House and the Democrats keep the Senate (maybe even picking up a seat).
_______
Lastly, a poll that I found interesting suggests that two thirds of independents are afraid to tell people (especially friends and family) who they are voting for? Read into that what you will... but I suspect that it does not bode well for the accuracy of our current polling.
