Thursday, October 22, 2020

What Republican internal polling can actually tell us

 

begin quote from:

https://www.cnn.com/2020/10/22/politics/republican-internal-polling-analysis/index.html

What Republican internal polling can actually tell us

(CNN)Former Vice President Joe Biden has held a steadfast lead in high quality nonpartisan public polling. Still, some conservatives and President Donald Trump claim the polls are off. Trump himself has said he is up in key swing states.

There's no reason to believe Trump. An examination of publicly released internal Republican and conservative group polling reveals they're also showing Trump clearly underperforming his 2016 showing.
I looked at more than a dozen of these partisan polls released to the public from House and Senate races since the major party conventions in August. These partisan polls are notoriously unreliable, and none of them meet CNN standards for reporting.
    The reason is simple: Partisans don't want to release polls that are bad for their side. That means the polls sponsored by a party, candidate or partisan organization tend to be biased in favor of the side releasing the poll.
    That's why it was amazing to find that on average, Trump was doing 5 points worse than he did in 2016 in the states and districts in released Republican and conservative polls.
    If Trump actually did 5 points worse than he did in 2016 in the swing states, it would mean he'd lose Arizona, Florida, Michigan, North Carolina, Pennsylvania and Wisconsin. Those states, plus the ones Hillary Clinton won in 2016, would be more than enough for Biden to get over 270 electoral votes.
    But remember: These are only the polls conservatives and Republicans were willing to put into the public sphere. There's good reason to believe it's worse for Trump in the numbers they're not releasing.
    A study from FiveThirtyEight reveals that internal polls are 4 or 5 points more favorable on average to the side for whom the poll was conducted than what you'd expect from a nonpartisan pollster in the same race.
    This meshes with what was previously reported by Dave Wasserman of the Cook Political Report -- mainly, that internal polls by both sides that have not been released are even worse for Trump than the internals that are being put out by conservatives and Republicans.
    Indeed, one of the more interesting aspects of the internal polls so far put out by Republicans is how few of them there are.
    As I noted back in the summer, the number of internal polls each side releases publicly is usually a good indicator of how their side is doing. If one side is doing well, they're more likely to release internal polling than if they're doing poorly.
    Since the conventions, Democratic candidates and groups trying to get them elected have put out about 75% of the House internal polls released by either side. That isn't nearly as lopsided as it was for a period in the summer, but it's still large by historical standards.
    Democrats also put out about 75% of the House internal polls in both 2006 and 2018.
    Those were years in which the national environment heavily favored the Democrats and they took back control of the House.
    This year, the Democrats already have the House, and it's unlikely we'll see the same national vote swing in the Democrats' direction. But we'd likely see a lot more internal polls released by Republicans if they were doing well.
      The bottom line is that there's really no reason to think that the Republicans have some secret polling putting Trump in a better position than the public polling does. The signs, in fact, point in the opposite direction.
      Trump and his fellow Republicans seem to be running behind no matter what polling you examine.

      No comments: