Incredibly and against all odds, the polling community has shown itself to be even more untrustworthy than journalism and political punditry. The pollsters are dishonest, their methodology stinks, and everyone who cites them as authority engages in deceptive cherry-picking.
To Hell with them all.
When I check in on MSNBC, which isn’t embarrassed about being virtually all-Epstein all-the-time for weeks now, they are flashing poll numbers showing that President Trump is wildly unpopular regarding, oh, everything. Of course, I can easily find the opposite analysis in the conservative media, and President Trump will, as is his wont, cite exaggerated or completely made up poll numbers whenever it suits his mood.
Two recent examples of this nonsense come to mind. Extreme left-biased, Trump Deranged sports pundit Jamele Hill blathered on CNN regarding the Epstein “client list” distraction, “What’s always interesting to me is, like, what it takes people to get off the Trump train. It’s not the xenophobia, it’s not the racism, it’s not the bigotry. It is this.” “This” meant that polls showed diminishing support for the President because of contrived hysteria over his association with Epstein that has been well-known for decades.
What always is interesting to me is, like, why is such a persistently wrong, dishonest, race-obsessed hack given a platform for her hate and stupidity at all? At least DEI provides an explanation, but if it isn’t that, then it’s a mystery on the level of the Mary Celeste.
So CNN’s Harry Enten, whose weird niche is using untrustworthy polls to debunk Axis narratives using different untrustworthy polls, “factchecked” Hill.
“When it comes to his approval rating with Republicans, it’s basically as solid as it ever was,” the biggest dork on broadcast news declared. “If you look at the average of polls, his approval rating with Republicans [is] still at about 90%.”
Notably, Enten didn’t factcheck Hill over her assertion that enforcing immigration laws constitutes xenophobia, racism or bias, but that’s another issue.
Then the recently-removed-from-mothballs MSNBC propagandist Chris Matthews suffered a flashback to his days of being an independent and objective liberal pundit and blurted out,
“To be honest with you, the country is moving towards Trump. These polls, they come out and show him not doing well — I don’t buy that. He is a stronger public figure than the Democratic people. Obama still has tremendous charisma — but Trump has strength. [Voters] want a President who is a strong figure. And he’s got it. It’s just there. And half the country buys it.”
Honesty on MSNBC! Watch out, Chris, it could be back to the mothballs!
I think Chris is right, but I couldn’t prove it by polls. Matthews also respects strong Presidents and always has; I find it hard to believe that half the country doesn’t. They prefer demented, puppet Presidents? Really?
Well, again, who knows? It would be nice to have some reliable data, but our current political research community values confirmation bias over truth.
Ethics Alarms 
I have considered any political poll, aka statistical analysis, of any kind to be pure manipulation and therefore hokum. They set out to prove a hypothesis and they do the things necessary to “prove” it and then extrapolate their results to absurdity. Any statistician that tells you otherwise is lying to you.
You can use statistics to “prove” just about anything, it just depends on the subjects that participate in the survey, how the information/questions are presented, how the data is collected, and how the data is compiled; all you have to do is use your imagination.
Consider the following:
A statistical analysis was done to find out what cat turds taste like, and the results showed resounding evidence that cat turds tastes like chicken.
Conditions of the analysis:
They used hungry dogs as the polling subjects, gave half of them small bowls of cat turds and the other half small bowls of chicken pieces, and collected 0-10 rating (0 being negative reactions and 10 being positive reactions) details of how energetically the dogs ate the contents of the bowl in front of them, did they eat it all, did they want more if offered. The results were all 10’s so the statistical analysis was 100% positive in all categories; therefore, the statistics clearly show that cat turds taste like chicken.
Conclusion:
Just because a statistical analysis has been conducted and a conclusion reached, does not mean that what “The Survey Says” represents anything close to actual reality.
The only thing statistics can reasonably show, if they are conducted 100% honestly which they aren’t, is the possibility of a trend which does not prove that a trend actually exists. In politics, statistical polling trends are treated as if they extrapolate with 100% accuracy across the entire population and that is a bald-faced lie.
I’m reasonably well versed in statistics and I thoroughly reject statistics, that’s probably my all time top bias, and for good reason. All critically thinking people should completely reject all statistics, especially political polls, and make their minds up based on actual facts.
I became disenthralled with polls and polling generally after completing courses in “Data Analysis for Public Policy” and “The Logic of Survey Analysis” over thirty-five years ago. A poll can be designed and administered so as to “prove” basically anything the pollster (or the entity paying the bills) wants.
Gutfeld recently dismissed polls because they “poll the narrative.”
in other words, figure out what you want to discuss and do a poll that will support that narrative.
I like the way he framed it.
-Jut