RightDataUSA

Demographics and Elections Commentary tagged with Alternate Universe

8/23/2024: Reverse Poll-arity [RightDataUSA]

Your humble author here at RightDataUSA can now see that he wasted his time many years ago getting a 4-year Bachelor of Science degree in Mathematics and Statistics, because it turns out that a person can become an expert on subjects like polling, sampling, margins of error, etc. without any expensive formal education. He can demonstrate his alleged expertise simply by parroting the same boilerplate drivel which those who are offended by unwelcome polling results routinely resort to.

As the chart above (from Real Clear Polling) shows, this year's presidential election polls were going fairly well (i.e. telling us what we wanted to hear) up to a certain point, and now most of them are absolutely bogus (i.e. telling us what we don't want to hear). These bogus pollsters obviously live in their own fantasy world. They are clearly Democrat puppets with no interest in reflecting reality, but instead are focused on trying to gaslight the public and shape reality to their liking (oh dear, there's some of that boilerplate drivel).

Of course Democrat pollsters damn well do engage in gaslighting, but that (along with the rest of the drivel) is merely a weak crutch; when these crutches are constantly used and abused in an attempt to explain away every poll we don't like -- just as "Frodd, frodd, frodd!!!" is used to explain away every election outcome we don't like -- these feeble rationalizations lose whatever factual impact they may actually have possessed, and they become a cogent explanation for precisely nothing.


Real Clear Politics polling average: Trump vs. Biden (click image to enlarge)

Donald Trump was doing reasonably well in what turned out to be the final polls against Joe Biden, but suddenly things are a lot tighter or have even flipped in some places. Weird, eh?

It's almost as if something important pertaining to the election has changed.

Like the identity of the Democrat nominee.


Real Clear Politics polling average: Trump vs. Harris (click image to enlarge)

Some who are not entirely clueless on the subject of polling claim -- with some justification -- that the reversals suffered by Trump and down-ballot Republicans lately do not necessarily mean that respondents have reconsidered whom they intend to vote for in November. But they proceed from that valid assertion to declare that the numbers have begun heading the wrong direction merely because the pollsters are "cooking the books" -- meaning that they have baselessly altered their underlying sampling schemes in various ways which appear to energize the left and demoralize the right. What these folks identify as the pollsters' motivation for this (e.g., "setting up the Democrat steal in November") descends back into boilerplate drivel territory, but regardless of motivation the dynamics of the upcoming election have changed and the forecasting models which are employed by pollsters therefore must also change.

Polling companies vary in their methods for determining the composition of the American electorate. They must make alterations in their samples regarding respondents' sex, race, political identification, geographic location, education level and a myriad of other factors whenever necessary. Some companies make subtle alterations (which can still be sufficient to generate significant movement); some companies make more blatant alterations; and some perhaps make none at all.

Like the captain of the Andrea Doria, there are pollsters who do not accept that danger lies ahead; they refuse to change course and continue on just as if conditions haven't changed. Continuing that analogy, there are a tiny number of pollsters including Rasmussen Reports which those on the right cling to like a life raft when everything else around them seems to be sinking. Like some others in the business, Ras will only reveal the recipe for their secret sauce for a price, so we can't determine whether the relatively happy (outlier) polling results which they continually provide -- for now -- are based on ignorance of reality or on something else which might be justifiable. Other life rafts for the right (or perhaps just flotsam and jetsam) currently include Fox News (!) and Trafalgar.



Outliers notwithstanding, most pollsters have recalibrated their surveys to reflect the fact that the presidential race is quite obviously not the same as it was prior to the Democrat coup which forced Biden's exit from the ticket. Still, there are those who refuse to accept that these recalibrations are necessary and instead see nothing but sinister motives for the changes.

Well then, let's turn this around 180 degrees and gauge the reaction. What would happen if the identity of the Republican candidate suddenly changed in mid-stream? What would happen if we dumped some lackluster presumptive nominee and switched over to our own "rock star"?


Photo credit: CNN

Picture, if you will, an alternate universe where Nikki Haley easily won all of the 2024 GOP primaries because she was unopposed except by some pissant candidate like whoever the Republican equivalent of Dean Phillips is (some alleged "moderate" who nobody's ever heard of).

Then the conservatives in the GOP stage a "coup" and force Haley out of the race in favor of Donald Trump, who wasn't even on the ballot in the primaries.

Now let's say the pollsters do not change their forecasting models, and therefore they show Trump doing no better than Haley against the Democrat, or perhaps doing even worse than her.

What would we be shrieking about then?


Photo credit: Ethan Hyman

Among other things, we'd be hearing:

"Pollsters are still 'oversampling' wimmen! But Trump is the nominee now and that's going to bring out more men as a percentage of voters! They need to account for that!"

"Trump is White and Haley is a minority (either Asian or black depending on whichever is most helpful at any given moment), but pollsters are still oversampling non-Whites!"

"They aren't acknowledging our exponentially-increased enthusiasm and that's the biggest factor of them all! They've tried to wave that off by claiming that Trump is just enjoying a brief 'honeymoon' period and they believe our enthusiasm will greatly diminish by November. No way, Jose! Trump really is a rock star -- just look at his rallies -- and our excitement is going to peak on election day! It's never going to wear off!"

"We grudgingly voted for Haley in the primaries because we had no other choice; she wasn't inspiring at all. Her support was a mile wide and an inch deep, and yet she wasn't faring too poorly in the polls against the Democrat. But now that Trump is our guy we have discovered the meaning of 'joy'! The pollsters still aren't budging and are refusing to accept what will surely be a dramatic turnout spike on the GOP side! We're pumped! We're stoked! We haven't seen passion like this since Ronald Reagan was running! We'd crawl over broken glass, blah blah blah...."

"In short, everything is different now, but these lying partisan Democrat pollsters haven't changed one thing. They truly live in a bubble!"

But not us. Definitely not us.

We don't like the polls but we can't change reality if we can't even bear to face reality. This election is far from over, but doing an impression of an ostrich from now until November is not the recommended way to try to achieve the best result for America.

Tags:

2024 Polls Alternate Universe