“They didn't just change the tack to put Hillary ahead[over Donald Trump]. They went back and changed the outcome,” the late Democratic pollster and strategist Pat Caddell said in 2016. “They didn't tweak the process, they fudged it.” Caddell wasn't talking about an unnamed news organization when he made that comment. He was talking about Reuters, the world's second-largest media company.
Caddell added that he had never seen a journalism organization “behave so dishonestly” in his life, a question that arises given that Reuters is also ranked the “second most trusted news brand.”
What would Caddell say about polls today that show Kamala Harris leading Trump in the 2024 presidential election? Can we trust the polls? It's worth noting that until very recently, Harris was one of the least popular vice presidents in American history.
One person who knows his position well is commentator John Kudla. “Stop Worrying About the Polls: Kamala's Lead is Fake,” he said in a headline on Sunday.
Currently, Harris is running “above” the RealClearPolitics (RCP) polling average by 1.8 points, and ahead by FiveThirtyEight (FTE) by 3.2 points. Kudla says Harris could win, but only if the public buys into the mainstream media's propaganda that portrays Trump as the problem and Harris as the novel solution. (Of course, it's not unheard of for people to fall victim to propaganda; it's not surprising that the low-information stratum of the electorate is referred to as the “idiot vote.”)
Smoke and mirrors
Kudla begins by pointing out that Trump still has a “secret weapon”: the large number of “non-voting” Trump supporters. They are in this state because, as with politically incorrect positions, many Americans are unwilling to admit that they support Trump. They are a hidden but very real electorate.
As for the actual poll results, let's start by considering RCP's 1.8-point lead. It's worth noting here that only three of the 11 polls used “targeted voters” (LV); the other eight asked “registered voters” (RV). Historically, LV surveys are more accurate. Additionally, some of these polls were conducted before or on the day Robert F. Kennedy Jr. endorsed Trump (August 23). Thus, the RCP average cannot fully reflect the impact of that endorsement.
As for Harris' 3.2-point lead among FTEs, Kudla is downright baffled. For starters, many of the polling agencies cited are no-name names, “including Big Village, Kaplan Strategies and Angus Reid Global,” Kudla writes. “And there's a jumble of polls of adults, registered voters and likely voters, many of which still view RFK Jr. as a separate candidate.”
And how unbiased are these polls? After all, it's easy to artificially inflate support for Harris — just “sample more Democrats, more women, more minorities, or sample more in metropolitan areas,” Kudla points out. And in fact, oversampling of Democrats is common.
Not just incompetent
Kudla then gives several examples of book tampering.
According to a RedState article, a recent New York Times/Siena poll was allegedly biased. The poll showed Kamala with sizable leads in Michigan, Wisconsin and Pennsylvania, but those leads were questionable at best…
Another potential accuracy issue is that most polling agencies are ideologically left of center. Could they be purposely inflating their numbers to help Kamala? According to pollster John McLaughlin, this appears to be happening.
“So what they're doing is undermining the polling of Republican voters. They're disproportionately polling voters who supported Biden in the 2020 election. It's ridiculous. So what they're doing is trying to pump up Harris. They're trying to suppress our vote. And, you know, this is something smart people are doing, so I think it's intentional.”
This is not a fantastical conclusion. Remember, most pollsters are passionate about polling for the same reason musicians are passionate about music: passion. They are political creatures, and as Kudla points out, they tend to lean left.
History repeats itself
This isn't new either. Consider that Democrat Rep. Caddell, mentioned at the beginning of this article, expressed astonishment at the Reuters poll's manipulation. “They changed the result by nine points,” Breitbart quoted Caddell as saying in 2016. “'This is what the media is trying to do to get her elected,' Caddell said,” the site continued. “'This poll is just part of a media offensive.'”
The offensive continues, and some of it is a kind of “jumping the shark” polling equivalent. For example, Kudla cites a July 28 Bloomberg/Morning Consult five-way Michigan poll showing Harris leading by 12 points. He also highlights the aforementioned New York Times poll, which gives Harris a 5-point lead. “If you remove both polls from the average,” Kudla argues, “it shrinks from 2.3 points to just 0.4 points, effectively making it a tie.” (Note: He does not specify which average he is referring to.)
Are you talking to the wrong person?
Another relevant factor is that Democrats outnumber Republicans by 3-4 percentage points in the US. But this is due to the very liberal and populous states of New York and California. This matters because opinion polls are a measure of the popular vote, whereas presidential elections are decided by the electoral vote. Apart from California and New York, the Trump-Harris race is a draw.
Finally, Kudla makes one more point, writing:
The real cause of Kamala's polling surge may be response bias. Educated Democrats are three to four times more likely to respond to polls than non-college educated Democrats, according to Mark Harris of the political consulting firm ColdSpark. They also over-survey high-turnout voters. Trump tends to get more support from working-class Democrats and low-turnout voters. Harris also argues that “historic response bias in surveys is what's causing big polling mistakes this fall.”
National surveys of likely voters overestimated Democratic support by 1.3 percentage points in 2016 and a whopping 3.9 percentage points in 2020.
In short, the polling business is not about truth but about marketing, the drug called leftism packaged in a Democratic Party package.
Protected from stupidity, they became fools.
Part of the problem is also a lack of accountability. Think about it: after Reuters broke in in 2016, shouldn’t this organization have been shunned? Shouldn’t RCL, FTE and others have said, “We have taken your manipulation into consideration and will not be quoting you again until further notice”? This would, first of all, be the right thing to do, and would likely improve polling averages as well. But just as importantly, it would also be a deterrent to misconduct. Instead, the message sent is that deception pays.
Which brings me to another possible, rather sinister, reason for poll propaganda: “When Democrats cheated to get across the finish line, the Democratic candidate needs a credible victory story,” asserted the top commenter on Kudla's piece. “The polls are all about cheating.”
In fact, the “flaws” in the current polls may actually help reflect whether election results include voter fraud. And for those who say this is MAGA delusion, know that Democrats don't agree. Former Hamtramck Mayor Karen Majewski, a Democrat, said after her loss in the 2021 election that she is “absolutely certain” her office was stolen.
Oh, and by the way, it was allegedly stolen by other Democrats.
