Live Fact Checking is Journalistic Malpractice

September 2025

A large swath of the public, mostly Republicans, don't trust the mainstream news media. They believe it's biased.

So you would think that Network and Cable News outlets would cease and desist from displays of hubris that confirm the public's worst suspicions.

Case in point: live fact checking during televised presidential debates.

This practice debuted in 2012 when CNN's Candy Crowley "corrected" a statement by Mitt Romney during a Town Hall with President Obama.

Unfortunately her attempt was more distracting than enlightening, and it evoked a firestorm of criticism. The controversy led the media elite to concede that live fact checking is NOT an appropriate role for debate moderators.

For the next 11 years debate moderators curbed their urge to dispute the fibs of politicians.

But by March 2024 the verdict of the Republican primaries was in: the orange charlatan clinched the opportunity to lie and slander his way into a second Presidential term. The prospect that the debate stage would be dominated by his torrent of delusional rhetoric was too frightening for left-leaning newsrooms to endure.

So, on September 10, 2024, ABC's moderators for the Trump-Harris debate resurrected live fact-checking for the noble cause of leveling the playing field in favor of the truth.

Obviously, this intervention effects the viewers' perception of the candidates' credibility. Every correction issued by the moderator erodes a candidate's credibility, while emerging from the debate completely unscathed affirms it. So the integrity of the debate depends crucially on the accuracy and fairness of live fact checking.

The problem is, the execution of fact checking in real time is prone to gross random errors as well as personal bias; and because of this, it is more likely to distort the perception of the candidate than sharpen it. In short, for live fact checking to be reliably accurate and unbiased, the moderator must approximate omniscience.

Here's why...

A moderator who screens verbiage for factual accuracy is like a radiologist who screens for breast cancer using mammography. Both are plagued by two types of error: false positives and false negatives.

Mammogram screening is highly accurate because the rate of false positives and false negatives is incredibly low, 12% and 1.5% respectively.

By contrast, live fact-checking is debilitated by incredibly high error rates, especially false negatives. That's because the semantic memory capacity of even the smartest fact-checker is insufficient to ensure accuracy.

False Negatives = False Statements that are Not Called Out.

Live fact checking results in a false negative when a candidate lies, but the moderator fails to catch and correct it.

Even the smartest moderator is highly likely to miss a false statement for an obvious reason: limited semantic memory.

The facts immediately accessible from semantic memory are a miniscule fraction of the corpus of factual knowledge related to the topics covered in the debate. That's why modern humans rely on external repositories of knowledge, like books. And that's why professions are increasingly specialized. And that's why it's stupid to expect TV personalities to be experts on every issue. In short, on the scale of omniscience, debate moderators score close to zero.

For example, in the 2012 debate Obama was never corrected by Crowley. Likewise, in the 2024 debate the moderators badgered Trump, while Harris got off scot-free. However, both Obama and Harris actually committed several Pinocchios, as revealed by rigorous fact-checking after the debates. They got away scot-free not because they were truthful, but because of they weren't caught.

This illustrates live fact checking's susceptibility to bias. The viewers' perception of the Democrat candidates was enhanced by false negatives. By disputing Romney and Trump, but leaving Obama and Harris unscathed, the moderators implicitly affirmed the Democrats' credibility. Rampant false negatives made them look like models of truthfulness.

Live Fact Checking is Vulnerable to Bias

Over the course of a debate the inaccuracies of live fact checking inevitably end up tilting the score in favor of one candidate over the other. Whether the tilt benefits the left or the right is partly an artifact of random failures to detect a lie when it occurs, i.e., luck.

However, that tilt also results from systematic bias. A moderator's political ideology effects the composition of her semantic memory. (She is more likely to remember MeToo complaints against Trump than Biden's sniff-and-touch record). Therefore she is primed and prepped to snare the fibs of one side more than the other.

Let's face it, the resurrection of fact checking in 2024 was instigated by a smidge of Trump Derangement Syndrome among liberal media professionals.

False Positives : the Ambiguity of Facts

A false positive occurs when the moderator falsely corrects a candidate's assertion. This means that the proffered correction is not supported by empirical evidence from a reputable and unbiased source. Rather, the correction is unverifiable, irrelevant, ambiguous or plainly false.

Candy Crowley was a victim of poor quality evidence. Her correction of Romney was bungled because, without the actual transcript of Obama's press conference at hand, her ad-lib from memory slightly misquoted what he said. After the debate she admitted that Romney “was right in the main, I just think that he picked the wrong word.”

In 2024, moderator David Muir challenged Trump's rant about cat-eating Haitians in Springfield with a statement from the City Manager, "there have been no credible reports...of pets being harmed, injured or abused by individuals within the immigrant community."

This evidenced is flawed in two ways.

First, it obviously violates a basic principle of logic: "the absence of evidence is not evidence of absence."

Second, the evidence is not from a neutral third party, but an official who has a interest in burnishing the image of the community. And Trump saw it that way when he retorted "maybe that's a good thing to say for a city manager."

Trump then countered with contrary evidence, "I've seen people on television say my dog was taken and used for food" to which Muir responded "I'm not taking this from television. I'm taking it from the city manager." (Wow, this prominent TV-news personality advised the TV audience not to trust what you hear on TV).

Since Muir had already challenged Trump on immigration, Harris was able to sidestep the toxic topic when she was asked to respond. Instead she recited a script about Trump's unpopularity among notable Republican and former staff.

The point is, Trump had to navigate a tag-team format in which he had to debate several opponents - Harris plus the moderators. The problem is, the tag-team format was never unleashed on Harris.

For example, when asked by Linsey Davis "why have so many of your policy positions changed," Harris addressed only one of the many flip-flops, and then proceeded to evade the question by bloviating on an entirely different subject. Davis never tried to corral her back to the specifics of the question asked. Yet, Davis was willing to badger Trump for not answering a direct question about abortion. That kind of fact-checking is legitimate, since the failure to answer a question is a verifiable fact.

The most glaring false positive occurred when Muir attempted to dispute Trump's tirade on the flood of criminals crossing the border. Referring to an update on crime rates from 2023 through September '24, Muir stated "as you know, the FBI says overall violent crime is coming down in this country."

Apparently, Muir was completely unaware that the crime rate statistics he was waving actually tend to support Trump's contention, not refute it. For instance, the decline from '23 through '24 corresponds to a slowdown in border crossings, thanks to Biden's reversal on immigration policy. But during the surge in immigration from 2021 through 2023, the violent crime rate increased, as did the rate of property crimes.

However, these official statistics on crime rates are irrelevant anyway because they don't directly address Trump's argument.

His argument, minus the bombast and slurs, is that a significant percentage of unvetted immigrants are "bad people." So, when Biden's policies generated a flood unauthorized immigrants, more criminals landed on America's soil. Muir might have disputed Trump's exaggerations of this fact, but it's essential truth is indisputable.

For example, the data on felony convictions show that 8% of American adults are certified criminals. If that percentage also holds for the 4 million additional border crossings under the Biden administration, then America's criminal population will have increased by 320,000 (0.08 x 4 million). In this example, immigration would have no effect on the overall crime rate, but the additional criminals would certainly increase the number of crimes.

The Takeaway.

Live fact checking is journalistic malpractice. It is a thumb-on-the-scale scheme disguised as a public service. The moderator intrudes on the debate as a co-debater pretending to be a neutral arbiter of the truth. But no human moderator possesses the knowledge and objectivity needed for live fact checking to be reliably accurate. So the moderator becomes an instrument of bias.

Crowley's attempt fact-checking in 2012 was a case of impulsive stupidity. But, in 2024, the practice was resurrected because a left-leaning team at ABC decided that the audience needed protection from the tyranny of Trump's mendacity. Fair and balanced, this is not. So Donald was forced to debate three opponents: Harris plus two pretentious moderators. This is bias regardless of the magnanimous intentions.