These are the logical fallacies we flag in every analysis on this site. Each one has a definition, an obvious example so the pattern is unmistakable, and a subtler version showing how it tends to appear in a video essay. Some of these fallacies go by different names in different contexts, and several can overlap with each other.
This list is not exhaustive, but it covers the patterns that appear most often in political video content, frequently used by creators who may not even recognise they are doing it.
HOW FALLACIES ARE CATEGORIZED
- Structural The logical structure is broken regardless of whether the facts are true.
- Evidence The facts cited are real but are used to create a false impression.
- Framing & Causation There is truth here but the conclusion goes further than the evidence allows.
Jump to: Genetic Fallacy | Strawman | False Dichotomy | Slippery Slope | Composition/Division | Circular Reasoning | Tu Quoque | Red Herring | Appeal to Personal Experience | Cherry-Picking | Hasty Generalization | Anecdotal Evidence | Moving the Goalposts | Survivorship Bias | Texas Sharpshooter | Equivocation | Misleading Framing | Appeal to Authority | Loaded Language | Appeal to Emotion | Overgeneralization | Post Hoc | Single-Cause | Nirvana Fallacy | Correlation vs. Causation
Structural Fallacies Structural
The logic is broken at the foundation. It doesn’t matter if every cited fact is accurate. The conclusion still doesn’t follow.
FALLACY #1
Judging the Argument by Where It Came From
(Genetic Fallacy)
Dismissing or accepting a claim based on its source instead of its content. Where an argument comes from tells you nothing about whether it is true.
Obvious example: “A tobacco company funded that study. You can’t trust it.” The funding source is worth noting. But it doesn’t make the data wrong. The results still need to be checked on their own terms if possible.
In a video essay: A creator spends thirty seconds on a critic’s political affiliation instead of engaging with their numbers. The implication is “these people funded it, so move on.” The data is never addressed. The source replaced the argument.
FALLACY #2
Misrepresenting the Other Side’s Argument
(Strawman Fallacy)
Replacing your opponent’s real argument with a weaker, easier version, then defeating that version. The original position is never actually refuted.
Obvious example: “Conservatives want to cut school funding.” “So you want children to be uneducated?” Nobody said that. The objection was about budget allocation, not education itself. The real argument is still standing.
In a video essay: A researcher argues that sugar consumption contributes to obesity. A critic responds: “So you want to ban all food enjoyment and control what people eat.” Nobody said that. The argument was about one dietary factor. The response attacked a position nobody made.
FALLACY #3
Pretending There Are Only Two Options
(False Dichotomy)
Framing a complex issue as having only two possible positions when more exist. Forces a choice between two poles that the creator constructed, not discovered.
Obvious example: “Either you support this policy completely, or you hate working people.” There is no third option offered, like supporting workers while opposing this specific policy. The middle ground is disappeared.
In a video essay: A video argues that since incremental policy reforms haven’t solved inequality, the only remaining option is systemic revolution. Stronger regulations, expanded public ownership, antitrust enforcement, and other structural reforms that fall between “tweak the system” and “overthrow it” are never acknowledged. The binary wasn’t discovered. It was built.
FALLACY #4
Assuming One Thing Leads to an Extreme Outcome
(Slippery Slope Fallacy)
Claiming one action will trigger a chain of increasingly extreme consequences, without showing why each step in that chain would actually occur.
Obvious example: “If we allow corporations to donate to political campaigns, they will eventually own every politician and democracy will become meaningless.” Each step in that chain requires its own evidence. The endpoint doesn’t follow automatically from the starting point.
In a video essay: A video argues that allowing any platform moderation will inevitably lead to total censorship of all dissent. The mechanism, why moderate content policies must escalate to full suppression, is assumed, not argued.
FALLACY #5
Assuming the Group Represents Every Individual
(Composition / Division Fallacy)
Assuming what is true of the whole applies to each part, or that what is true of each part defines the whole. Groups and their members are different things.
Obvious example: “This company is the most profitable in the country, so every employee there must be rich.” The company’s profit says nothing about any individual worker’s salary.
In a video essay: A video points out that several studies on a policy produced flawed results and concludes the entire body of research on that topic is unreliable. A few bad studies are used to dismiss a field of hundreds. The parts are doing the work of the whole.
FALLACY #6
Proving the Point by Assuming It Is Already True
(Circular Reasoning / Begging the Question)
Using the conclusion as one of the premises. The argument assumes what it is trying to prove. No matter how it sounds, no actual evidence has been offered.
Obvious example: “This is a safe neighbourhood because crime doesn’t happen here. How do you know crime doesn’t happen here? Because it’s a safe neighbourhood.” The conclusion and the premise are the same statement in different clothes. You haven’t moved anywhere.
In a video essay: A video argues that anyone who defends capitalism is “just defending their own class interests,” and uses that framing to dismiss every counter-argument without engaging with it. The dismissal only works if the premise is already true. But the premise is what is being disputed.
FALLACY #7
Deflecting by Pointing Out Someone Else Did It
(Tu Quoque / Appeal to Hypocrisy)
Responding to a criticism by pointing out that the critic is also guilty of the same thing. Hypocrisy is worth noting. But it doesn’t make the original criticism wrong.
Obvious example: “You can’t criticize drunk driving. You drove drunk once in college.” That past behaviour doesn’t make drunk driving acceptable. The criticism still stands.
In a video essay: A video challenging corporate tax avoidance gets countered with “your favourite politicians also took corporate donations.” That might be true, but it doesn’t address whether the tax avoidance was legal, ethical, or harmful. The subject changed from the accusation to the accuser.
FALLACY #8
Introducing an Unrelated Fact to Discredit the Other Side
(Red Herring / Guilt by Association)
When a separate and unrelated fact is brought into an argument to make someone look bad, rather than addressing the actual claim being made. The introduced fact may be true. It just has nothing to do with the question on the table.
Obvious example: During a debate about a company’s environmental record, someone says “before we talk about emissions, let’s remember that the CEO went through a messy divorce.” The divorce tells you nothing about the environmental record. It is introduced to make the CEO look bad so the audience is less willing to hear the defence.
In a video essay: A video challenges a researcher’s conclusions about immigration and wages. Rather than engaging with the data, it notes that the researcher once appeared on a podcast considered controversial. The podcast appearance says nothing about whether the wage data is accurate.
FALLACY #9
Using Personal Background to Make a Position Unchallengeable
(Appeal to Personal Experience / Credential Shield)
When someone uses their own lived experience or background to declare their position beyond challenge, rather than letting evidence do that work. Personal experience is relevant context. It is not a substitute for argument, and it does not settle questions about populations, institutions, or data.
Obvious example: “I grew up poor so I know exactly how poverty works and nobody who hasn’t lived it can tell me anything different.” Personal experience gives you insight. It doesn’t give you immunity from data that contradicts your conclusions.
In a video essay: A creator says “I’ve been researching this topic for five years, so when I tell you this is how it works, trust me.” Five years of research should produce five years of evidence. When the research is cited but the evidence isn’t shown, the credential is doing the argumentative work.
Evidence Fallacies Evidence
The facts cited are real. The problem is how they are selected, weighted, or arranged to create an impression the full picture doesn’t support.
FALLACY #10
Picking Only the Examples That Support the Point
(Cherry-Picking / Selective Evidence)
Presenting only the evidence that fits your conclusion while ignoring evidence that contradicts it. Every fact shown may be true. The lie is in what is left out.
Obvious example: A pharmaceutical company publishes the three trials that showed a drug worked and buries the fifteen that showed it didn’t. Every published result is accurate. The overall picture is false.
In a video essay: A video on healthcare privatisation shows three cases where private hospitals charged patients more than public ones. It never mentions countries where mixed systems produced better outcomes than fully public ones. The three cases are real. The selection is the problem.
FALLACY #11
Drawing a Big Conclusion from Too Few Examples
(Hasty Generalization)
Reaching a broad conclusion from a sample that is too small or not representative. The examples might be real. There just aren’t enough of them to support a universal rule.
Obvious example: “I’ve met three people from that country and they were all rude. Everyone from there is rude.” Three people out of millions isn’t a sample. It’s a handful of encounters turned into a rule.
In a video essay: A video on tech culture shows four founders from wealthy families, then concludes “success in tech requires inherited wealth.” No wider data is cited. The word “all” is doing work that four examples can’t support.
FALLACY #12
Using One Story to Prove a Universal Rule
(Anecdotal Evidence)
Treating a personal story or single case as proof of a broader pattern. Individual experiences are real. But they can’t stand in for evidence across a whole population.
Obvious example: “My grandfather smoked two packs a day and lived to 94. Smoking isn’t that dangerous.” One person’s outcome can’t override population-level health data. Individual results can deviate wildly from statistical risk.
In a video essay: A video uses the story of one factory worker who developed a serious illness after years of exposure to industrial chemicals as proof that the entire industry operates without safety standards. The story may be real and genuinely troubling. But one case can’t establish what is true across an entire sector.
FALLACY #13
Changing What Counts as Proof Mid-Argument
(Moving the Goalposts)
When a demanded standard of proof is met, the standard gets raised or changed. The target keeps moving so the evidence can never actually win.
Obvious example: “Show me one peer-reviewed study.” The study is provided. “That journal isn’t prestigious enough.” A better journal is found. “That was funded by the wrong group.” Each standard is met, then discarded. If no evidence could ever satisfy the bar, the bar isn’t about evidence.
In a video essay: A creator says mainstream economists support their view. When an opponent cites mainstream economists who disagree, the response is “those are establishment economists, we need grassroots voices.” The type of source that counts as valid changed the moment it stopped being useful.
FALLACY #14
Claiming Credit for Successes While Ignoring Failures
(Survivorship Bias)
Only analyzing cases that survived a selection process while ignoring all the cases that didn’t. The survivors look like proof of a pattern. The failures would change the picture entirely.
Obvious example: Every successful business owner interviewed says hard work and a good idea are all you need. But for every person who started a business and succeeded, several others worked just as hard with just as good an idea and ended up in debt. You never hear from them. The advice only comes from the survivors.
In a video essay: A video profiles ten successful small business owners as proof that “the market rewards hard work and vision.” The thousands of equally hard-working people whose businesses failed in the same period don’t appear. Ask who didn’t make it onto the list.
FALLACY #15
Choosing the Target After Seeing Where the Shots Landed
(Texas Sharpshooter Fallacy)
Drawing a pattern from data after the fact, finding a cluster, then declaring it a bullseye. The pattern wasn’t predicted. It was spotted after the fact and treated as meaningful.
This is related to Cherry-Picking, but with a key difference. Cherry-Picking means you deliberately hide evidence that contradicts your conclusion. The Texas Sharpshooter doesn’t hide anything. Instead, they look at all the data, notice a cluster that happens to support a point, and then declare that cluster to be the target. The problem isn’t what was hidden. It is that the bullseye was drawn around where the shots already landed.
Obvious example: A man fires randomly at a barn wall, finds the biggest cluster of holes, draws a target around it, and calls himself a sharpshooter. Any large dataset will contain clusters. The question is whether the cluster was predicted before the data was collected.
In a video essay: A video shows twenty years of wage data and highlights a five-year window where wages stagnated, presenting it as proof of a long-term trend. All twenty years are visible on the chart. But the five-year window was chosen because it fit, not because it was defined before looking.
Framing & Causation Fallacies Framing & Causation
There is partial truth here. The problem is the conclusion goes further than the evidence allows, through framing, loaded language, or bad assumptions about cause and effect.
FALLACY #16
Switching the Meaning of a Word Mid-Argument
(Equivocation)
Using a word with two different meanings as if both uses mean the same thing. The argument appears valid because the same word appears throughout, but the meaning quietly shifts.
Obvious example: “Only man is rational. No woman is a man. Therefore no woman is rational.” The word “man” shifts from “human being” to “male person” between the first and second premises. Swap in the correct definition at each step and the argument collapses.
In a video essay: A video argues “capitalism is just trade,” cites ancient markets as evidence humans are naturally capitalist, then critiques “capitalism” as a 19th-century industrial property system. Three distinct definitions are used interchangeably throughout.
FALLACY #17
True Facts Arranged to Create a False Impression
(Misleading Framing)
Using accurate facts but presenting them in a context or order that leads the audience to a conclusion the facts don’t actually support. Nothing said is technically false. The deception is structural.
Obvious example: A tabloid runs “Politician Seen Leaving Hotel at 2 AM.” The politician was at a fundraiser that ended late. Every word is true. The implication of scandal is not supported by a single fact in the article.
In a video essay: A video places accurate crime statistics directly after a description of a violent crime committed by an immigrant, with no transition or context. No false statement is made. The juxtaposition is designed to imply a connection that the data doesn’t show.
FALLACY #18
One Expert’s Opinion Settles the Debate
(Appeal to Authority)
Treating a claim as true because an authority said it, without examining the evidence behind it. Expert consensus matters. But a single expert’s opinion isn’t consensus, and authority in one field doesn’t carry over to another.
Obvious example: “This Nobel Prize-winning physicist says vaccines are dangerous. You can’t argue with a Nobel laureate.” Credentials in physics don’t confer expertise in immunology. The prize says nothing about the vaccine claim.
In a video essay: A video cites one well-known historian to settle a debate about whether a specific policy caused an economic collapse. That historian’s view is presented as “what historians agree on.” Several historians who studied the same event and reached different conclusions go unmentioned.
FALLACY #19
Winning by Defining the Terms in Your Favor
(Loaded Language / Definitional Retreat)
Embedding assumptions into the language itself so the conclusion follows from the definitions rather than from evidence. Once the loaded terms are accepted, the argument wins automatically.
Obvious example: A news anchor describes protestors as a “mob” while describing a counter-protest as “concerned citizens.” Both groups are doing the same thing. The word choice decided who was legitimate before a single argument was made.
In a video essay: A video consistently says “wage theft” instead of “unpaid overtime violations.” Both describe the same thing. The first embeds a moral verdict before any argument is made. The viewer accepts the framing before the evidence appears.
FALLACY #20
Emotional Weight Makes a Weak Argument Feel Strong
(Appeal to Emotion)
Using emotional impact to substitute for logical evidence. Emotional content isn’t inherently wrong. But when it is doing the work that evidence should be doing, the argument has a problem.
Obvious example: A political ad shows grieving families, a waving flag, and swelling music, then ends with the politician’s name. No policy position is stated. The emotional response is borrowed to create a positive association. The feeling of conviction is not the same as a reason to be convinced.
In a video essay: A video spends ten well-sourced minutes on historical suffering, then uses that emotional weight to push a present-day policy conclusion the earlier evidence never actually supported. The audience feels the connection. The logical link was never made.
FALLACY #21
Treating a Partial Pattern as a Universal Rule
(Overgeneralization)
Taking a real pattern that applies in some cases and declaring it applies in all cases. The original observation may be correct. The error is in the scale of the conclusion drawn from it.
Obvious example: “Politicians are always corrupt.” Some politicians have been corrupt. That doesn’t mean every politician in every country at every level is corrupt. The universal claim requires universal evidence.
In a video essay: A video documents real failures of financial deregulation, then concludes “deregulation always produces worse outcomes.” The examples are real. The conclusion applies them to sectors and contexts where that evidence doesn’t exist.
FALLACY #22
X Caused Y Just Because X Came Before Y
(Post Hoc Ergo Propter Hoc)
Claiming that because event A happened before event B, event A caused event B. Sequence in time is not the same as a causal relationship.
Obvious example: “I wore my lucky socks twice and we won the game both times. The socks work.” The win followed the socks. But wearing socks has no mechanism for determining game outcomes. Two things happening in sequence doesn’t mean one produced the other.
In a video essay: A video notes that inequality rose after a country adopted free trade agreements, then uses that sequence as proof the trade policy caused it. The sequence is real. But other major economic shifts happened in the same period, none of which are examined.
FALLACY #23
One Cause Assigned to Something with Many Causes
(Single-Cause Fallacy / Causal Reductionism)
Attributing a complex outcome to one cause when many contributing factors are likely at play. The identified cause might be real. The error is claiming it is the only one.
Obvious example: “The Roman Empire fell because of one bad decision an Emperor made.” That is one of many hypotheses. Military overextension, economic instability, administrative collapse, and external pressures are all established contributing factors. Complex events almost never have a single cause.
In a video essay: A video attributes declining Gen Z mental health entirely to social media. Social media may well be a factor. But economic anxiety, academic pressure, reduced physical activity, and changing diagnostic rates are all documented variables, none of which appear in the video.
FALLACY #24
Comparing a Real, Flawed Thing to a Perfect Ideal That Doesn’t Exist
(Nirvana Fallacy / Perfect Solution Fallacy)
Dismissing an existing system by comparing it to an idealized alternative that is never held to the same scrutiny. The real system is judged by its worst outcomes. The alternative is judged by its best intentions.
Obvious example: “The current medical system isn’t perfect, so we should replace it with one where wait times don’t exist, costs are zero, and every patient gets immediate personalised care.” The existing system is judged on its real flaws. The alternative exists only as a description of perfection, with no examination of how it would actually function.
In a video essay: “Capitalism produces poverty and exploitation. Under communism, resources would be distributed based on need.” The first half cites real, documented problems. The second half describes communism as it exists in theory, not as it has ever functioned in practice. One side gets the historical record. The other gets the brochure.
FALLACY #25
Two Things That Move Together Must Cause Each Other
(Correlation vs. Causation)
Treating a statistical relationship between two variables as evidence that one causes the other. Correlation measures whether two things rise and fall together. It says nothing about why.
Obvious example: Ice cream sales and drowning rates are positively correlated. Ice cream doesn’t cause drowning. Both rise in summer when it is hot and more people swim. A third variable, temperature, explains both.
In a video essay: A video shows that countries with higher union membership have lower poverty rates and concludes unions cause reduced poverty. The correlation may be real. But those same countries also tend to have stronger social safety nets and different political histories, none of which are controlled for.
Questions to Ask Yourself
These checklists are not about rejecting every argument that cites an expert or uses emotion. They are about noticing when those things are doing work that evidence should be doing instead.
Structural Fallacies Structural
- Does the conclusion actually follow, or did I just accept it because the facts sounded right?
- Is this attacking the real argument, or a version of it nobody actually made?
- Are there other options the video isn’t offering me?
- Is this saying something bad will definitely happen, or just that it could?
- Is the criticism being answered, or is the critic being attacked instead?
- Is the speaker’s personal background being used to make their position unchallengeable rather than letting the evidence do that work?
Evidence Fallacies Evidence
- How many examples are there? Is that enough to make a universal claim?
- What is left out? Are there cases that would complicate this?
- Who produced this data and why might they have an interest in the conclusion?
- How is the source described? “Leading expert,” “award-winning,” and “renowned” are praise, not credentials. What are they actually an expert in?
- Were these examples chosen before or after the creator knew what point they wanted to make?
- Is this one person’s story standing in for an entire population?
Framing & Causation Framing & Causation
- Are words like “always,” “never,” or “inevitable” doing work that the evidence doesn’t support?
- Is there music or emotional imagery alongside the conclusion? Does the argument hold without it?
- Did two things happen at the same time, or did one actually cause the other?
- Is the language neutral, or has a verdict already been built into the words?
- Is the alternative being compared a real system with a real track record, or an ideal that has never been tested?
- Is one cause being named for something that clearly has many?
Questions to Ask Yourself
These checklists are not about rejecting every argument that cites an expert or uses emotion. They are about noticing when those things are doing work that evidence should be doing instead.
Structural Fallacies Structural
- Does the conclusion actually follow, or did I just accept it because the facts sounded right?
- Is this attacking the actual argument, or a weaker version of it?
- Are there other options the video isn’t offering me?
- Is this saying something bad will definitely happen, or just that it could?
- Is the information being critiqued, or is a person being attacked instead?
Evidence Fallacies Evidence
- How many examples are there? Is that enough to make a universal claim?
- What is left out? Are there cases that would complicate this?
- Who produced this data and why might they have an interest in the conclusion?
- How is the source described? “Leading expert,” “award-winning,” and “renowned” are praise, not credentials. What are they actually an expert in?
- Is this one person’s story being used an example for an entire group?
Framing & Causation Framing & Causation
- Are words like “always,” “never,” or “inevitable” doing work that the evidence doesn’t support?
- Is there music or emotional imagery alongside the conclusion? Does the argument hold without the ‘feelings’?
- Did one thing actually cause the other or did two things happen at the same time that could have been caused by something else?
- Is the language neutral, or ‘leading’ which has a verdict already been built into the words?
- Is the alternative being compared to a real system with a real track record, or an ideal system that is otherwise perfect?
- Is only one cause being named for something that can have many causes?