What Fallacy Is This?

Find a fallacy by what happened. Pick the description that fits, and the entry it points to has the name, a worked example, and a short link you can paste. Some descriptions lead to more than one entry. The line under each name says which is which.

The reply is about the person

The argument is still on the table. The response went for whoever made it.

  • Attacking the person instead of the argument

    Also described as: insulting someone, attacking someone's character, name calling

    • Ad Hominem Logical fallacy

      The claim is rejected because of who made it, not because of what it says.

  • Pointing out that the critic does it too

    Also described as: accusing someone of hypocrisy, "you also", "you do it too"

    • Tu Quoque Logical fallacy

      The critic's own behaviour is offered as the reason the criticism is wrong.

    • Whataboutism Rhetorical device

      The reply raises a different wrong, by the critic or their side, and never returns to the first one.

  • Dismissing a claim because of where it came from

    Also described as: judging an idea by its source, guilt by origin

    • Genetic Fallacy Logical fallacy

      The source or history of the idea is the whole case against it.

  • Explaining someone's behaviour by their character

    Also described as: assuming bad intent, blaming personality over circumstances

  • Turning the accusation back on the accuser

    Also described as: playing the victim, reversing the roles

    • DARVO Debate tactic

      The response is to deny it, attack the accuser, and claim to be the injured party.

  • Treating a denial as proof

    Also described as: denying it proves it, an accusation that cannot be answered

    • Kafkatrap Debate tactic

      Saying no is counted as more evidence for the accusation.

The subject changed

The original point never gets answered, because the conversation is now somewhere else.

  • Changing the subject

    Also described as: changing the topic, diverting attention from the issue, dodging the question

    • Red Herring Logical fallacy

      A new topic is introduced that has no bearing on the point being argued.

    • Whataboutism Rhetorical device

      The new topic is somebody else's wrongdoing.

  • Changing the argument partway through

    Also described as: shifting position when challenged, changing what the claim was

    • Motte and Bailey Debate tactic

      A bold claim is swapped for a safer one under pressure, then swapped back later.

    • Moving the Goalposts Debate tactic

      The evidence asked for is supplied, and a harder demand replaces it.

    • Equivocation Logical fallacy

      A key word quietly changes meaning between one step and the next.

  • Asking for more and more evidence

    Also described as: never satisfied with the proof, "just asking questions"

    • Sealioning Debate tactic

      The requests are polite and endless, and the aim is to wear the other person out.

    • Moving the Goalposts Debate tactic

      Each demand that gets met is replaced by a new one.

  • Making too many points to answer

    Also described as: overwhelming someone with claims, rapid-fire arguments

    • Gish Gallop Debate tactic

      The claims arrive faster than anyone could check them, and any left unanswered are counted as wins.

  • Ending the discussion with a stock phrase

    Also described as: "it is what it is", "agree to disagree", shutting down the conversation

The position got distorted

Something is being argued against, but it is not quite what the other side said.

  • Misrepresenting or exaggerating someone's argument

    Also described as: distorting the argument, arguing against something nobody said, using an extreme version

    • Straw Man Logical fallacy

      A weaker or more extreme version stands in for the real position and gets knocked down.

  • Presenting only two options

    Also described as: either-or thinking, black and white thinking, false dilemma

    • False Dichotomy Logical fallacy

      More choices exist than the two on offer.

  • Asking a question with an accusation built in

    Also described as: a trick question, a question you cannot answer without admitting something

    • Loaded Question Logical fallacy

      Any direct answer concedes a point that was never established.

  • Making a false comparison

    Also described as: comparing two things that are not alike, "both sides are the same"

    • False Equivalence Logical fallacy

      Two things are treated as the same despite a difference that matters to the argument.

    • False Balance Rhetorical device

      Two positions get equal weight although the evidence behind them is lopsided.

  • Assuming the truth is somewhere in the middle

    Also described as: splitting the difference, compromise as proof

    • Middle Ground Logical fallacy

      The halfway point is taken as correct only because it sits between two positions.

The evidence is thin or selective

There is evidence. It is too little, too hand-picked, or pointed the wrong way.

  • Generalizing from one example

    Also described as: jumping to conclusions, generalizing from a few cases, using an extreme example

    • Hasty Generalization Logical fallacy

      A broad rule is drawn from a sample too small or too unusual to carry it.

    • Small Sample Size Statistical fallacy

      The striking result comes from a small group, where chance alone produces extremes.

  • Using personal experience as proof

    Also described as: using your own experience, "it worked for me", arguing from a story

    • Anecdotal Evidence Logical fallacy

      One story is asked to settle a question about what generally happens.

  • Ignoring evidence that does not fit

    Also described as: using incomplete evidence, only showing one side of the data

    • Cherry Picking Logical fallacy

      Supporting evidence is presented and the rest is left out.

    • Confirmation Bias Cognitive bias

      It is not deliberate. Agreeable evidence simply gets noticed and trusted more.

  • Looking only at the successes

    Also described as: ignoring the failures, learning only from winners

    • Survivorship Bias Cognitive bias

      The cases that did not make it are missing from the picture.

  • Saying it is true because nobody has disproved it

    Also described as: absence of evidence, "you can't prove it's false"

The cause is assumed

Two things are linked. The claim is that one produced the other.

  • Assuming that association is causation

    Also described as: correlation and causation, assuming one thing caused another

    • Correlation Is Not Causation Statistical fallacy

      A statistical link is read as cause without ruling out confounders, reverse causation, or chance.

    • False Cause Logical fallacy

      The general mistake of inferring cause from a link or a sequence alone.

  • Assuming it caused it because it came first

    Also described as: "after this, therefore because of this", blaming whatever happened before

    • Post Hoc Logical fallacy

      Order in time is the only evidence that one thing produced the other.

  • Predicting a chain reaction from one step

    Also described as: "where does it end", "next thing you know", the thin end of the wedge

    • Slippery Slope Logical fallacy

      Each link in the chain is asserted, and none is shown.

  • Crediting a change that would have happened anyway

    Also described as: a result that returned to normal, improvement after a bad run

    • Regression to the Mean Statistical fallacy

      An extreme result was followed by an ordinary one, and something else got the credit or the blame.

  • Expecting a streak to even out

    Also described as: "it's due", believing luck has to turn

    • Gambler's Fallacy Cognitive bias

      Past random outcomes are expected to change the odds of the next one.

An appeal stands in for a reason

Something persuasive is offered. It is not evidence for the claim.

  • Appealing to emotion

    Also described as: using emotion instead of facts, playing on fear or pity

    • Appeal to Emotion Logical fallacy

      A feeling is doing the work an argument about the facts should do.

    • FUD Rhetorical device

      The feeling is vague fear about the alternative, with no case made for your own option.

  • Saying it is true because everyone believes it

    Also described as: relying on popularity, appeal to the people, "everyone knows"

    • Bandwagon Logical fallacy

      The number of believers is offered as the evidence.

  • Saying it is true because an authority said so

    Also described as: using authority as proof, "because I said so", "experts say"

    • Appeal to Authority Logical fallacy

      Status replaces the evidence, or the authority is speaking outside their field.

  • Saying it is good because it is natural

    Also described as: chemical-free, natural versus artificial

    • Appeal to Nature Logical fallacy

      Natural is treated as proof of safe or good, and unnatural as proof of harm.

  • Saying it is right because it has always been done this way

    Also described as: "we've always done it this way", comparing an old time to now

    • Appeal to Tradition Logical fallacy

      The age of a practice is the entire justification for it.

The claim cannot lose

Whatever comes up, the conclusion survives. That is a design feature, and a warning sign.

  • Assuming the conclusion

    Also described as: circular reasoning, begging the question, assuming the answer

    • Circular Reasoning Logical fallacy

      The support for the claim is the claim again, in other words.

  • Redefining the group to exclude the counter-example

    Also described as: "no real one would do that", changing the definition

    • No True Scotsman Logical fallacy

      A case that breaks the rule is declared not a true member of the group.

  • Making an exception for your own case

    Also described as: a double standard, "that's different"

    • Special Pleading Logical fallacy

      A standard applies to everyone else, and no principled reason is given for the exemption.

  • Rejecting a fix because it is not perfect

    Also described as: the perfect solution, "it won't solve everything"

    • Nirvana Fallacy Logical fallacy

      A workable option is compared against an ideal that is not actually available.

  • Rejecting a conclusion because the argument for it was bad

    Also described as: pointing out a fallacy as proof, "that's a fallacy, so you're wrong"

    • Fallacy Fallacy Logical fallacy

      A flawed argument is taken to show that its conclusion is false.

The wording is doing the work

The facts are the same either way. The words are chosen to settle the question early.

  • Using a word in two different senses

    Also described as: using different definitions, playing on a double meaning

    • Equivocation Logical fallacy

      The argument only holds if nobody notices the word changed meaning.

  • Using emotionally charged words

    Also described as: biased wording, slanted language

    • Loaded Language Rhetorical device

      The word choice delivers the verdict before any argument is made.

  • Using softer words for something harsh

    Also described as: sugarcoating, corporate language

    • Euphemism Rhetorical device

      The softened wording hides what is actually being described.

  • Using vague phrases that commit to nothing

    Also described as: "some people say", "studies show", "many believe"

    • Weasel Words Rhetorical device

      A claim is implied, and the speaker stays free to deny having made it.

  • Using coded language

    Also described as: a hidden message, saying it without saying it

    • Dog Whistle Rhetorical device

      The phrase sounds harmless in general and carries a second meaning for one audience.

  • Making someone doubt their own memory

    Also described as: "that never happened", denying what was said

    • Gaslighting Rhetorical device

      The denial is sustained, and its aim is to make the other person distrust their own recall.

Your own judgment is tilted

Nobody is arguing in bad faith here. The mind takes a shortcut, and the shortcut leads somewhere wrong.

  • Continuing because of what has already been invested

    Also described as: being in too deep, "we've come this far", throwing good money after bad

    • Sunk Cost Fallacy Cognitive bias

      Past spending, which cannot be recovered, is driving a decision about the future.

  • Sticking to the first number mentioned

    Also described as: the opening offer, the original price

    • Anchoring Cognitive bias

      A starting figure pulls later estimates toward it, even when it was arbitrary.

  • Saying it was obvious all along

    Also described as: "I knew it", being wise after the event

    • Hindsight Bias Cognitive bias

      Knowing the outcome makes it feel as though it was predictable from the start.

  • Judging how likely something is by what comes to mind

    Also described as: fear of rare events, overestimating what is in the news

  • Overestimating your own competence

    Also described as: a little knowledge, not knowing what you do not know

    • Dunning-Kruger Effect Cognitive bias

      Limited skill in a field comes with limited ability to see the limits.

The numbers are real, the reading is wrong

Nothing is made up. The figure is accurate and the conclusion drawn from it is not.

  • Quoting an average that hides the spread

    Also described as: mean versus median, the average person

    • Misleading Average Statistical fallacy

      A skewed distribution means the mean describes almost nobody in the data.

  • Ignoring how rare something is to begin with

    Also described as: false positives, misreading a test result

    • Base Rate Fallacy Statistical fallacy

      A result is read without accounting for how common the thing being tested for is.

  • Applying a group statistic to one person

    Also described as: stereotyping from averages, judging an individual by their group

    • Ecological Fallacy Statistical fallacy

      A number that describes a population is used to describe one member of it.

  • A trend that reverses when the groups are combined

    Also described as: pooled data telling a different story

    • Simpson's Paradox Statistical fallacy

      The pattern holds inside every subgroup and flips in the combined total.

  • Finding the pattern first and the hypothesis second

    Also described as: drawing the target after shooting, seeing patterns in noise

    • Texas Sharpshooter Statistical fallacy

      A pattern spotted in the data is presented as the thing that was being tested.

    • P-Hacking Statistical fallacy

      Many analyses were run, and only the one that crossed the significance line is reported.

  • A link created by how the sample was chosen

    Also described as: selection effects, "all the good ones are taken"

    • Berkson's Paradox Statistical fallacy

      Selecting on two traits at once makes them look negatively related inside the selected group.

  • Counting the visible benefit and ignoring the hidden cost

    Also described as: opportunity cost, "it creates jobs"

    • Broken Window Fallacy Logical fallacy

      What the same resources would otherwise have done is left out of the sum.

Before you paste the link

A name is a starting point. Every entry has a section on when the pattern is not a fallacy, and it is worth thirty seconds before you cite one. A critic's record is sometimes relevant. Citing an expert is usually fine. Some slopes really are slippery.

Finding a fallacy in an argument does not make its conclusion false. It means that argument does not get there. Treating the label as a verdict has its own entry: the fallacy fallacy.

If the description that fits is something you said, the entry is written for you. Each one ends with how to fix the argument, and the repair is usually smaller than it sounds.

If nothing here fits, the full list is grouped by category, and the quiz is a quick way to learn the common patterns.