Cherry-picking
Every fact they cited was true. The picture they built was false.
Also known as Selective evidence / card stacking
Selecting the evidence that supports a conclusion and leaving out the evidence that does not. Unlike lying, cherry-picking uses only true statements, which is what makes it durable and hard to challenge. A graph is truncated to show a rise that reverses over the full period. A quote is lifted from a paragraph that contradicts it. A string of successes is listed while the larger string of failures is omitted. The audience is given facts and led to a falsehood, and the gap between the two is built entirely from what was left on the cutting-room floor. The deception is in the selection, and the selection is invisible from inside the presentation.
Truth-adjacency
Truth-adjacent: the pattern's significance depends on whether the claim is true
Where it shows up
Public discourse
What to watch for
The phrases and tells that mark this pattern in the wild:
a claim supported by real facts that still feels misleadinga graph whose axes start conveniently close to the interesting parta quote that seems to say something its source does nota list of examples that are all in one direction, presented as if they are all of thema conclusion that survives only as long as you do not see the full dataset How to recognize it
The tell is the relationship between the shown facts and the hidden ones. Cherry-picking is invisible in the presented material, because the presented material is true. It becomes visible only when you ask what is not there. Watch for conclusions that feel larger than their evidence, graphs whose axes or windows do the persuading, and lists offered as representative that are all pointed one way. The deeper signal is the presenter’s relationship to the full dataset: a honest summarizer can tell you what they left out and why, and their conclusion survives the showing. A cherry-picker needs the omitted half to stay omitted, and the conclusion collapses the moment the full picture is laid down beside the curated one.
What to ask
- What is not being shown? The presented facts are only half the evidence. The other half is whatever was available and not included. Ask what the full dataset, the full quote, the full time period says.
- Would the conclusion survive the full picture? If showing the rest reverses or materially weakens the claim, the selection was doing the lying, and the truths were only the material.
- Is the selection representative, or one-directional? Everyone emphasizes. The difference is whether the emphasized examples are a fair sample or a curated string. If every chosen example points one way and the presenter implies they are typical, check whether the unchosen ones point the other.
What it looks like when you’re wrong about it
You call “cherry-picking” on a person who summarized, led with their strongest evidence, and emphasized the facts that support a position they genuinely hold, but whose conclusion would survive the full dataset being shown. All communication selects; exhaustiveness is impossible. The pattern requires the selection to be designed to produce a false impression: the omitted evidence materially reverses the conclusion, the selection is one-directional and presented as representative, and the full picture is withheld because it is fatal. If the presenter can show you the rest and still stand on their claim, you are looking at emphasis, not the pattern.
What it feels like from the inside
- Like the evidence is overwhelming, but you cannot quite say why it feels dishonest.
- Like the conclusion is bigger than the facts shown.
- Like asking 'what else is there?' would be rude or unnecessary.
How it starts
A presenter leads with the strongest facts for their position. The audience sees true statements and assumes the conclusion follows. The omitted evidence, which would reverse or weaken the conclusion, is never mentioned.
How it progresses
- Graphs get truncated: axes start close to the interesting part, time windows exclude the reversal.
- Quotes get lifted from paragraphs that contradict them.
- Lists of examples all point one direction, presented as if they are all of them.
- The conclusion grows more confident as the curation gets more aggressive.
Common signs
- Every cited fact is true. The picture built from them is false.
- Graphs whose axes start conveniently close to the interesting part.
- Quotes that seem to say something the broader source does not.
- The presenter cannot tell you what they left out and why.
Why it's hard to leave
Because every fact shown is real. Challenging the conclusion feels like denying the facts. The deception is in the invisible half, and you cannot argue against what you have not been shown.
Do this now
- Ask: what is the full dataset? The shown facts are only half the evidence.
- Check the axes. If the graph's window does the persuading, the data is being used as a prop.
- Ask the presenter: what did you leave out and why? An honest summarizer can answer. A cherry-picker needs the omission to stay invisible.
What people realize later
Later, people realize the lie was not in the facts. It was in the gap between the facts shown and the facts available. The presentation was a curated exhibit, and the conclusion only survived as long as the exhibit stayed closed.
Recognized this in public information?
This pattern in the wild
Field notes where this pattern was identified:
Misuse Guardrails
How this pattern gets misused
Someone dismisses any selective presentation as cherry-picking, treating the ordinary and necessary act of choosing which facts to lead with as if it were deception. All communication selects. The term becomes a way to demand impossible exhaustiveness and to frame any framing as dishonest, which makes it harder to identify the cases where the selection is genuinely designed to mislead.
What it looks like when you're wrong about it
Choosing which facts to lead with, summarizing, and emphasizing the strongest evidence for a position you genuinely hold is normal communication, not cherry-picking. The pattern requires the selection to be designed to produce a false impression: the omitted evidence materially reverses or undermines the conclusion, the selection is one-directional rather than representative, and the full picture is withheld rather than merely unmentioned. If the presenter's selection could be corrected by showing the rest and the conclusion would survive the full data, it is emphasis, not the pattern.
Not sure? Describe the situation to someone outside it. If they do not see the pattern, pause before you name it.
Related Patterns
Context narrowing
You looked where the light pointed. You did not notice the rest of the room going dark.
Source obfuscation
You trusted it because it sounded official. 'Official' was the costume, not the credential.
Historical revision through omission
You read the history and it made sense. It made sense because the part that would have stopped you from believing it was not there.
The name is designed to spread. The hook is designed to stick. If you recognized something, share the name.