Sturgeon's law
Sturgeon revelation · 90% of everything is crud
Ninety percent of everything is crud. Theodore Sturgeon said it in the 1950s defending science fiction, and the defense is the point: the ninety percent is not special to any field, so judging a field by its average is judging nothing.
In practice
Whatever you are comparing yourself against is the visible ten percent of its field against the whole of yours, including the parts nobody was ever going to see.
The common mistake
Using it as a put-down. Sturgeon coined it as a defense, and the operative clause is the second one — that every field is like this — which makes it an argument against dismissing a field rather than for it.
Theodore Sturgeon was a science fiction writer, and in the 1950s he was tired of critics who reviewed the best of every other genre and the worst of his. His response, which he later wrote up properly: yes, ninety percent of science fiction is crud — and ninety percent of everything is crud.
The first clause is the one people quote. The second is the one that does the work.
Why the second clause matters
It converts an insult into a base rate.
If ninety percent of everything is poor, then finding poor examples in a field tells you nothing about the field. You have measured the baseline, which is the same everywhere. What distinguishes fields is their top decile, and that is the only comparison worth making.
Which is why the argument cuts against dismissal. A critic who reads three bad novels of a kind and concludes the kind is bad has performed no measurement at all. The same procedure disposes of literary fiction, of classical music, and of philosophy.
The two things it is actually good for
Calibrating your own output. If nine in ten attempts at anything are poor, then producing poor work is the ordinary condition of producing work, and it is not evidence of anything about you. The implication is about volume: a person who writes ten things and a person who writes a hundred have, at the same hit rate, one good piece and ten. This is the same reason deliberate practice is measured in attempts rather than in hours of intention.
Correcting the comparison. What you see of any field is its surviving tenth, because the other ninety percent was filtered out before it reached you — by editors, by time, by whoever decided what to put on the shelf. Comparing your unfiltered whole against someone else's filtered remainder is survivorship bias with a personal cost attached, and it is the most common form the error takes.
Where the number comes from
Nowhere. Sturgeon picked ninety percent because it is a plausible large fraction, and no measurement supports it as a constant. Nothing in the argument depends on the number being right: the claim only needs the poor fraction to be large and roughly similar across fields, which is hard to dispute and impossible to quantify.
Treating it as a finding rather than a rhetorical figure is the way to get it wrong. It is an argument about where to look, not a statistic.
The limit
It says nothing about whether the top decile of two fields is comparable. That a field has good work in it does not establish that the good work is as good as the good work elsewhere, and Sturgeon's law is sometimes used as though it did — as a general amnesty for any field that can produce a few strong examples.
The honest use is narrower and still worth having: do not judge by the average, and do not judge yourself by someone else's edit.
Concept web
Open the full webQuestions
What is Sturgeon's law?
Ninety percent of everything is crud. Theodore Sturgeon coined it in the 1950s defending science fiction, and the second clause — that this is true of every field — is what makes it an argument rather than an insult.
Why is Sturgeon's law a defense rather than an attack?
Because if the poor fraction is the same everywhere, finding poor examples in a field measures nothing about that field. What distinguishes fields is their best work, which is the only comparison worth making.
Is the ninety percent figure accurate?
No measurement supports it, and nothing in the argument needs it. The claim only requires the poor fraction to be large and roughly similar across fields. Treating it as a statistic rather than a rhetorical figure is how it gets misused.