All corrections
X May 19, 2026 at 08:36 PM

x.com/suchenzang/status/2056450518206361964

1 correction found

1
Claim
if you're top 10% in 3 different areas, that already makes you top 0.1%
Correction

This percentile math is not automatic. Multiplying 10% × 10% × 10% only gives 0.1% when the three traits are statistically independent, which the post does not establish.

Full reasoning

The statement treats the overlap of three "top 10%" groups as if it must equal 0.1%. That is not true in general.

Standard probability rules say the general formula for overlap is P(A and B) = P(B)P(A|B), not simply P(A)P(B). The product rule P(A and B) = P(A)P(B) applies only when the events are independent. The same logic extends to three events: multiplying 0.1 × 0.1 × 0.1 is valid only if being top-10%-level in each area is independent of being top-10%-level in the others.

But skill areas are often correlated rather than independent. So being top 10% in three areas does not automatically make someone top 0.1% overall. As a simple counterexample, if the same 10% of people are top 10% in all three areas, then the overlap is 10%, not 0.1%.

So the claim is incorrect as written because it presents a conditional probability result as an automatic one.

2 sources
Model: OPENAI_GPT_5 Prompt: v1.16.0