Discernable

Every line below is statistically true. The only variable is which one you'll let me say:

1. Men are, on average, more physically capable than women.
2. Women are, on average, kinder than men.
3. Inequality in vocation selection is a feature, not a bug.
4. Paralympians are less capable than Olympians.
5. Asian kids are academically smarter than white kids. Their pen*ses are also smaller.

Stereotypes are useful because they're accurate. When you meet a stranger and have no other information, the population-level claim is the best prediction you have.

The moment you have individuating information about the person in front of you though, the prior gets overwritten by local data.

Your Asian friend may be the tallest, strongest, least academic, biggest-pen*sed man you know. But the stereotype hasn't been falsified. It's been overridden by better information for a single case.

Refusing the stereotype in the first place doesn't protect anyone. It just makes you operationally blind.

Why are we training everyone to refuse statistical thinking on the grounds that it might be bigoted? Racist?

And hey, if you're stuck in a stereotype, become an outlier. It's within your ability. You can be the data that overrides it.