Item — HN Reader

Probability is hard to understand and it means different things in different contexts. (is it only about knowledge or is it out there in the world)

If it is hard for Andrew Gelman it may be too hard for most of us.

"Court of Appeal bans Bayesian probability (and Sherlock Holmes)"

https://understandinguncertainty.org/court-appeal-bans-bayes...

1 comments

The problematic thing about probability is, it is still there even if you don't understand it.

When your fingerprint gets matched with the one of a murderer, there is a certain propability that you are in fact a false-positive. When they put you into a polygraph the same is true (but much liklier).

The same is true for any evidence, also if multiple of them are involved.

Any legal system that has the interest of delivering just outcomes should be happy to be able to rely on such a methodology. The methodology "being hard" is not an argument against it. Judges already need to weigh evidence and reliability based on their experience or precedent. Having hard numbers for doing this doesn't make it harder, but easier.

I fully agree. Every judge and laywer should understand:

base rate fallacy and cold hits.

(and everyone should understand: Simpson's paradox, Regression toward the mean, NHST problems )