Simpson's paradox

Ah, yes.

I think a lot of people vaguely remember Simpson's paradox from school as a quirky math trick.

Like a trend appears in different groups of data but then magically disappears when you combine them.

Yeah, and it is often taught as a quirk, which honestly does a massive disservice to how dangerous it actually is in empirical research.

But the regulators would not back off on it.

They were threatening a referral to the DOJ for redlining purposes, even though we educated them about Simpson's paradox, et cetera, and just said, look it, just isolated to the only area where they can possibly be accused of discrimination, and they looked fine.

But when you add all their lending elsewhere in counties where there's no minority tracts, it diluted their penetration rate in the aggregate to the point where it looked statistically significant, but it really, the reality was it wasn't.

What are your thoughts about a situation? Have you ever run into anything like that in your experience, Josh?

We can just kind of try to rifle through them here.

One that we hear about a lot, but I often, I always need to jog my memory on is Simpsons paradox.

Oh, this is the one

Simpsons.

I think you'd probably agree for a lot of use cases, but, but thinking through and understanding.

Some of these little trickaroos is still part of what we still have to do.

Not to say that an AI couldn't think about Simpsons Paradox.

Yeah, I don't know.