ICGrad wrote: ↑Mon May 09, 2022 10:14 pm
Homer wrote: ↑Mon May 09, 2022 9:58 pm
Nobody thinks RPI is a good metric. I've never seen anyone in my lifetime leap to the defense of RPI.
Some people clearly think it's a good and valid metric, given it's significance on selection Sunday and on how frequently it's cited throughout the season. In fact, I would say it's
the most widely-cited metric on this and other Lacrosse sites.
Look, if we all just agree it's sh/t and to stop using it, then argument over. I'm in.
When I post around this time of year, I spout RPI left, right, and center like it's the Trevi Fountain. It's an immensely important metric because it's indisputably the way the Committee slots teams for purposes of determining what's a good win and what's a bad loss. How much the Committee actually looks to each team's own RPI in comparing them for selection and seeding purposes is an endlessly debated topic; probably the answer genuinely varies from year to year.
But like I just said, I don't think RPI is a particularly good measure of anything. It just happens to be the metric the Committee's chosen to treat as definitive for a particular purpose. That's the only reason why I and others talk about it at all.
When I say nobody thinks it's a good metric, I mean that nobody who's thought at all about the issue thinks RPI is an informationally robust measure of actual team quality. It's arguably "good" in other ways that are relevant to the NCAA's continued use of it:
It's simple. It's transparent. It involves virtually no judgment about inputs or weighting that could be seen as biased in favor of a particular conference or region. It's been around forever, so people aren't surprised when they tune in and hear this number being thrown around, even if they have no clue what it actually means.
And it's not like RPI's a completely atrocious formula that always gets everything wrong. It's a bit like a science fair project: sort of crude and dumb, not quite how anyone working in the field would choose to set things up, but broadly on the right track. Should the NCAA have done better in the last 30 years? Yes, they should've, but here we are.