Quote:
Originally Posted by mjw0586
To me, it's always good to recognize the limitations of certain stats, when those limitations exist. I find that far more interesting than a bunch of white people talking about how black people should feel about certain things.
It's certainly applicable if you're making projections that rely on exact data, rather than estimates. Those small statistical errors can become problematic if you're looking to make exact projections over large sample sizes.
Quote:
Originally Posted by Banzai-
It's just tightening up one part of the calculation by removing an approximation that is only there so that the formula is entirely calculable from the box score. The data is there elsewhere, might as well actually calculate it properly.
yeah, fine, just seems like if I went through all that trouble and I see DeAndre being the worst case guy going from super efficient to super efficient just a little less, I'd be pretty meh.
And as far as projections, if you're using pbp you shouldn't care that much about individual efficiency anyway. Everything in basketball is context dependent. Speaking theoretically since I don't have the hard data in front of me, but, hypothetically, the fact that DeAndre is really a 60.5 realTS% guy instead of 61.1%.. that 0.6% discrepancy might not matter due to the clippers constantly getting in the bonus due to the fouls he draws, and the extra OReb rates from his wild form.
On the flip side, it wouldn't shock me if guys we considered efficiency studs like Kevin Martin and Corey Maggette didn't really boost overall team ORtg that much due to other problems with their game (lack of drawn doubles, poor passing, ball-stopping, etc).