Quote:
Originally Posted by ninjacookies
Luka is amazing offensively. 42 points on a much higher efficiency night than usual for him.
But was a starter-worst -12. Warriors are going at him defensively.
|
The advanced box score shows a lot more useful info than raw plus-minus:
https://www.basketball-reference.com...205200GSW.html
You can tell by BPM more or less who was producing in the game.
I keep saying, without anyone showing a shred of evidence otherwise, that raw plus-minus is useless, I wouldn't even call it information.
If you play most minutes of a game, your raw plus-minus is likely to look a lot like the score in the game. Gee, Luka's -12 was pretty darn close to the team's -9. You'll find the pattern repeat time and time again, dare I say without exception?
So Luka being -12? That tells me his team lost by roughly that margin, but nothing else of any explanatory value. His teammates didn't show up well enough to win despite his high-BPM (~19) performance, is all.
Unless of course you can tell me the reasons you know
why you reference raw plus-minus. Imagine a Bill James of basketball looking at all kinds of numbers trying to determine what ones convey the most useful information. I can't imagine such an analyst coming up with anything about raw plus-minus saying it's useful/informative. Do you think GMs use raw plus-minus to assess a player's value? If so, they're either dumb or I'm dumb for missing what raw plus-minus tells us that other numbers don't/can't.
My hunch based on accumulated experience is that it's useless when applied to individual players, and only became cited as by folks because the concept behind it ("plus-minus") is certainly sound, but it's measures like BPM that actually give you mostly/generally useful info on roughly how much value a player brings (aside from "intangibles" that somehow don't manifest in box scores I guess?).
MJ's 63 point game against the Celtics? I would guess that his raw +/- in that game was somewhere close to -4 despite his obviously legendary value-added there.