PDA

View Full Version : In an all-time season, Steph Curry leads the league in Real Plus-Minus, LOWRY SECOND


AirTupac
12-10-2015, 02:45 PM
In an all-time season, Steph Curry leads the league in Real Plus-Minus at +10.90; Lowry is #2 at +9.94

Lowry top 5?
mehyaM24
12-10-2015, 02:46 PM
rpm only goes back a few years iirc.

the other stuff is xRAPM - similar but not quite the same thing.
Legends66NBA7
12-10-2015, 03:00 PM
It's only been 23 games. He was playing amazing early in the year last season too. Difference is that he's in better shape this time around and hopefully won't breakdown from all the injuries he sustained either.

I still wouldn't say he's Top 5 player. Could argue he's a fringe Top 10 player, but I think he'll cool down a bit. Can't really say with a straight face I would take him over guys like Curry, LeBron, Durant, Westbrook, and George... nor players like Griffin, Paul, Davis (yeah he's struggled, but you still really wouldn't take Kyle over him), etc...

I'll say this, he doesn't really have the most talented team around him if were comparing star players. He has DeMar and Jonas but the rest are high to below end role players. Jonas is injured, otherwise those 3 were playing really well with each other. Cory Joseph has been nice surprise too.
ArbitraryWater
12-10-2015, 03:03 PM
rpm only goes back a few years iirc.

the other stuff is xRAPM - similar but not quite the same thing.

whats the difference? one adjusts for lineups?
Pushxx
12-10-2015, 03:04 PM
I didn't think Lowry was gonna be as impactful or motivated after losing all that weight and having last season end the way it did.

I still think they'll struggle in the playoffs. Bad coach will lead to anemic offense once the regular season ends.
mehyaM24
12-10-2015, 03:06 PM
whats the difference? one adjusts for lineups?

they both adjust for lineups. xRAPM from, say, the 90s, is a bit different because i dont believe all the data we have today is included.

its basically the same stat as rpm, but with more noise.
pm fpliii and he'll explain it better than i can.
fpliii
12-10-2015, 03:27 PM
they both adjust for lineups. xRAPM from, say, the 90s, is a bit different because i dont believe all the data we have today is included.

its basically the same stat as rpm, but with more noise.
pm fpliii and he'll explain it better than i can.
RPM is by J.E.'s (J.E. being the creator who supplies it to ESPN) own admission comparable to xRAPM (he said so in this post (http://apbr.org/metrics/viewtopic.php?f=2&t=8964&start=15#p24729); I think the difference is he did some fine-tuning with the SPM, or box score prior). You are correct, the 90s xRAPM is 'fake RAPM' (link for details (http://apbr.org/metrics/viewtopic.php?t=8067)).

This link (https://web.archive.org/web/20150408042813/http://stats-for-the-nba.appspot.com) contains both fake RAPM (90-91 through 99-00) and xRAPM (00-01 and beyond).

Anyhow, you are 100% correct. All of RPM/xRAPM/RAPM contain lineup adjustments. All of the 10-man (team A vs team B) lineup combinations are used to create a data matrix, and each player is treated as a separate variable. So all players are being adjusted for quality of lineups (teammates and opponents).