Trey Lyles is a 19 yr old rookie.
Real +/- NBA ratings so far......
- Thread starter tleikheen
- Start date
Trey Lyles is a 19 yr old rookie.
GS is starting with a +13.4 points per game so even when you deduct for other great players on the floor you are going to wind up with top ratings (or there is something very wrong with the stat).
The Blue Elephant
Well-Known Member
Well RPM looks at your team with you vses without you. Kanter defense was compared against Rudy Gobert in DRPM last year.
Now it is compared against the pairing of Serge Ibaka and Steven Adams.
Not very many centers have such good defensive counterparts in the same position as Kanter does.
Kanter is bad at defense. And in the conversation as top 5 worst in the NBA for a starting caliber center. If he was better he would be an Allstar considering his good scoring and elite rebounding.
Jazz gets a bunch of young talent and loses/gets rid of them as soon as they start to get a little bit older so Jazz fans are able to recycle the "oh he is only just young" excuse when they play absolutely abysmal. Very hard to find people who played as bad as Lyles is outside the lottery. Jan Vesely and Thomas Robinson maybe?
The early returns of Lyles are horrible, regardless of age.
Sorry this is wrong. What happens in the game when the rated player is on the bench is irrelevant to their ranking. All that matters is how your team fares when you are on the floor and how good are the other 9 players who are on the floor with you.
It does adjust for all 5 opponents equally, not just the player you are defending and who is defending you.
It most definately does account for chemistry to the degree that chemistry leads to more points for your team and fewer points for your opponents. which is the bottom line.
The Blue Elephant
Well-Known Member
"RPM estimates how many points each player adds or subtracts, on average, to his team's net scoring margin for each 100 possessions played. The RPM model also yields separate ratings for the player's impact on both ends of the court: offensive RPM (ORPM) and defensive RPM (DRPM)."
https://espn.go.com/nba/story/_/id/10740818/introducing-real-plus-minus
RPM adjusts for your teamates by comparing how they do with and without you. When Kanters teamates aren't playing with Kanter they are playing with a defensive stud at center which hurts his score. One of the reasons Gobert doesn't look as elite this year compared to last year on DRPM.
vegas
Well-Known Member
I don't see any links to the exact formula. There is some vague stuff about what it claims to do. So you are taking all of the claims of the salesman at face value. I'm looking at the outcomes and saying, you know, I don't agree that's a great metric when KG is considered a far superior Defender to Rudy THIS YEAR.
It seems to do a decent job with putting Lebron, Steph and co at the top, but after that it gets off imo.
Regarding who you are playing against is the age old starters vs scrubs. Does it account for garbage time, how does it weigh against the rating of the opponent? Scoring 30 on Rudy vs dropping 30 on Pleiss in my mind are two very separate achievements. It says it does adjust for each teammate and opposing player, but exactly how. Based on results I say it is off somewhere in that.
Regarding Chemistry. What if you are on Big Als team and its iso ball all the time and you never get set up, then you get a few mins while he is out that you have better chemistry with his replacement, so all of the sudden you are much more effective. Since you spent 25 mins with Al watching him and 10 mins with somebody else, does the formula adjust for that? Sure you got good marks for the 10 mins where you were in the game log, but the other 25 when it was sparse, was that weighted to eliminate the affect of that chemistry?
IMHO it gets some right and misses a lot. Something to take with a grain of salt. I could trust more if we knew the formula, since we could better rule out situations it can't measure.
Thats the full explanation of what it does. It leaves me with many more questions than answers.
It seems to do a decent job with putting Lebron, Steph and co at the top, but after that it gets off imo.
Regarding who you are playing against is the age old starters vs scrubs. Does it account for garbage time, how does it weigh against the rating of the opponent? Scoring 30 on Rudy vs dropping 30 on Pleiss in my mind are two very separate achievements. It says it does adjust for each teammate and opposing player, but exactly how. Based on results I say it is off somewhere in that.
Regarding Chemistry. What if you are on Big Als team and its iso ball all the time and you never get set up, then you get a few mins while he is out that you have better chemistry with his replacement, so all of the sudden you are much more effective. Since you spent 25 mins with Al watching him and 10 mins with somebody else, does the formula adjust for that? Sure you got good marks for the 10 mins where you were in the game log, but the other 25 when it was sparse, was that weighted to eliminate the affect of that chemistry?
IMHO it gets some right and misses a lot. Something to take with a grain of salt. I could trust more if we knew the formula, since we could better rule out situations it can't measure.
Thats the full explanation of what it does. It leaves me with many more questions than answers.
The Blue Elephant
Well-Known Member
Rudy Gobert had a much better DRPM last year when he was opposed to Kanter.
Yes, it weights who you are on the floor with by time played. So if you are on the floor with an ineffective player (with a poor +/-) it increases your RPM.
Plus/minus ignores who you are on the floor with. So if you rack up point while on the floor with 4 starters while the opposing team plays scrubs, RPM is adjusted downward.
So I think it is an improvement over simple plus/minus, which is more flawed and misleading. But, like all statistics, they give you limited information and you can't just turn your brain off and say "because player X has a higher RPM than player Y, therefore player X is better" Instead combine various stats with other factors and make that (subjective) decision.
