Liver_Pooty wrote:jeroka wrote:Vanderbilt_Grad wrote:Regarding RPM
1) It's not open source. So without knowing how exactly it's calculated we can't really know what it's flaws are (and it has them all metrics do).
2) RPM hasn't really been shown to be predictive yet in a scientific way. We will need more years for that ... either past years of the NBA calculated (the stuff MKA has been asking for) or to simply see how it does for the next few years.
I'm not saying that RPM is BAD, but I am saying that we just don't know if it's GOOD or not yet.
On Kemba
So this year we get to see what he can do with more talent around him. I'm honestly interested to see how that goes. One of the things that gets talked about by the analysts is how good players can develop bad habits on losing teams. With Kemba he's been asked to carry this team's scoring more than once and in games it seems like he gets over focused on that from time to time. He would stop passing as much and would start forcing shots. This year he shouldn't have to do that nearly as much and I'm interested to see if we will get less Heat Check Kemba as a result or if he will still go into Hero mode most games out of habit.
if anything, the RPM is more reliable than the traditional +/- stat.
Can you explain why? Not questioning you. Just curiois.
as the other poster said, it's not open source, so we can't really determine the "how" exactly. but for instance, it takes into account strength of the team on the floor, as well as strength of competition... while traditional +/- does not.
a wild guess to help us understand would be, if player A has a traditional +/- of +10, and the other 4 guys on the floor has a net +/- of -1, player A's RPM would be adjusted accordingly... I also read that they take into account other analytics to make further adjustments, such as hustle rates, eFG%, TS%, etc... so I put more value to RPM than the old +/-... although I still put some value on the latter since it's still widely used today.
