What's new
By:

plus/minus stat per minute after 7 games

MT Steve said:
It doesn't even tell you who played well that particular game. There are so many variables that go into that stat that have nothing to do with the individual player's performance that it makes it essentially worthless when measuring an individual player.
Click to expand...

I disagree. A large value still means that your team did well when the player was on the floor, and a small (i.e. negative) value still means that your team did poorly when you were on the floor. And if you look at the league leaders they tend to be players who everyone would recognize as the best in the NBA. Obviously there are exceptions, particularly at the start of the year, but I still think the stat has a lot of value.
 
colton said:
I disagree. A large value still means that your team did well when the player was on the floor, and a small (i.e. negative) value still means that your team did poorly when you were on the floor. And if you look at the league leaders they tend to be players who everyone would recognize as the best in the NBA. Obviously there are exceptions, particularly at the start of the year, but I still think the stat has a lot of value.
Click to expand...

Great players being at the top is because superstars have a massive impact on their teams relative to everybody else. Superstars drive the +/- stat exponentially more than the other players. That doesn't make it a good stat - and to prove my point, let's look at 2016-17:

Five of the top six in the stat last year were Stephen Curry, Draymond Green, Klay Thompson, Kevin Durant, and Andre Iguodala. Zaza Pachulia ranked 14th overall in plus/minus. This is a function of the Golden State Warriors being by far and away the best team, and does not mean that Thompson and Iguodala are top 6 players. It also sure as hell doesn't mean Zaza is the 14th best player in the NBA. The year before, Zaza Pachulia wasn't even a top 200 player per plus/minus. Playing with much better players vaulted him from 201st to 14th, without his individual skill-set really changing at all.

On the other end of the spectrum, you'll have superstars in the NBA that are surrounded by a garbage roster (Anthony Davis). Plus/Minus doesn't even regard him as a top 100 player.

It's a bad stat for measuring the value of individual players, plain and simple.
 
+/- is so dumb for evaluating individual players. It's good for evaluating how 5-man lineups do against other 5-man lineups, and that's about it.
 
MT Steve said:
Great players being at the top is because superstars have a massive impact on their teams relative to everybody else. Superstars drive the +/- stat exponentially more than the other players. That doesn't make it a good stat - and to prove my point, let's look at 2016-17:

Five of the top six in the stat last year were Stephen Curry, Draymond Green, Klay Thompson, Kevin Durant, and Andre Iguodala. Zaza Pachulia ranked 14th overall in plus/minus. This is a function of the Golden State Warriors being by far and away the best team, and does not mean that Thompson and Iguodala are top 6 players. It also sure as hell doesn't mean Zaza is the 14th best player in the NBA. The year before, Zaza Pachulia wasn't even a top 200 player per plus/minus. Playing with much better players vaulted him from 201st to 14th, without his individual skill-set really changing at all.

On the other end of the spectrum, you'll have superstars in the NBA that are surrounded by a garbage roster (Anthony Davis). Plus/Minus doesn't even regard him as a top 100 player.

It's a bad stat for measuring the value of individual players, plain and simple.
Click to expand...

In this thread I'm looking at individual players on the same team though, not individual players compared from team to team. So that complaint doesn't really hold.
 
colton said:
In this thread I'm looking at individual players on the same team though, not individual players compared from team to team. So that complaint doesn't really hold.
Click to expand...

Even when you limit it to looking at players on the same team it doesn't hold up. Unless you want to make the argument that last year, Klay Thompson was better than Kevin Durant, JJ Redick was better than both Blake Griffin and DeAndre Jordan, Jae Crowder was Boston's best player, and hundreds of other examples.

Getting back to the Utah Jazz, Rudy Gobert is 7th in plus minus per the first post. This is a reflection of the starting lineup as a whole being ineffective - it certainly doesn't mean that Rudy Gobert has been only the 7th best player on the Jazz. It really reveals nothing about Gobert's individual play.
 
Plus/minus can certainly be something to look at to determine the value and production of a player, team and certain lineups. It shouldn't be used by itself to make a conclusive evaluation, but certainly should be used. I guarantee you front office NBA dudes look into it.
 
Archie Moses said:
Plus/minus can certainly be something to look at to determine the value and production of a player, team and certain lineups. It shouldn't be used by itself to make a conclusive evaluation, but certainly should be used. I guarantee you front office NBA dudes look into it.
Click to expand...

It is good to use to determine the relative effectiveness of different lineups.
 
MT Steve said:
It is pretty worthless in evaluating the performance of an individual player, for all the reasons I've already stated.
Click to expand...

It can be worthless, but like I said before, it can be a useful tool to see if some players are really the common denominator in line up success and to see some, under valued players.
 
Archie Moses said:
It can be worthless, but like I said before, it can be a useful tool to see if some players are really the common denominator in line up success and to see some, under valued players.
Click to expand...

Yeah, if you're comparing the +/- of line-ups involving Players A, B, C, D, E to Players A, B, C, D, F, you might find that the latter lineup is more effective. Could mean Player F is a better fit for that specific lineup.

That's a bit different though than simply listing the +/- for every player on the team and attempting to use that to judge the performance of those players individually - it's essentially useless in that context.
 
Lol ESPN dude just called out Al Horford's individual plus/minus on national TV. Has nothing to do with individual play though. Ground breaking news. MT Steve should be a GM or journalists or not skewed minded. :)



Hehe








Peepee






****
 
MT Steve said:
Yeah, if you're comparing the +/- of line-ups involving Players A, B, C, D, E to Players A, B, C, D, F, you might find that the latter lineup is more effective. Could mean Player F is a better fit for that specific lineup.

That's a bit different though than simply listing the +/- for every player on the team and attempting to use that to judge the performance of those players individually - it's essentially useless in that context.
Click to expand...

Plus or minus does not equal 100% factual lineup and individual play. It gives a good measure of both. Not hard to understand.
 
MT Steve said:
Even when you limit it to looking at players on the same team it doesn't hold up. Unless you want to make the argument that last year, Klay Thompson was better than Kevin Durant, JJ Redick was better than both Blake Griffin and DeAndre Jordan, Jae Crowder was Boston's best player, and hundreds of other examples.

Getting back to the Utah Jazz, Rudy Gobert is 7th in plus minus per the first post. This is a reflection of the starting lineup as a whole being ineffective - it certainly doesn't mean that Rudy Gobert has been only the 7th best player on the Jazz. It really reveals nothing about Gobert's individual play.
Click to expand...

We're going to have to continue to disagree here. The starting lineup being ineffective is absolutely due in part to Gobert's individual play. Is he the 7th best player on the Jazz? Of course not. But the stat does still absolutely mean something valuable.
 
MT Steve said:
That's a bit different though than simply listing the +/- for every player on the team and attempting to use that to judge the performance of those players individually - it's essentially useless in that context.
Click to expand...

No one has claimed that +/- is a perfect measure of the value of individual players. But it does mean something because an individual player's ability is definitely correlated with the team +/- when the player's on the floor.
 
There is a loose correlation, but given how many other variables go into the stat, it's nowhere near enough to draw the conclusions that you attempted to do in the opening post.

Rather than going in circles over this, I'll just point out that there are much better statistics (ie more accurately predictive) such as RPM, RAPM, and BPM, that attempt to do the same thing as raw +/-. except they use a logical regression to adjust for the strength of teammates/opponents. Even if you believe that raw +/- can be useful when measuring individual performances, why not use a better stat that adjusts for a lot of the noise?
 
MT Steve said:
There is a loose correlation, but given how many other variables go into the stat, it's nowhere near enough to draw the conclusions that you attempted to do in the opening post.
Click to expand...

Um, I intentionally drew essentially no conclusions in the OP. I just posted the numbers with a couple of minor observations of things that surprised me.

Rather than going in circles over this, I'll just point out that there are much better statistics (ie more accurately predictive) such as RPM, RAPM, and BPM, that attempt to do the same thing as raw +/-. except they use a logical regression to adjust for the strength of teammates/opponents. Even if you believe that raw +/- can be useful when measuring individual performances, why not use a better stat that adjusts for a lot of the noise?
Click to expand...

Because this is a thread about +/-, namely how the team has done when particular players are on the floor. When you say "use a better stat", I think you are reading way too much into my posts. I'm only using this stat for what it is.
 
You must log in or register to reply here.

Share this page
Back
Top