Is Jeff Green finally realizing his potential

Lately I have been thinking/hoping that Uncle Jeff is finally showing some consistency, and developing into the type of player many hoped he could be - maybe not a star, but at least a player who regularly realizes his considerable athletic potential. Forsberg recently wrote a "good-bad-ugly" analysis of Green's game:

http://espn.go.com/boston/nba/story/_/id/12006386/jeff-green-work-translating-success-boston-celtics

AForsberg basically says that although JG has been more consistent this year, "advanced metrics" suggest the Celtics production is better when he is off the floor. I feel Forsberg relies too much on advanced metrics with little understanding of how a metric is determined and whether or not the assumptions underlying the metric are correct. For example, Jeff's "advanced metrics" for "team" production will look worse than the other starters simply because Jeff plays many minutes with the bench. Although the bench was spectacular the other night, this is not always the case. Anyhow, you all can read the article yourself. I am interested to other people's take on JG this year...

Advanced metrics have the same characteristics as the +/- stat in one important respect:  The metrics of players who play a lot of minutes tend to reflect the metrics of the club as a whole.

Russ and Wilt used to play almost every minute.  Therefore, their +/- was basically the same as the entire club's +/- for a given game.  The better a player is, and the more minutes he plays, the more heterogeneous will be the mix of teammates he plays with, and the less advanced metrics will reflect his own performance as opposed to the collective performance of the teammates with whom he plays.

Jeff is particularly susceptible to the biases inherent in metrics because he tends to play small (with one group of teammates) when they go big, and he tends to play big (with another group of teammates) when they go small.  That's just one of the myriad variations that come to bear on the team's performance when he's on the floor vs. being off the floor.  And metrics really reflect the team's performance (not his) when he's on or off the floor because the metrics cannot isolate his own performance (and all its nuances) within what is inherently a team sport.

I'm sure metrics have merit—especially to the extent that they measure combinations of players rather than individual players.  But, as measures of individual players, they're really limited at best.

Sam

Sam

Sam, yes, that is absolutely correct. And something like PER, if you actually look at the equation (at least 34 terms) - it is clear that it favors scorers. I know Hollinger tried to make it more general, but it is a metric designed to reinforce that someone like Jordan is a great player. The problem is that most sports writers (and fans) don't have the training to truly understand the numbers. Forsberg for instance, throws out numbers based on ridiculously small sample sizes. To his credit, he usually states that the sample size is small, but the logical equivalent of his qualifying statement is "this stat is not valid, but here is what it is and i am going to draw conclusions from it anyways". Don't mean to single out Forsberg, because he seems to try and do a good job, just want to point out that advanced metrics need to be taken with a grain of salt...

This is why I like stats somewhat, but hate too much overanalytical neurosis of stats and I don't care for stat geeks in general....for instance stats would tell you Kendrick Perkins is a marginal player, but when you see people slow up and move very carefully around him when he sets a bruising pick or see him relentlessly shoulder, hip, forearm constantly knock players off balance in the post.....anyway you get my point.

These aren't the depth of stats you get from Synergy Sports, but they are something.


http://www.hoopdata.com/player.aspx?name=Jeff%20Green




bob



.

The article itself makes a case for throwing the stats aside when it points out he's purely a scorer.  That's his strength, that's what we need from him.  

Of course the stats reflect a more productive, balanced attack with him off the floor; what would one expect when you're running so much of the motion offense for him, or any one player for that matter? The numbers are simply pointing out what they should, given the number of times we're getting the ball to him in this offense, by design. As simplistic as this sounds, I don't think the author gets how the offense we're trying to use works.

I sincerely hope Uncle Jeff is finally hitting his stride.  I temper my optimism with the fact he's had these streaks before, however.  If he indeed is getting there at long last, that's one more piece of the puzzle we already own, and that would be big.