Celtic NET PER's after 22 Games (20-2)

[Jammer]

NET PER is your PER minus the PER of the man who you are guarding.

You could title the 2009 Celtics The Great Divide,
because after the top 8 players,
the NET PER's fall off a cliff.

+11.2 Ray Allen
+11.1 Kevin Garnett
+06.9 Tony Allen (starting to come back to earth)
+06.3 Rajon Rondo (deserves Most Improved Player Award)
+03.9 Paul Pierce (I don't think anything wrong, he's letting his teammates self actualize)
+00.5 Leon Powe

-00.1 Kendrick Perkins
-01.0 Eddie Houise

-11.1 Glen Davis (doing the best that he can, can you say MUTOMBO)
-15.9 Brian Scalabrine (still, he seems to be playing better than last year)
-16.6 Gabe Pruitt (could lose his spot on the active roster to Stephon Marbury)
-18.0 Patrick O'Bryant (???)

-42.3 Bill Walker (an SG's size with an SF's quickness)

[TheSheriff]

-42.3 Bill Walker (an SG's size with an SF's quickness)

That has more to do with the fact he has played 6 minuets in his career

[Rocky5000]

^Same for Pruitt and POB. Not enough minutes to make the stat really telling.

[BillessuR6]

PER shows exactly what I see on the court. We are the worst at back up SF and center. Danny needs to add two players that can fill these positions and are an upgrade over Scal and Davis.

[GuyClinch]

^^^ Agreed. Backup center should be the main focus though.

BBD is not a rotational level NBA player. He is more a situational/depth guy. One of your 11 - 15 guys.

But Giddens/Walker might be able to help and Tony Allen is great against small SF's. Now if Doc insists on playing Scabs instead of Walker/Giddens against big SFs maybe we do have to get someone there too.

Pete

[celticfan42487]

I think BBD would look fine at the PF spot.

Defentily back-up material. But right now Centers are having trouble backing him down. At PF he could lean on those guys all night and wear them down as well as keep them out of their favorable position. Not to mention he'd probuly find it easier boxing them out on the offensive end of the floor for some rebounds in good position.

He knowes how to create contact when he attempts layups and he could really be a disruptor in that sense.

He completely disrupts Maxiell's game when we face DET and takes a big part of their bench away from them. And dispite his length, his body is large enough to mnake Varejo stuggle... I guess he's the garbage point PF killer.

[humblebum]

I agree with celticfan... Davis would be a lot more decent if he wasn't being played out of position.

The ideal scenario for the Celtics would be to bring in a good backup center (Mutombo or Joe Smith/PJ Brown who would essentially function as Centers) which would allow the Celtics to use Davis and Powe situationally as the 4th and 5th bigs, defense/offense type of pairing.

It's great to see Ray doing so well here. He just makes the Celtics a much more dangerous offensive team.

[drza]

It's interesting to follow these posts over time. At 3 games Allen's PER was at 7.2, KG's at 16.7, and Pierce's at 17.6. After 8 games the net PERs were +9 (Allen), +5.2 (KG), and +2.2 (Pierce). By 15 games we were at +10.5 Allen, +9.6 KG, +6.2 Pierce. And now, after 22 games, we see KG at +11.1 (PER of 23.4), Alen at +11.2 (PER of 21.1), and Pierce at +3.9 (PER 17.8).

KG is clearly returning to his expected high value, as he has steadily improved in chunks after a slow start and probably will be back in his customary position atop the league before all is said and done. Allen has been playing at a consistently high level since a very slow start, a level much higher than last season but almost exactly the same as his 06-07 pre-injury production. And Pierce...despite getting some early MVP buzz after the first few games and getting "what's wrong with him?" buzz these days, is actually producing almost the exact same PER now as he was to start the season. So if anything, he's been the most consistent of all, though that is consistently well below the marks that he's set in the past.

Thanks for the updates.

[UGA Hayes]

I have to say I'm not a fan of using net PER as an analytic tool. Our defense is predicated on what a great help defensive team we are. Subjectively Ray is clearly the worst defender of our starting 5 though his offensive production far exceeds my expectations for this year and makes me, as one of the bigger wet blankets of the trade for him, more optimistic that he will indeed be able to transition to a reggie miller time late career.

[threrf23]

PER vs Counterpart PER needs to be taken into account alongside other metrics, and all metrics need to be taken in context.

I also am unsure of 82games.com's data. Ray Allen, when playing alongside Tony Allen, appears to get credit as being our SF. Powe, when playing alongside Scal, gets credit as being our SF.

Now....interestingly, on/off court data would actually seem to paint Ray as the best defender on our team. He has the best defensive + / - on the team, with him on the court, our opponent scores less per possession (I'll add here, I think per possession stats are overrated, pace is its own reward when dealing with team stats) than with any of our other starters on the floor. Even in context, this would on the surface indicate that he has been our best defender, or at least one of our best defenders.

But, I agree with UGA. He definitely is not our best defender, and clearly is the worst natural defender in our starting 5. And this begs me to ask the question - when Ray is on the floor, are we doing something to compensate for a perceived defensive weakness @ SG? If we are, should we do it more often or in different contexts?

[Jammer]

PER vs Counterpart PER needs to be taken into account alongside other metrics, and all metrics need to be taken in context.

I also am unsure of 82games.com's data. Ray Allen, when playing alongside Tony Allen, appears to get credit as being our SF. Powe, when playing alongside Scal, gets credit as being our SF.

Now....interestingly, on/off court data would actually seem to paint Ray as the best defender on our team. He has the best defensive + / - on the team, with him on the court, our opponent scores less per possession (I'll add here, I think per possession stats are overrated, pace is its own reward when dealing with team stats) than with any of our other starters on the floor. Even in context, this would on the surface indicate that he has been our best defender, or at least one of our best defenders.

But, I agree with UGA. He definitely is not our best defender, and clearly is the worst natural defender in our starting 5. And this begs me to ask the question - when Ray is on the floor, are we doing something to compensate for a perceived defensive weakness @ SG? If we are, should we do it more often or in different contexts?

You are correct in the opponents get adjusted during the course of the game,
so you are not just labeled an SF or SG.

When Pierce is out, Ray Allen is the de facto SF.

When Pruitt and House are in the backcourt,
and Tony Allen is on the floor (only seen in blowouts);
Tony Allen is the SF for matching up against an opponents PER.

Still, the SG's in this league are high scoring.
Joe Johnson, Richard Hamilton, Kobe Bryant, Manu Ginoboli, Dwayne Wade,
Ben Gordon, Jamal Crawford, Kevin Martin, Tracy McGrady, OJ Mayo,
Kevin Durant (recently moved back to SF); Mike Miller, Rashad McCants, Allen Iverson.
So, although Ray may not be the greatest defender,
he's a heck of a lot better player than most posters on this board give him credit for being.

Interestly enough, I'm surprised at the naivete of posters in not recognizing KG's intelligence to refuse to negotiate an extension with the Celtics until AFTER the Ray Allen trade was complete.

[KyleCleric]

Davis just has been struggling as he's been getting use to playing regular minutes in the NBA. I expect with more minutes and practice, he'll do much better. Right now, it's not really important how guys like Pruitt, Davis, O'Bryant, and the duo in Utah play, just that they get better. How they play in the playoffs will be the key. The struggling those guys are doing now will help then. Scal of course is a different story. Doc is making a mistake by wasting minutes on him when instead Patrick O'Bryant or Gabe Pruitt could be on the floor learning how to help the team.

[Dave_From_NB]

Davis just has been struggling as he's been getting use to playing regular minutes in the NBA. I expect with more minutes and practice, he'll do much better. Right now, it's not really important how guys like Pruitt, Davis, O'Bryant, and the duo in Utah play, just that they get better. How they play in the playoffs will be the key. The struggling those guys are doing now will help then. Scal of course is a different story. Doc is making a mistake by wasting minutes on him when instead Patrick O'Bryant or Gabe Pruitt could be on the floor learning how to help the team.

Davis is getting 3 more minutes a game than last year, I don't think the minutes are an issue. Unfortunately he's playing out of position, and there is a scouting report against him now so teams are increasingly knowing what to do about him.

Davis and Scal are in the bottom 5% of players for PER, it's not a surprise that their net-PER's are so bad.

[drza]

PER vs Counterpart PER needs to be taken into account alongside other metrics, and all metrics need to be taken in context.

Absolutely, 100% agree. No stat stands alone as a "gold standard", there should instead be several independent stats used as well as corroboration with other means when making a real "decision".

FWIW, I do appreciate 82games.com's Roland rating, which combines Net PER with +/-, as it attempts to correct for some of this. Net PER says, in general terms, whether one player is outproducing their opposing player individually. +/- says, in general terms, whether the team as a whole is better or worse when that player is on the court. Used together, these two stats act as sanity checks for each other. It's fairly uncommon for a fluke player to be extremely strong in both areas, so if a player's Roland Rating is pretty high there's a better chance that they are playing well than just looking at Net PER or +/- alone.