Something wrong with Kobe?

Wannabe MEP · Post #1 » by **Wannabe MEP** » Sun Feb 20, 2011 2:49 pm

Adjusted Plus-Minus says Kobe is absolutely killing his team this year -- basketballvalue.com has him at the third worst player in the NBA (given at least median minutes). Regardless of how you feel about APM, it's pretty hard to ignore the negative unadjusted net rating: the team is playing better when he is not on the court. That's not true of any other Laker starter, even though they play a ton of minutes together.

It looks like the difference is pretty much on the defensive end, and it doesn't look like his offensive stats are unusual for him, except...

...his usage rate is the highest it's been in years, and highest in the league. Which leads me to:

Theory 1
When Kobe ball-hogs, teammates are disengaging defensively. Perhaps they're saying, "I'm sick of this sh*t!" or perhaps it's not really conscious. But one way or another, teammates slack when Kobe is on the court.

Theory 2
Kobe is simply playing bad defense. He's lost a step, but he's still going for aggressive plays and getting burned in the process. Teams are spreading the court with good shooters more and more, and Kobe tries to do too much, abandoning the perimeter to help. I know this happened when the Suns made 22 3-pointers.

Theory 3
He's unnecessary. Shannon Brown, Matt Barnes, and to a lesser extent Ron Artest are more than adequately filling his spot. These guys are solid defensively and the Lakers as a whole are deep enough and complete enough that Kobe just doesn't really help that much. When five skilled Lakers come together to run the triangle and hustle on defense, Kobe is irrelevant, and he actually disrupts the flow at times.

Theory 4
There's a flaw in the statistics.

Theory 5
Some combination of the above.

Thoughts? Anyone who has watched the Lakers more than I have this year have any ideas?

Chicago76 · Post #2 » by **Chicago76** » Sun Feb 20, 2011 5:02 pm

It's definitely some combination of the above.

1-Bryant has definitely lost a step, and he needs to swallow a bit of his pride and adjust his defensive style accordingly.

2-Bryant's usage has increased, but it hasn't corresponded to a huge loss for most of the regulars. Fisher, Bynum, and Artest are down. The one that concerns me is Artest, for several reasons:

a-he tends to wear out his welcome everywhere he goes.
b-he finally got a title, so the fire may be gone.
c-he frequently complains about lack of touches wherever he goes, and had openly admitted he doesn't understand his role with LA.
d-His effort on the court is more directly related to his emotional state/general happiness than probably any other player I can think of.

I haven't seen enough of the Lakers this year to know if this is the difference, but it really wouldn't surprise me at all if Artest is letting his lack of touches influence his effort when Bryant is on the court with him.

3-This could also largely reverse itself with a bigger sample size, i.e., a complete season of data.

Vinsanity420 · Post #3 » by **Vinsanity420** » Mon Feb 21, 2011 12:08 am

I don't know if the increase in his usage rate is a good or bad thing... people keep ranting about this, but the Lakers are still the #2 offense in the league, and with Kobe on floor, they still produce a 115 O Rating. Payne posted something earlier in the year that their loss to Boston had little to do with Kobe shooting a lot, when Kobe was criticized for doing so.

Kobe has lost a step --- his prime ended in the december of 2010, when he and LeBron were running neck to neck for MVP, and Kobe had an injury. This is evident on his defense, but I am definitely not ready to believe that he is as bad as APM makes it out to be -- SG's simply don't impact the defense that much. His opponent PER on 82games looks about average as well. I would say it's got something to do with the sample size - Gasol's APM was a negative last year as well.

Wannabe MEP · Post #4 » by **Wannabe MEP** » Mon Feb 21, 2011 3:56 pm

My best guess is that it's a little bit of everything, but mostly theories 4 and 3.

I studied the units on this page:
http://basketballvalue.com/teamunits.php?year=2010-2011&team=LAL

Fish-Kobe-Artest is still very, very effective. The thing is, so is Blake-Brown-Barnes, so the APM response has been to give the credit to the bigs. The problem for Kobe's APM is that he hasn't mixed well with Blake-Brown-Barnes.

Blake, Steve - Brown, Shannon - Bryant, Kobe - Odom, Lamar - Gasol, Pau (unit) -6.99
Fisher, Derek - Bryant, Kobe - Barnes, Matt - Odom, Lamar - Gasol, Pau (unit) -25.60

Take a look at that first unit again, and look at this comparison:

Blake, Steve - Brown, Shannon - Bryant, Kobe - Odom, Lamar - Gasol, Pau (unit) -6.99
Blake, Steve - Brown, Shannon - Barnes, Matt - Odom, Lamar - Gasol, Pau (unit) +15.71
Blake, Steve - Brown, Shannon - Walton, Luke - Odom, Lamar - Gasol, Pau (unit) +7.04

It seems that Kobe and the Lakers will be fine as long as there is clear separation between the first and second units. It's certainly interesting to wonder whether this team would still be elite without Kobe, with perhaps this starting five:

Blake, Steve - Brown, Shannon - Barnes, Matt - Odom, Lamar - Gasol, Pau (unit) +15.71

Vinsanity420 · Post #5 » by **Vinsanity420** » Tue Mar 1, 2011 12:24 am

You have to take into account the minutes played of each lineup as well. The lineups that don't have Kobe had significantly lesser minutes... chances are, those lineups played when the game was out of reach.

It's like drawing conclusions out of the Spurs lineup this year

Hill-Neal-Manu-Bonner-Dice... posts up an APM of 33.3!

mysticbb · Post #6 » by **mysticbb** » Tue Mar 1, 2011 9:07 pm

Vinsanity420 wrote:You have to take into account the minutes played of each lineup as well. The lineups that don't have Kobe had significantly lesser minutes... chances are, those lineups played when the game was out of reach.

APM adjusts for different strength of opponents and teammates. In the case of basketballvalue.com the values are even adjusted for garbage minutes. Thus your idea doesn't apply here.

Vinsanity420 · Post #7 » by **Vinsanity420** » Tue Mar 1, 2011 9:28 pm

mysticbb wrote:
Vinsanity420 wrote:You have to take into account the minutes played of each lineup as well. The lineups that don't have Kobe had significantly lesser minutes... chances are, those lineups played when the game was out of reach.

APM adjusts for different strength of opponents and teammates. In the case of basketballvalue.com the values are even adjusted for garbage minutes. Thus your idea doesn't apply here.

You are right. I must have been drunk while writing up that post. :-?

Even though the minutes point was still valid... the small sample size shows a massive possible error.

DSMok1 · Post #8 » by **DSMok1** » Wed Mar 2, 2011 5:25 pm

Better to look at RAPM, which is ridge-regressed APM. It pulls everybody toward 0 to get rid of some of the multicollinearity problem.

In that, Kobe is +0.1 on the year, so basically average.
http://stats-for-the-nba.appspot.com/

Jimmy76 · Post #9 » by **Jimmy76** » Wed Mar 2, 2011 8:24 pm

DSMok1 wrote:Better to look at RAPM, which is ridge-regressed APM. It pulls everybody toward 0 to get rid of some of the multicollinearity problem.

In that, Kobe is +0.1 on the year, so basically average.
http://stats-for-the-nba.appspot.com/

is this yours?

CoffeeJanitor · Post #10 » by **CoffeeJanitor** » Thu Mar 3, 2011 2:24 am

Kebo is a chucker.

DSMok1 · Post #11 » by **DSMok1** » Thu Mar 3, 2011 2:22 pm

Jimmy76 wrote:
DSMok1 wrote:Better to look at RAPM, which is ridge-regressed APM. It pulls everybody toward 0 to get rid of some of the multicollinearity problem.

In that, Kobe is +0.1 on the year, so basically average.
http://stats-for-the-nba.appspot.com/

is this yours?

No, it's Back2newbelf's: see http://sonicscentral.com/apbrmetrics/vi ... php?t=2669