Does Kobe make his teammates better? Finally an Objective look.

[dirno2000]

Kevin Pelton
82games.com

Imagine a team that owns, at the same time, both the NBA's best defense and its worst offense. This scrappy bunch of overachievers, not blessed with a ton of offensive talent, nonetheless manages to stay in most games before falling just short.

This team exists not just in your imagination but in reality - sort of. This team is the 2005-06 Los Angeles Lakers, sans Kobe Bryant. Through Friday, this Web site's database indicated that, without Bryant in the game, the Lakers were scoring at a rate of just 90.2 points per 100 possessions while surrendering just 95.7 points per 100 possessions. No full NBA team is scoring less than 100 points per 100 possessions; the only team with a Defensive Rating below 100 is San Antonio (99.6).

You might think this is a common effect for a team with its superstar off the floor, but the Bryant-less Lakers take this to an extreme. Bryant's presence has been worth 22.0 points per 100 possessions to the Lakers offense this season, far and away the biggest difference made on offense by any NBA player:

Players Tm Diff
---------------------
Bryant LAL +22.0
Wade MIA +17.0
Szczerbiak MIN +15.0
Howard ORL +14.9
Nowitzki DAL +14.9

Last year, Steve Nash led the NBA with a 17.6 difference in Offensive Rating, so what Bryant is doing is rather remarkable. It's also probably not as dramatic as it seems, given that Bryant is usually out of the game for short stretches -- he's averaging 40.4 minutes per game -- and often with other starters also on the bench.

When Bryant missed two games because he was suspended for elbowing Memphis' Mike Miller in the throat, the Lakers offense struggled, but faired better than it has overall, scoring at a rate of 103.1 points per 100 possessions -- still well below their mark of 112.2 points with Bryant. (It was the defense, which surrendered 113.8 points per 100 possesssions, that really let the Lakers down as they lost both games to the Jazz.)

No matter what measure you use, Bryant's importance to the Lakers offense is evident. Yet this hasn't silenced critics that demean Bryant as a ballhog. To the extent that the term is defined as a player who shoots a lot, it is inarguable. Bryant is attempting an amazing 39.4% of the Lakers' shots while on the court this season, twice as many as an average player. Tracy McGrady (35.6%) is second in terms of percentage of his teams shots and only 15 other players are attempting even 30%.

Bryant's so-so efficiency makes it tempting to label him a ballhog. After all, his True Shooting Percentage of 54.1% is lower than fellow Lakers starters Lamar Odom (54.9%), Smush Parker (55.3%) and Chris Mihm (56.4%) as well as occasional starter Brian Cook (57.6%). It's also scarcely above the league average of 53.2%.

While I'm a big believer in the importance of True Shooting Percentage, it doesn't tell a complete story. For one, it ignores the role of turnovers. Bryant's turnover rate is in the NBA's top 15. If we look at points scored per possessions used (FGA + (.44*FTA) + TO), Bryant -- who averages 98.8 points per 100 possessions -- shoots past every Lakers regular save Cook (103.8) in terms of efficiency. Still, this alone is not enough to explain Bryant's dramatic impact on the Lakers offense.

For that, we have to look deeper into Bryant's on-court/off-court statistics available on this site to take a look at how his presence has affected the performance of his teammates. Of the 10 other Lakers who have played at least 100 minutes this season, only two -- Mihm and reserve forward Devean George -- have shot a lower field-goal percentage with Bryant on the court.

We can take this a step forward by going to the points per 100 possessions rating I referenced earlier, which takes into account the extra turnovers other players pick up with Bryant on the bench. Here is how the Lakers regulars' Offensive Ratings look with and without Bryant:

Player with w/o diff
-----------------------------
Brown 76.8 68.5 + 8.3
Bynum 66.9 49.5 +17.4
Cook 104.6 100.9 + 3.8
George 92.0 80.2 +11.8
Mihm 96.4 81.1 +15.3
Odom 92.9 91.9 + 1.0
Parker 100.5 73.3 +27.2
Profit 90.5 86.1 + 4.4
Vujacic 95.9 72.7 +23.2
Walton 81.2 64.2 +17.0

Remarkably, every single player has improved his Offensive Rating, if only by a little in a couple of cases. However, Parker and his backup, Sasha Vujacic, have seen their efficiency cut into by about a quarter when Bryant leaves the court.

George is a particularly interesting case that illustrates why field-goal percentage is not sufficient to address this issue. While George's field-goal percentage has basically been the same with and without Bryant, his Offensive Rating drops dramatically without Bryant. George hits twice as many 3s per field goal with Bryant on the court, and turns the ball over far, far less. He has just eight turnovers in 433 minutes when teamed with Bryant, 13 in 236 by himself.

Is it possible that other players make this kind of difference? Perhaps, but a glance at the statistics with and without other key offensive players, including Nash and Shaquille O'Neal from last year, reveals a less consistent pattern. Nash, for example, had a major influence of the shooting of Amaré Stoudemire -- perhaps adding context to Stoudemire's "breakout" season -- but had no effect on Shawn Marion.

Without Bryant using as many possessions as he does, the Lakers become a much more balanced team on offense. Here's how their percentage of possessions used compare with and without Bryant on the court:

Player with w/o diff
---------------------------
Brown 12.2 21.4 + 9.2
Bynum 11.2 27.6 +16.4
Cook 17.5 17.0 - 0.5
George 12.6 20.5 + 7.9
Mihm 17.6 24.6 + 7.0
Odom 17.3 24.4 + 7.0
Parker 16.0 19.3 + 3.3
Profit 16.4 23.0 + 6.7
Vujacic 10.1 14.4 + 4.2
Walton 14.2 17.9 + 3.7

It's a bit surprising that Odom hasn't stepped forward as more of a go-to player with Bryant off the court. While he's never been an aggressive player, Odom used about as many possessions in 2003-04 in Miami despite playing with Dwyane Wade. He's probably got the capability to step up his role in the offense even more.

I see the difference in his teammates' performance with and without Bryant as relating to the age-old notion that a player can "make his teammates better." To many APBRmetricians, that line of thinking is anathema, but I've always had more problem with the haphazard manner in which the claim has been used to denigrate star players blessed with untalented teammates than the concept itself. It's theoretically obvious that playing with a star player should help a player's statistics. Not only does it produce more open shots because of double-teams drawn by the star, it also forces the teammate to take fewer contested shots because of being asked to create his own shot.

Kurt from the definitive Lakers blog, Forum Blue and Gold, sees the other Lakers adjusting to and playing off of Bryant much better recently.

"Brian Cook has developed a very consistent outside shot since he came into the league, so you have started to see him run a pick-and-pop with Kobe that works well. Parker and Vujacic are figuring out where to be to get kickout passes. Mihm in particular but also the much-maligned Brown have started to find good spacing under the basket when Kobe penetrates, leading to them getting easy baskets off passes or putbacks on misses.

"Regular Laker watchers have long thought Kobe does make his teammates better - did you see the year Chucky Atkins had in 2004-05?"

Really, you can't look at statistics to determine whether a player is a ballhog. Like pornography as defined by Supreme Court Justice Potter Stewart, I know a ballhog when I see one. Given the positive effect Bryant has had on his teammates, it seems to me that the amount Bryant is shooting is not only not hurting the Lakers, it's the reason for the team's offensive success. Considering how much Bryant has helped his teammates, they certainly can't complain.

[snoop]

I think the most amazing Kobe stat has to be his 1.33 grpv. (girls raped per vacation)

[grndmstr_c]

The offensive +/- Kobe's got is fantastic. The defensive numbers are another story. He's currently a -12.3 on that end of the court.

And if I can be allowed to indulge for but a moment, that table indexing the other players' offensive efficiency with/without Bryant is eerily reminiscent (minus the TO stats) of an analysis I did on Dirk's ability to make his teammates better a few months back. Kudos to one of my favorite stat geeks on the internet (Pelton) for emulating my genius.

[dirno2000]

And if I can be allowed to indulge for but a moment, that table indexing the other players' offensive efficiency with/without Bryant is eerily reminiscent (minus the TO stats) of an analysis I did on Dirk's ability to make his teammates better a few months back. Kudos to one of my favorite stat geeks on the internet (Pelton) for emulating my genius.

I thought about that when I read the article. I said this a couple of years ago when you tracked +/- in the Sac series: you should be working for those guys.

The thing is, it's a little more intuitive with Dirk. With Kobe, there are still quite a few people on this board and in the mainstream media that actually think that he hurts his team and teammates by shooting so much.

[grndmstr_c]

fwiw, Kobe's not generally had the kind of absurd impact numbers he's had so far this season. Last year he was a +5.1, the year before +7.1, and in 02/03 he was +9.7 (per 48 minutes). The most interesting comparison is this year to last. His offensive on-court this season (112.8) is only moderately better than last year's 110.5. It's the off-court numbers that have really taken a tumble (89.9 this year vs. 105.4 last year). I think you have to at least question whether he might not be shooting so much this year that he's preventing his teammates from getting any kind of a rhythm going that they might be able to carry over when he takes a seat on the bench. After all, last year he was taking 23.6 fga's per 48 minutes and 11.9 fta's, whereas this year he's at 31.9 and 12.4, respetively. Of note, by my calcluations the Lakers averaged ~103 points per 100 possessions in the two games they played against the Jazz when Kobe was suspended.

[dirno2000]

I think you have to at least question whether he might not be shooting so much this year that he's preventing his teammates from getting any kind of a rhythm going that they might be able to carry over when he takes a seat on the bench.

I think you can argue that he has worse teammates. They essentially traded Caron Butler and Chuckie Atkins for Kwame Brown and Smush Parker. Butler was their 2nd leading scorer and the veteran Atkins is probably better equiped to play without Kobe than Paker (he was 5-19 during Kobe's suspension).

[grndmstr_c]

Here's the rub, though: Atkins, for his career, boasts a ts% of only 51.7%, and Butler's even worse at 50.3%. Do you really think that the difference betwen those two guys and Parker and Brown is worth 15.5 points per 100 possessions sans Kobe?

[kg_veteran]

I love Pelton's work, and this article contains some great analysis, but a couple of thoughts:

1. Undoubtedly, the numbers are going to show that Kobe's teammates have an easier time scoring when he's on the court than when he's off the court. I think, however, that the problem that most have with Kobe is his mistaken belief that he can be the lone superstar on a title contender. Sure, I've heard a lot of people say that Kobe shoots so much because, hey, what other choice does he have? In a sense that's true, but it's hard to feel sorry for him when he essentially made his bed by running Shaq out of town. The reason his teammates are not as good as they used to be is that he demanded that Shaq be traded and the team be built around him.

Kobe's a great individual player, but he's not winning a title as the lone superstar on that team.

2. Pelton's "points scored per possessions used (FGA + (.44*FTA) + TO)" stat is eerily like the points per possession stat I came up with and discussed in my blog some time ago -- except that his is much easier to calculate.

[dirno2000]

Here's the rub, though: Atkins, for his career, boasts a ts% of only 51.7%, and Butler's even worse at 50.3%. Do you really think that the difference betwen those two guys and Parker and Brown is worth 15.5 points per 100 possessions sans Kobe?

But when they were in LA, Atkins shot 55.8 and Butler shot 52.8. Kwame's shooting 48.5 and Smush is shooting 49.3. Butler's a good 2nd or 3rd option whereas Brown is a bust that the defense doesn't really have to worry about.

I don't know if that explains the full 15 points but the way I see it, there are only two players on that team capable of creating offense and one of them hasn't shown the willingness to consistently do it.

[dirno2000]

In a sense that's true, but it's hard to feel sorry for him when he essentially made his bed by running Shaq out of town. The reason his teammates are not as good as they used to be is that he demanded that Shaq be traded and the team be built around him.

There's some truth to that and for that reason I don't feel sorry for Kobe. I wouldn't even call myself a Kobe fan. I just don't think you can knock his game based on that.

Also Mitch has done a poor job of putting a team around him. There's a chance that we could have offered a better package for Shaq than they ultimately got but Mitch wouldn't even take our calls.

Can he win a title without another superstar? I guess that depends on what you mean by superstar. I do think that you can win a title if Kobe's you're best player. The Lakers have a ton of cap room next summer and I can't imaging that they won't be able to get a really good player to come to LA (I hope it's not Dirk). So in a couple of years a lot of the Kobe questions will be answered.

[#1MavsFan]

Sportscenter pointed out an interesting little fact last night that the Lakers are 0-7 when Odom gets more than 15 shots. That basically just proves Kobe has to shoot over and over again for that team to stand a chance.

[grndmstr_c]

Sportscenter pointed out an interesting little fact last night that the Lakers are 0-7 when Odom gets more than 15 shots. That basically just proves Kobe has to shoot over and over again for that team to stand a chance.

No it doesn't. First tell me who those games were against, second tell me whether the Lakers lost those games because of their defense or their offense, third tell me how many scoring attempts Kobe had and what his efficiency was, and then _maybe_ we can talk about what that stat does and does not prove.

[grndmstr_c]

Actually I went and checked it out myself.

The losses (I count 0-8 when Odom gets 15+ fgas) included the following:
last night to Sac
two to the Jazz when Kobe was suspended
to the Wolves on Dec. 2
to the Spurs on Nov. 29
to the Clippers on Nov. 18
to the Wolves on Nov 9
to the Suns on Nov. 3

So first off, toss out the Jazz losses, since Kobe didn't even suit up.

After that, and with the exception of Sac, the rest of those losses all came against teams that are either comparable to the Lakers (Wolves and Clippers), or clearly superior (Spurs and Suns).

In the Kings game, Kobe took 35 fgas and 13 ftas. I think it's safe to say that game doesn't really support the argument since Kobe shot plenty and the Lakers still lost. Allowing the Kings to score at a rate of ~119 pp100 probably had something to do with the loss as well.

In the Dec. 2 Wolves game, Kobe only had 16 fgas and 2 ftas, so maybe that one supports the contention that Kobe needs to shoot all the frickin' time? Nope. The Lakers scored just fine in that game, with fg% of 54.3%, including 41.2% from three. The problem, again, was the defense, which allowed the Wolves to score at a rate of better than 130 pp100.

In the Spurs game Kobe had 33 fgas and 8 fgas for 25 points.

Against the Clipps, Kobe had 35 fgas and 12 ftas for 36 points

In the first Wolves game Kobe had 26 fgas and 6 ftas for 28 points

In the Suns game Kobe had 26 fgas and 14 ftas for 39 points (making this one of only two of these 7 losses in which Kobe's inefficient offense can't in part be blamed for the loss), and the Lakers allowe the Suns to score at a rate of ~124 pp100.

So from my perspective the idea that those seven losses prove Kobe has to shoot all the time is completely bunk. If it's true that he does (I remain skeptical) there will be other ways to show it; my suggestion would be to check the relationship between Kobe's scoring attempts (or points*ts% if you want to get a sense of how important a high and efficient point total is necessary) on a game-to-game basis and whether or not the Lakers win, perhaps with defensive pp100 thrown in to control for what's going on at the other end of the court, but I'll let someone who's motivated to demonstrate that Kobe's selfishness is actually a winning trait do that. Alternatively, you could regress the Lakers' ts% as a team against Kobe's scoring attempts.