Kobe and D-Wade's Defense +/-

I wanted to investigate further why Kobe's On/Off defensive impact last season didn't look too good.

According to 82games, when Kobe was on the court the Lakers gave up 107.8 points per 100 possessions, or 97.9 points per 48 minutes. When Kobe was off the court, the Lakers gave up only 101.2 points per 100 possessions, or 88.6 points per 48 minutes. This gives a net difference on defense of 6.5 points per 100 possessions for Kobe. That's near the bottom of the league.

Many Laker fans have pointed out that the main reason for this is opponents faced. Kobe is usually on the floor when the best offensive lineup for the opposing team is playing, and he rests when the opposing team's best offensive units are not on the floor. Makes sense, but up to now I haven't been able to confirm this statistically.

Thanks to a new website that just turned up, basketballvalue.com, I finally got the data I needed to confirm this. It provides points scored and allowed for every 5-man unit, along with the opposing 5-man units faced.

I used this information to determine the strength of the opposition faced when Kobe was on the floor versus off the floor. I essentially took a weighted average of the offensive efficiency of the opposing 5-man units (weighted by number of minutes they played against the Lakers).

Here's a breakdown for the Lakers. Note that their possession calculation is slightly different than 82games.com, so points per 100 possessions doesn't quite match the the data there.

<font color="navy"><font size="2">
Key:
PA_48m: points allowed per 48 minutes
Opp48: opposition's points scored per 48 minutes throughout season
PA_100P: points allowed per 100 possessions
Opp100P: opposition's points scored per 100 possessions throughout season</font></font>
<div class='codetop'>CODE</div><div class='codemain'><br/>Lakers<br/>PA_48mOpp48mPA_100POpp100p<br/>on97.82 96.72 102.68 102.63<br/>off 87.54 89.24 95.7395.63<br/>Total 96.08 95.46 101.55 101.44<br/><br/>League average points per 48m:96.16<br/>League average points per 100P: 101.94<br/></div>

You can see that the opponents faced with Kobe on the court is over 7 points per 48 minutes better offensively compared to when he's off the court. That's a big part of why his On/Off defensive numbers look so poor -- he has to face much better players. From a per possession perspective, it is evident that the Lakers gave up about as many points as would be expected given the opposition with or without Kobe on the floor. Could it be that Kobe had to play with worse defensive teammates? Perhaps. But from this, it appears that his On/Off impact on defense over the whole season was essentially negligible.

I did a similar break down for Dwyane Wade on the Miami Heat:

<div class='codetop'>CODE</div><div class='codemain'><br/>Heat<br/>PA_48mOpp48mPA_100POpp100p<br/>on95.50 95.81 100.24 101.74<br/>off 96.10 93.32 101.85 98.71<br/>Total 95.66 95.14 100.66 100.92<br/><br/>League average points per 48m:96.16<br/>League average points per 100P: 101.94<br/></div>

You can see that there isn't as drastic a difference in opponent faced for the Heat; still, opposition Wade faced was 2.5 points per 48 minutes better offensively that the opposition faced when Wade was sitting. This makes Wade's On/Off defensive impact look more impressive. The Heat gave up less points per 100 possessions with Wade on the floor, and they did so facing better offensive units.

 

Very interesting Durvasa, it took me a while to understand the difference between PA48m and Opp48m, but once I understood it, it made sense.

If you had time it would be interesting to read about other players: Some I would be interested in are: Artest, Shane Battier, Bowen, AK47, Ben Wallace, Yao Ming, Dekembe, Shaq, Zo.

It would be interesting to see who had the biggest defensive impact between Artest, Bowen, AK47 and Battier.

But another thing I was thinking about, was that this tests compares how good the player is behind x-player.

Like lets say I had Nash on offense, vs. Nash on the bench. If Jason Kidd was behind Nash on the suns team, it would look like Nash wasn't good.

If Nash had Tierre Brown as his back-up his impact would look a lot better.

So its really (Kobe compared to his backup + players who play with backup) vs. (Wade vs. his backup + players around backup.)

I know you kind of addressed this. But when doing Shaq/Zo and Yao/Dekembe comparisons, if Yao had Juwan Howard as his backup center he would look a lot more impressive than having Dekembe, if Shaq had Doleac as his backup he would also appear more impressive than having Zo. So it seems to be kind of relative to player quality behind the stars.

 

<div class="quote_poster">Quoting SkiptoMyLue11:</div><div class="quote_post">Very interesting Durvasa, it took me a while to understand the difference between PA48m and Opp48m, but once I understood it, it made sense.

If you had time it would be interesting to read about other players: Some I would be interested in are: Artest, Shane Battier, Bowen, AK47, Ben Wallace, Yao Ming, Dekembe, Shaq, Zo.

It would be interesting to see who had the biggest defensive impact between Artest, Bowen, AK47 and Battier.</div>

I'll try to get to some of those at some point.

An attempt to clarify for others:

PA_48m refers to points allowed per 48 minutes by a team -- essentially their team defense.

Opp48m is like an expected points allowed per 48 minutes, given the 5-man units a team has faced and based on the season scoring rate for each of those 5-man units.

A relatively simple example might illustrate this:

Over the course of a game, a team faces 4 separate 5-man units: U1, U2, U3, and U4. Let's say that when the star player is on the court (40 out of 48 minutes), the team gives up 80 points, and when he's off the court the team gives up 10 points.

This means that PA_48m would break down as:

On: 80 *48/40 = 96 pts per 48m
Off: 10 * 48/8 = 60 pts per 48m
Total: 90 pts per 48m

Now, in the 40 minutes the star player is on the court, he faces U1 for 20 minutes, U2 or 15 minutes, and U3 for 5 minutes. And when he's off the court, the team faces U3 for 5 minutes, and U4 for 3 minutes.

Let's say that for the entire season, U1, U2, U3, and U4 have the following scoring rates:
U1: U1_pp48 = 105 pts per 48m
U2: U2_pp48 = 100 pts per 48m
U3: U3_pp48 = 90 pts per 48m
U4: U4_pp48 = 95 pts per 48m

Based on all this information, I determine the offensive strength of the opposition faced (Opp48m) as:

On: (20*U1_pp48 + 15*U2_pp48 + 5*U3_pp48 ) / 40 = 101.3 pts per 48m
Off: (5*U3_pp58 + 3*U4_pp48) / 8 = 91.9 pts per 48m
Total: (20*U1_pp48 + 15*U2_pp48 + 5*U3_pp48 + 5*U3_pp48 + 3*U4_pp48)/ 48 = 99.7 pts per 48m

So, in total, we got:

<div class='codetop'>CODE</div><div class='codemain'><br/><font color=""Navy"">PA_48mOpp48m</font><br/>On96101.3<br/>Off6091.9<br/>Total9099.7<br/></div>

For this game, we observe that they played worse "defense" without the star player on the court (albiet, small sample size), but they also faced better offensive units.

That's pretty much it, except I did it for the whole season (there are several thousand 5-man units to consider in this case). For calculating PA_100p and Opp100p, I do something similar except I consider possessions instead of minutes. That's more useful if you're comparing numbers for players who play on differently-paced teams.

<div class="quote_poster">Quoting SkiptoMyLue11:</div><div class="quote_post">
I know you kind of addressed this. But when doing Shaq/Zo and Yao/Dekembe comparisons, if Yao had Juwan Howard as his backup center he would look a lot more impressive than having Dekembe, if Shaq had Doleac as his backup he would also appear more impressive than having Zo. So it seems to be kind of relative to player quality behind the stars.</div>

Absolutely. Your essentially comparing how much more effective a team is defensively when the player is on the court versus when he's off the court, if you're doing the On/Off thing. It is very difficult to measure the defensive strength of the teammates accurately. That's why I focused more on the strength of the opponents -- it's easier to estimate how good they are offensively. I have some ideas on how to adjust for teammates a player plays with, but it will take some time for me to try it out and it may not even work right.

Let's examine the numbers I got for Kobe a bit more closely:

<div class='codetop'>CODE</div><div class='codemain'><br/>Lakers<br/><font color=""Navy"">PA_48mOpp48mPA_100POpp100p</font><br/>on97.82 96.72 102.68 102.63<br/>off 87.54 89.24 95.7395.63<br/>Total 96.08 95.46 101.55 101.44<br/><br/>League average points per 48m:96.16<br/>League average points per 100P: 101.94<br/></div>

I'll focus on the per 100 possession numbers. The way we usually would guage his defensive impact would be to take the difference in points given up when he's on the floor versus off the floor. In this case, it looks like the Lakers are 6.95 points per 100 possessions worse.

Alternately, we could focus on what Kobe and the Lakers held the opposition to versus what they could have been expected to score (Opp100p). In that case, there's almost no difference -- 0.05 points per 100 possessions. Finally, we could also bring both together to come up with a semi-adjusted On/Off defensive rating:

(102.68 - 102.63) - (95.73 - 95.63) = -0.05

We could interpret this as the Lakers are 0.05 points per 100 possessions better defensively with Kobe on the court versus without him, as opposed to being 6.95 worse.

 

<div class="quote_poster">Quoting SkiptoMyLue11:</div><div class="quote_post">
I know you kind of addressed this. But when doing Shaq/Zo and Yao/Dekembe comparisons, if Yao had Juwan Howard as his backup center he would look a lot more impressive than having Dekembe, if Shaq had Doleac as his backup he would also appear more impressive than having Zo. So it seems to be kind of relative to player quality behind the stars.</div>

Absolutely. Your essentially comparing how much more effective a team is defensively when the player is on the court versus when he's off the court, if you're doing the On/Off thing. It is very difficult to measure the defensive strength of the teammates accurately. That's why I focused more on the strength of the opponents -- it's easier to estimate how good they are offensively. I have some ideas on how to adjust for teammates a player plays with, but it will take some time for me to try it out and it may not even work right.

Let's examine the numbers I got for Kobe a bit more closely:

<div class='codetop'>CODE</div><div class='codemain'><br/>Lakers<br/><font color=""Navy"">PA_48mOpp48mPA_100POpp100p</font><br/>on97.82 96.72 102.68 102.63<br/>off 87.54 89.24 95.7395.63<br/>Total 96.08 95.46 101.55 101.44<br/><br/>League average points per 48m:96.16<br/>League average points per 100P: 101.94<br/></div>

I'll focus on the per 100 possession numbers. The way we usually would guage his defensive impact would be to take the difference in points given up when he's on the floor versus off the floor. In this case, it looks like the Lakers are 6.95 points per 100 possessions worse.

Alternately, we could focus on what Kobe and the Lakers held the opposition to versus what they could have been expected to score (Opp100p). In that case, there's almost no difference -- 0.05 points per 100 possessions. Finally, we could also bring both together to come up with a semi-adjusted On/Off defensive rating:

(102.68 - 102.63) - (95.73 - 95.63) = -0.05

We could interpret this as the Lakers are 0.05 points per 100 possessions better defensively with Kobe on the court versus without him, as opposed to being 6.95 worse.

 

<div class="quote_poster">Quoting shapecity:</div><div class="quote_post">
I think it has to do with two reasons.

1. Kobe doesn't respect their ability to shoot, and he'd rather them take a low percentage shot than dump in the post. Or he gambles too, often and gives up easy looks.

2. The game plan is for Kobe to use more energy on offense in those matchups.

Another factor to consider is the type of defenses a team uses. The Miami Heat play A LOT of zone defense and force teams to beat them from the perimeter. The Lakers play man to man defense almost exclusively and use a lot of backcourt pressure to disrupt a team's offensive rhythm.

When you pressure in the backcourt you have to gamble a lot and if you anticipate wrong it leads to easy baskets by the opponent.
</div>

These help explain why Kobe's defensive talents didn't totally translate to positive impact on team defense. I generally evaluate a player's defensive performance in a season by how much he helps his team limit the opposition from scoring, so I don't think these points would in any way excuse that. They point to Kobe putting more concentration and effort on the offensive end (I consider excessive gambling on defense as much of an offensive tactic as defensive tactic).

I'm not saying that's necessarily a bad thing ... the Lakers needed Kobe's offense much more than they needed his defense. But that doesn't mean we should give Kobe extra credit for his defensive performance.

<div class="quote_poster">Quote:</div><div class="quote_post">I'd be interested to see the 1st and 2nd half comparisons of the Lakers season. In the 1st half last year the Lakers were terrible rotating over and Kobe was usually the guy leaving his man to help out.</div>

Agreed. Just as with offense, I think extra consideration should go to performance at the end of the games when each play means more towards determining the final outcome. The site I mentioned has downloadable play by play information, so I might be able to find this out. I'll look into it.

 

First off....if this is Wade vs Kobe defensivley, Kobe wins. Kobe is a better defender, how do i know? because anybody who watches to game knows that Kobe has the ability to singlehandlely shut down his opponent on most nights. That my friend, might not show up in the stat column but if you watch a Laker game you'd see the difference. Stats aren't the answer for EVERYTHING. Kobe is the night.

 

<div class="quote_poster">Quoting Rock4life:</div><div class="quote_post"> because anybody who watches to game knows that Kobe has the ability to singlehandlely shut down his opponent on most nights. </div>

Are you talking about "ability", or what he actually did? Because if he actually did that over the course of the entire season, then it will show up somewhere in the play-by-play record. Right? If not, then what you think you see and what actually happened in reality aren't the same.

Further, when I talk about defense I'm not simply talking about shutting down the opposing player. There's much more to defense than that.

 

IMO Kobe's a better man-to-man defender than Wade no doubt

<div class="quote_poster">Quote:</div><div class="quote_post">Many Laker fans have pointed out that the main reason for this is opponents faced. Kobe is usually on the floor when the best offensive lineup for the opposing team is playing, and he rests when the opposing team's best offensive units are not on the floor. Makes sense, but up to now I haven't been able to confirm this statistically.</div>That's questionable at best, just something that people used as evi, look at all the G-Fs in the East that Wade has to defend: Vince Carter, Richard Hamilton, Michael Redd, Ben Gordon, Stephen Jackson.

The East is loaded with G-Fs. While I still think Kobe's man-to-man defense is better, Wade's man-to-man defense will be tested more than Kobe's.

I also think that Kobe could be a serious contender for DPOY if he played in the East.

 

<div class="quote_poster">Quoting shapecity:</div><div class="quote_post">
I'd be interested to see the 1st and 2nd half comparisons of the Lakers season. In the 1st half last year the Lakers were terrible rotating over and Kobe was usually the guy leaving his man to help out.</div>

I did a breakdown for each period:

<div class='codetop'>CODE</div><div class='codemain'><br/> | min PA_48m Opp48m| poss PA_100p Opp100p<br/> ---------------------------------------------------<br/>1st Half | 164399.04 95.87| 3320 102.11101.45<br/>After 1st Half | 163496.34 97.67| 3177 103.24103.12<br/>1st period | 931 99.04 95.39| 1872 102.59101.47<br/>with Kobe 2nd period | 713 99.02 96.49| 1450 101.48101.41<br/>3rd period | 899 98.57 96.65| 1756 105.21102.39<br/>4th period | 702 93.15 98.31| 1358 100.33103.71<br/>overtime | 33103.64111.94 | 64 110.94110.66<br/>entire game| 327797.69 96.76| 6497 102.66102.27<br/> ||<br/>Laker Total| 397196.08 95.46| 7828 101.55101.23<br/></div>

And what they did without Kobe on the floor:

<div class='codetop'>CODE</div><div class='codemain'><br/> | min PA_48m Opp48m| poss PA_100p Opp100p<br/> ---------------------------------------------------<br/>1st Half | 325 84.2086.95 | 62291.71 94.28<br/>After 1st Half | 369 92.2891.36 | 71099.93 97.88<br/>1st period | 5473.5786.06 | 10380.00 92.80<br/>w/o Kobe2nd period | 271 86.2987.12 | 51994.03 94.57<br/>3rd period | 8593.1291.42 | 16897.91 97.11<br/>4th period | 282 91.8791.50 | 536100.7598.46<br/>overtime | 2 113.39 69.85 | 683.33 68.41<br/>entire game| 694 88.4989.29 | 1331 96.09 96.20<br/> || <br/>Laker Total| 397196.0895.46 | 7828 101.55101.23<br/></div>



Unless I screwed up somewhere organizing the data, it appears that the Lakers with Kobe were only marginally more effective defensively after the first half, if that (key numbers would be PA_100p and Opp100p. They struggled somewhat in the 3rd quarter, but did their best defensive job in the 4th quarter. Not so much in the overtimes (small sample there, however).

 

Wow very impressive, prop to you for taking the times to evaluate their defensive stat.

On some night, Kobe played as if he was the best defensive player in the league. Then the next night, he played as if he was the worst defensive player in the league. Like shapecity pointed out, Kobe doesn't respect his opponents if they can't prove to score on a consistent basis. To be honest, I think Kobe doesn't give a crap about guys like Raja Bell, Trent Hassel, Bruce Bowen, or those less talented scorers drop 40 or 50 on him. It is hard to measure Kobe's true defensive ability when he, himself, is a very inconsistent defender. This season is definitely not the best season for Kobe defensively but he still got a defensive honor because I think those writers wanted to give Kobe's something for becoming the best player in the league.

D-Wade is a solid defender but he's still not in the same level as Kobe yet. Also the player surrounds D-Wade has something to do with his overall team defensive ability. It is better to have Shaq and Zo blocking the paint than having Kwame Brown and Chris Mihm.

 

<div class="quote_poster">Quoting shapecity:</div><div class="quote_post">I meant first half of the season. </div>



<div class='codetop'>CODE</div><div class='codemain'><br/>-------------------------------------------------------------<br/> pacemin pf/48 pa/48opp48<br/>-------------------------------------------------------------<br/>beforew/ Kobe95.19 1706102.1698.0696.79<br/>2/1 w/o Kobe 91.69 436 80.34 84.0987.48<br/> 94.48 214297.73 95.2294.90<br/>-------------------------------------------------------------<br/>after w/ Kobe95.13 1571102.6197.3096.74<br/>2/1 w/o Kobe 92.77 258 80.87 95.9392.35<br/> 94.79 182999.54 97.1096.12<br/>-----------------------------------------------------------<br/>whole w/ Kobe95.16 3277102.3897.6996.76<br/>seasonw/o Kobe 92.09 694 80.54 88.4989.29<br/> 94.62 397198.56 96.0895.46<br/>-------------------------------------------------------------<br/><br/>----------------------------------------------------------------<br/> pace posspf/100p pa/100popp100p<br/>----------------------------------------------------------------<br/>beforew/ Kobe95.193384107.33103.01 102.46 <br/>2/1 w/o Kobe 91.69832 87.62 91.7194.00<br/> 94.484216103.44100.78 100.79<br/>----------------------------------------------------------------<br/>after w/ Kobe95.133113107.87102.28 102.05<br/>2/1 w/o Kobe 92.77499 87.17 103.41 99.87<br/> 94.793612105.01102.44 101.75<br/>----------------------------------------------------------------<br/>whole w/ Kobe95.166497107.59102.66 102.27<br/>seasonw/o Kobe 92.09133187.45 96.0996.20<br/> 94.627828104.16101.55 101.23<br/>----------------------------------------------------------------<br/></div>

If the Lakers were a better defensive team in the second half of the season, it's not apparent from these numbers. Without Kobe on the floor, they were actually worse in the second half, and with Kobe the difference seems marginal.

 

<div class="quote_poster">Quoting durvasa:</div><div class="quote_post">Are you talking about "ability", or what he actually did? Because if he actually did that over the course of the entire season, then it will show up somewhere in the play-by-play record. Right? If not, then what you think you see and what actually happened in reality aren't the same.

Further, when I talk about defense I'm not simply talking about shutting down the opposing player. There's much more to defense than that.</div>
What's more to defense that making ur opponent miss the shot? What stat does that go under?

 

There's also the ability to make the offensive player pass the ball and be a liability on the offense. Or the ability to deny a player the ball, etc.

 

Nothing new to add, but that's a good find Durvasa, nice work

 

<div class="quote_poster">Quoting Rock4life:</div><div class="quote_post">What's more to defense that making ur opponent miss the shot? What stat does that go under?</div>

How are you defining "opponent"? If you juts mean the opposing positional player, there's a lot more to it. No team plays straight man-to-man defense.

How about forcing turnovers? Defensive rebounding? Help defense?

Ultimately, a player's defensive ability should be measured by how he impacts the opposing team's ability to score. That's what I'm interested in understanding. Basketball, after all, isn't about winning one-on-one matchups.

 

<div class="quote_poster">Quoting durvasa:</div><div class="quote_post">How are you defining "opponent"? If you juts mean the opposing positional player, there's a lot more to it. No team plays straight man-to-man defense.

How about forcing turnovers? Defensive rebounding? Help defense?

Ultimately, a player's defensive ability should be measured by how he impacts the opposing team's ability to score. That's what I'm interested in understanding. Basketball, after all, isn't about winning one-on-one matchups.</div>
I understand ur point. but ultimatley if ur gonna compare the better defender out of two superstars, it comes down to who's better at stopping their opponent from scoring. But all the other factors are important

 

Yeah, but say someone is defending a guard who usually scores a lot, and he is holding the guard to much less points than the guard usually scores. So, the other team decides to run the offense through their center instead, and the center, who usually doesn't score a whole lot, gets a lot of points, while the guard who usually scores all the points either passes the ball to the center or someone else, and keeps to the weakside for most of the game.

The guy who is defending the guard can't do much to stop the center, since his own man is keeping to the other side of the court, and he knows he can't go play much help D because they might pass the ball to the now-open guard. The team with the scoring center wins, even though the other team played great defense on the usual high-scorer. What about that situation?


(Sorry for the very messy explaination)