Analysis of the Spurs' Perimeter Defense Against OKC (2016)

[Chinook]

Hey all,

I guess I'm planning on doing this again this season, even though I don't know where I'm going to find the time. For those of you who aren't familiar, threads like this are where I rewatch the games to score every possession of the opposing star to see which Spur is most responsible for the result. I've been considering ended possessions to be those that result in fouls, shots or turnovers. The end goal is to get PPP scores for players and the defenders.

This introduction post (since I'm not going to break down Game One until early Monday evening probably) is just for me to ask if there is anything in particular folks want me to evaluate or note other than the stated PPP discussion. We've done opponent role-player spotlights, three-point closeouts and shot charts in the past. I'm always open to suggestions for new angles. I am actually going to go to work in a couple of hours, so I'll be making final decisions on the Game One post after I get off (let's say noon CST). So if you have an idea, suggest it before then.

Let's hope for a good series, everyone, with lots of great defensive effort. I want to only have to do four breakdowns.

[DPG21920]

I would like to see the shot charts and breakdowns of shot % area like the ones I posted in my thread from NBA.com (how many shots in the paint from corner 3, straight away 3, mid-range).

Also, your own version of contested vs uncontested.

Looking forward to it :tu

[BillMc]

I'm sure it will be a great read. Thanks for doing this OP.

[Chinook]

Also, we can use this thread for general data pertaining to the perimeter D. So if someone has numbers or charts of their own or that they find, they are welcome to add them to the discussion.

[cutewizard]

thanks in advance!!!!

[YGWHI]

I've been considering ended possessions to be those that result in fouls, shots or turnovers. The end goal is to get PPP scores for players and the defenders.

This introduction post (since I'm not going to break down Game One until early Monday evening probably) is just for me to ask if there is anything in particular folks want me to evaluate or note other than the stated PPP discussion. We've done opponent role-player spotlights, three-point closeouts and shot charts in the past. I'm always open to suggestions for new angles.


I would like to see the shot charts and breakdowns of shot % area like the ones I posted in my thread from NBA.com (how many shots in the paint from corner 3, straight away 3, mid-range).

Don't see very often in this type of analysis, the % ISOs, % denied the ball, % shoot less, % drives lees, of opponents players against one defender.

Just read some of those before the DPOY voting but most of them were about ISOs and shoot less %. Would be great to see all them here.

[YGWHI]

Also, we can use this thread for general data pertaining to the perimeter D. So if someone has numbers or charts of their own or that they find, they are welcome to add them to the discussion.

Is there a particular reason why to do this only about the perimeter defense? I mean, Tim is still elite and he did a great job tonight against OKC bigs but also, protecting the rim. And LMA looks great, too.

I'd love to see in one place all the numbers.

[SAGirl]

Thanks for doing this Chinook. Looking forward to it. I haven't seen your prior threads so I have no idea how these look like but I guess the league keeping track of hustle stats is helpful to add a layer to it.

Since rebounding is huge this game, very underrated are rotations where perimeter players help to box out bigs, collect contested defensive rebounds or tip them to a teammate or recover loose balls. These are game winning efforts that don't necessarily reflect on individual defense, but good team defenders help the helper etc. It could be an added layer to evaluating our perimeter players. Just a suggestion, not sure how you can incorporate these but since you are watching the game again anyways.

[TampaDude]

Scoreboard.

[Chinook]

Is there a particular reason why to do this only about the perimeter defense? I mean, Tim is still elite and he did a great job tonight against OKC bigs but also, protecting the rim. And LMA looks great, too.

I'd love to see in one place all the numbers.

Rim protection is often part of what I consider perimeter defense. Help defense is going to be a major factor in my numbers. But really, I just said perimeter defense because that's what these threads have always been called. We'll probably be focusing on Durant, Westbrook, Ibaka and Kanter unless someone like Waiters or Morrow catches fire.

[UZER]

How much did the defense suffer when the Diaw / West combo we on the floor in the first half.

[GSH]

First quarters are going to be important in this series, but more so against the Warriors. (I'm giving the Spurs credit for getting to the WCF.) In the first 7 minutes or so, Westbrook missed quite a few shots from close. I'd like your take on how much of that was the Spurs defense, and how much was Westbrick just missing shots. Maybe whether there were other Thunder players wide open, if WB had decided to kick instead of doing it himself.

BTW - what you're doing is a lotta damn work. Thanks.

[Chinook]

Don't see very often in this type of analysis, the % ISOs, % denied the ball, % shoot less, % drives lees, of opponents players against one defender.

Just read some of those before the DPOY voting but most of them were about ISOs and shoot less %. Would be great to see all them here.

So these are actually extremely fascinating suggestions, and I do think having those numbers would be really helpful. I will say that I don't know that I can provide those numbers. In addition to not knowing the standards they use to come up with their findings, getting those numbers for each player seems extremely time-consuming, which makes it more impractical for me to do this time around. However, I will see how close I can get with the raw data I collect. If you see those anywhere, feel free to post them here. That is the exact type of thing I feel adds to the discussion as least as much as whatever I do.


I would like to see the shot charts and breakdowns of shot % area like the ones I posted in my thread from NBA.com (how many shots in the paint from corner 3, straight away 3, mid-range).

Also, your own version of contested vs uncontested.

Looking forward to it :tu

Well, remember that I don't do anything with graphics, so my version of shot charts are all text-based. If you find interesting images, post them. I do like the idea of pointing out whether the shots were contested or not.


Thanks for doing this Chinook. Looking forward to it. I haven't seen your prior threads so I have no idea how these look like but I guess the league keeping track of hustle stats is helpful to add a layer to it.

Since rebounding is huge this game, very underrated are rotations where perimeter players help to box out bigs, collect contested defensive rebounds or tip them to a teammate or recover loose balls. These are game winning efforts that don't necessarily reflect on individual defense, but good team defenders help the helper etc. It could be an added layer to evaluating our perimeter players. Just a suggestion, not sure how you can incorporate these but since you are watching the game again anyways.

Absolutely love this idea, especially given the storyline going into this series. This is really the type of suggestion I'm looking for, as it's a stat that's both not hard to track but also not easy to find.

Keep it up people.

[YGWHI]

Rim protection is often part of what I consider perimeter defense. Help defense is going to be a major factor in my numbers.
I get it :tu

[apalisoc_9]

Probably just a subtle way for the op to satisfy his Danny Green Fanboysim. I can through his tactics.

[GSH]

Probably just a subtle way for the op to satisfy his Danny Green Fanboysim. I can through his tactics.


There are a lot of people stepping up with some damn fine basketball takes and analysis recently. Then there are a few who just go from thread to thread pissing on everything, like dogs.

[Aztecfan03]

There are a lot of people stepping up with some damn fine basketball takes and analysis recently. Then there are a few who just go from thread to thread pissing on everything, like dogs.

He's needs a Snickers. Or maybe he's just a miserable person because he is vegan.

[GSH]

This introduction post (since I'm not going to break down Game One until early Monday evening probably) is just for me to ask if there is anything in particular folks want me to evaluate or note other than the stated PPP discussion.


One more thing. How about a breakdown of Duncan's defensive play, and whether he's still really anchoring the defense or just holding his own.

[will_spurs]

I remember these threads, they were great, the input will be welcome if you can find the time.

The main thing I'd like to understand is whether Ibaka was just hot, or the Spurs just dared him to beat them, or the Spurs had some defensive lapses on these plays.

Re: numbers, here are a few from http://espn.go.com/blog/statsinfo/post/_/id/117656/spurs-crush-thunder-behind-kawhi-leonardlamarcus-aldridge


Russell Westbrook (http://espn.go.com/nba/player/_/id/3468/russell-westbrook) was initially guarded by Kawhi Leonard (http://espn.go.com/nba/player/_/id/6450/kawhi-leonard) on 84 percent of the Thunder’s half-court offense while both were on the court Saturday. Westbrook was 3-of-16 in half-court and was held to two points on 1-of-7 shooting and committed a turnover when Leonard was the primary defender.
Danny Green (http://espn.go.com/nba/player/_/id/3988/danny-green) was Durant’s initial defender on 62 percent of the Thunder’s half-court possessions while Durant was on the court. When Green was the defender throughout the possession, the Thunder shot 11-of-30 (36.7 percent). Durant had nine points on 3-of-7 shooting.
Leonard started and finished on Durant on only three possessions. The Thunder were held to 0-of-3 shooting on those possessions, and Durant did not take a shot on them.

[Darius Bieber]

even though they score close to 100, Durant and Westbrook got frustrated..

[YGWHI]

Re: numbers, here are a few from http://espn.go.com/blog/statsinfo/post/_/id/117656/spurs-crush-thunder-behind-kawhi-leonardlamarcus-aldridge

Russell Westbrook (http://espn.go.com/nba/player/_/id/3468/russell-westbrook) was initially guarded by Kawhi Leonard (http://espn.go.com/nba/player/_/id/6450/kawhi-leonard) on 84 percent of the Thunder’s half-court offense while both were on the court Saturday. Westbrook was 3-of-16 in half-court and was held to two points on 1-of-7 shooting and committed a turnover when Leonard was the primary defender.
Danny Green (http://espn.go.com/nba/player/_/id/3988/danny-green) was Durant’s initial defender on 62 percent of the Thunder’s half-court possessions while Durant was on the court. When Green was the defender throughout the possession, the Thunder shot 11-of-30 (36.7 percent). Durant had nine points on 3-of-7 shooting.
Leonard started and finished on Durant on only three possessions.The Thunder were held to 0-of-3 shooting on those possessions, and Durant did not take a shot on them.
Thanks! Shooting less is one of the stats I want to see in %. Sometimes, Tom Haberstroh or Kevin Pelton tweet that type of stats but not tonight.

[Russo21]

Ibaka averaged 13.7ppg, 64%FG, 61%3PTFG against Dallas and just had 19 points, 53%FG, 50%3PTFG in game 1. I know when you play the Thunder it's a pick your poison kinda deal with Westbrook and Durant and they did a great job defending them. When Ibaka started hitting 3s I started thinking about the Thunders playoff series win against us a few years ago. Do not want Ibaka to go off on us from outside and I hope someone can cool him down, especially when they head to Oklahoma for game 3. Great game, just something to be careful of.

[Slomo]

Probably just a subtle way for the op to satisfy his Danny Green Fanboysim. I can through his tactics.

Posters have stepped up and are sacrificing time to bring good takes and threads to the forum improving it by leaps and bounds, instead of just whining about how great ST used to be. Unfortunately you are not among them - even worse you decide to make fun of people for doing it.

Go eat a steak, drink a beer and calm the fuck down.

[MI21]

So these are actually extremely fascinating suggestions, and I do think having those numbers would be really helpful. I will say that I don't know that I can provide those numbers. In addition to not knowing the standards they use to come up with their findings, getting those numbers for each player seems extremely time-consuming, which makes it more impractical for me to do this time around. However, I will see how close I can get with the raw data I collect. If you see those anywhere, feel free to post them here. That is the exact type of thing I feel adds to the discussion as least as much as whatever I do.



Well, remember that I don't do anything with graphics, so my version of shot charts are all text-based. If you find interesting images, post them. I do like the idea of pointing out whether the shots were contested or not.



Absolutely love this idea, especially given the storyline going into this series. This is really the type of suggestion I'm looking for, as it's a stat that's both not hard to track but also not easy to find.

Keep it up people.

I know why SAGirl suggested this, but I won't hate :)

I feel like Pop made sure Kyle was on court when Kanter was as his length helps clear the boards. Could have been coincidence, but Kyle was very helpful on the defensive boards.

[EVAY]

I'll be thrilled with whatever you come up with, based on your description. It sounds like a huge amount of work, and I thank you for it in advance.

[playbonner15]

Lay em out son

[Agloco]

Probably just a subtle way for the op to satisfy his Danny Green Fanboysim. I can through his tactics.


There are a lot of people stepping up with some damn fine basketball takes and analysis recently. Then there are a few who just go from thread to thread pissing on everything, like dogs.


Posters have stepped up and are sacrificing time to bring good takes and threads to the forum improving it by leaps and bounds, instead of just whining about how great ST used to be. Unfortunately you are not among them - even worse you decide to make fun of people for doing it.

Go eat a steak, drink a beer and calm the fuck down.

Just ban him already and be done with it. You know it's going to happen eventually tbh.

Props to Chinook for this thread :tu

[wildbill2u]

I'd like to see a stat on succcessful Pick n Rolls vs. well-defensed PNRs. We did a good job last night, often leaving the OKC Big 2 frustrated without a shot on this basic play.

[DPG21920]

Can you do this for OKC's defense of SA and set it to circus music?

[GSH]

Posters have stepped up and are sacrificing time to bring good takes and threads to the forum improving it by leaps and bounds


Speaking of that, I'm waiting for CryHavoc's post-game grades. I'm not starting a new thread to ask, but this is a good place for it.

So step it up, CryHavoc. What could possibly be more important that taking time out of your life to give us something to read?

[Chinook]

Here are the raw numbers for all players:



Player
Poss.
Makes
Misses
Fouls
TOs
Points
PPP


Westbrook
23
5
14
4
2
14
0.608696


Durant
17
6
9
2
1
16
0.941176


Ibaka
15
8
7
0
0
19
1.266667


Waiters
11
3
3
4
2
11
1


Adams
9
4
3
2
0
9
1


Kanter
9
3
4
1
1
6
0.666667


Payne
8
1
5
0
2
3
0.375


Foye
4
2
2
0
0
5
1.25


Morrow
2
1
0
1
0
5
2.5


Roberson
2
2
0
1
0
0
0


Collison
1
1
0
0
0
2
2


Mohammed
1
1
0
0
0
2
2


Singler
1
0
1
0
0
0
0



And here is how the Spurs players fared individually:



Player
Poss.
Makes
Misses
Fouls
TOs
Points
PPP


Aldridge
23
9
13
1
0
22
0.956522


Green
17
5
8
2
2
12
0.705882


Anderson
11
5
4
2
2
12
1.090909


Duncan
8
2
5
1
0
4
0.5


West
8
3
4
0
0
8
1


Diaw
6
3
2
1
0
7
1.166667


Leonard
6
3
2
3
1
10
1.666667


Marjanovic
6
2
3
1
0
4
0.666667


Martin
6
1
3
2
1
5
0.833333


Parker
5
4
1
1
0
4
0.8


Ginobili
3
0
1
1
1
2
0.666667


Miller
3
0
2
0
0
2
0.666667


Mills
1
0
0
0
1
0
0



What sticks out to me is how often Green was attacked and how often LMA found himself in help position. You can pretty much disregard Waiters, as he was garbage time. Adams did well on the boards and in garbage plays. Ibaka was the most efficient Thunder player, and hopefully he doesn't get too many games like that. Him making those threes completely drives down LMA's rating. Aldridge was by far the Spurs' best defender. He cleaned up so many Thunder drives. When Westbrook said the help men were his main focus, I can see why. Other thing is that Kawhi was only involved in six possessions, though to be fare, he also had three non-possession fouls. That's a shutdown corner if I've ever seen one.

Now to the stats for the Thunder's four best offensive players:



Durant
Poss.
Makes
Misses
Fouls
TOs
Points
PPP

Westbrook
Poss.
Makes
Misses
Fouls
TOs
Points
PPP


Aldridge
2
1
1
0
0
2
1

Aldridge
8
1
6
1
0
4
0.5


Anderson
4
1
3
0
0
2
0.5

Anderson
2
0
1
0
1
0
0


Green
10
4
4
1
1
11
1.1

Diaw
1
1
0
0
0
2
2


Parker
1

1
0
0
0
0

Duncan
3
0
3
0
0
0
0


Total
17
6
9
2
1
16
0.941176

Ginobili
1
0
0
1
0
2
2











Green
3
0
2
0
1
0
0


Ibaka
Poss.
Makes
Misses
Fouls
TOs
Points
PPP

Leonard
3
1
2
1
0
2
0.666667


Aldridge
9
5
4
0
0
12
1.333333

Parker
2
2

1

4
2


Diaw
1
0
1
0
0
0
0

Total
23
5
14
4
2
14
0.608696


Duncan
1
0
1
0
0
0
0











Green
1
0
1
0
0
0
0

Kanter
Poss.
Makes
Misses
Fouls
TOs
Points
PPP


Leonard
2
2
0
0
0
5
2.5

Diaw
1
1
0
0
0
2
2


West
1
1
0
0
0
2
2

Leonard
1
0
0
0
1
0
0


Total
15
8
7
0
0
19
1.266667

Marjanovic
3
1
1
1
0
2
0.666667











West
4
1
3

0
2
0.5











Total
9
3
4
1
1
6
0.666667



Some observations:

The Thunder were far too hesitant to have Westbrook attack Leonard. Especially after two fouls in the first half, it didn't make any sense to go away from that. For all the talk about Westbrook not being able to post up Leonard, it's strange that the Thunder didn't try it. Simply put, they won't win by trying to avoid the Spurs' premier defender. They just aren't versatile enough.

Webber is an idiot. The Thunder posted up Durant on at least 10 possessions. I would go back and count if the game weren't about to start. The issue was that they were posting up really far away from the rim -- like almost to the three-point line. Durant is just not built to be able to move his man back that far. Even short distances, he has to absolutely drive his body into his man to the point it could be an offensive foul. Asking him to back his man down consistently just begs for the Spurs to pull the chair on him.

The team defense did a great job at collapsing the paint. Whenever any Thunder player got near the rim, at least two black jerseys were around them. I've never seen a team have so many non-contest pressures in my life. Even though there were quite a few shots where there wasn't a hand in their face, the Thunder players were having to jump at weird angles to get there. The Spurs also had really active hands in preventing layups that began under the rim. So many blocks and deflections there.

All in all, we'll see what happens. I think Kanter will eventually have better games. I also think Roberson won't play as much this upcoming game. His defense is pretty wasted, and obviously he sucks on offense. It's possible that starting Waiters would give the first-unit that third guy they really need to keep Parker honest. Kawhi's going to have to avoid the cheap fouls he picked up last game, and the Spurs have to do a better job at boxing out guys like Adams. There's a path for the Thunder to win this series, but it will disappear if the Spurs just don't lose focus.

[Chinook]

I managed to find enough time to get the raw data for the game and crunch some of the numbers. I didn't have enough time to get to the special requests (I intend to get to them throughout the week), but I have done some work to make collecting and organizing data easier in the future.

For those of you who don't know, I award credit/blame based on who I think is most responsible for the result. There's some subjectivity in there, but it comes with basic guidelines:

1) Players own their own fouls and blocks and almost always their own steals: Getting credit for these on the official play-by-play means they get the credit for the result even if I think someone else may have been more responsible.

2) Credit is slanted toward help defenders. If a defender gets beat off the dribble and another rotates, the new defender will get credit for the play if they force a miss, but if they don't force a miss, then the original defender gets credit. This does not include preplanned rotations like in the PnR. The help guy is expected to defend their opponent, and failing to do so goes against them.

3) Almost everything is credited. There's only one stat in this game that doesn't have an owner, and that's Roberson's turnover off that weird inbounds pass. I don't have a "transition" or "multiple" category like some analyses do. If you fail to pick up your man on the break or fail to get the switch, then you're responsible for the score. If you get a tech, you're responsible for that point they score on the free throw.

If you have questions about the data, I have a more detailed spread sheet than I did last year. Of course there's room for debate on plays, and I welcome it. Fire away, but please don't JUST do that. If you think my method is flawed for what it's trying to do, have counter examples or suggestions on how to improve.

[BillMc]

I managed to find enough time to get the raw data for the game and crunch some of the numbers. I didn't have enough time to get to the special requests (I intend to get to them throughout the week), but I have done some work to make collecting and organizing data easier in the future.

For those of you who don't know, I award credit/blame based on who I think is most responsible for the result. There's some subjectivity in there, but it comes with basic guidelines:

1) Players own their own fouls and blocks and almost always their own steals: Getting credit for these on the official play-by-play means they get the credit for the result even if I think someone else may have been more responsible.

2) Credit is slanted toward help defenders. If a defender gets beat off the dribble and another rotates, the new defender will get credit for the play if they force a miss, but if they don't force a miss, then the original defender gets credit. This does not include preplanned rotations like in the PnR. The help guy is expected to defend their opponent, and failing to do so goes against them.

3) Almost everything is credited. There's only one stat in this game that doesn't have an owner, and that's Roberson's turnover off that weird inbounds pass. I don't have a "transition" or "multiple" category like some analyses do. If you fail to pick up your man on the break or fail to get the switch, then you're responsible for the score. If you get a tech, you're responsible for that point they score on the free throw.

If you have questions about the data, I have a more detailed spread sheet than I did last year. Of course there's room for debate on plays, and I welcome it. Fire away, but please don't JUST do that. If you think my method is flawed for what it's trying to do, have counter examples or suggestions on how to improve.

Many, many thanks for the excellent work. :bobo I'll hope to give a more informed comment after the game. :flag:

[rastaspur]

Thanks in advance. Good to see folks getting enthusiastic and posting something with substance. Love the breakdown. Excellent job.

[GSH]

Wow, that's going to take some time to digest. That looks like they didn't test Duncan much at all. It will be interesting to watch that tonight. And Danny did a pretty damn good job, considering what was thrown at him.

That's a hell of a lot of work. I want to come back to it after the game. :tu

[Chinook]

As a correction, I had Payne's three in Kawhi's row rather than Green's (they're adjacent, as you can see from the charts). Here is the corrected table for each:



Player
Poss.
Makes
Misses
Fouls
TOs
Points
PPP


Green
17
5
8
2
2
15
0.882353


Leonard
6
3
2
3
1
7
1.166667

[DPG21920]

You say LMA was the best defender but when you look at volume plus PPP it seems DG grades out best, no?

[Chinook]

You say LMA was the best defender but when you look at volume plus PPP it seems DG grades out best, no?

Yeah, as I mentioned, LMA got saddled with Ibaka going off for those few minutes in the third. He was awesome, awesome in his help.

[DPG21920]

Yeah, as I mentioned, LMA got saddled with Ibaka going off for those few minutes in the third. He was awesome, awesome in his help.

Got ya

[Em-City]

bump for game 3?

[spurraider21]

Green on Durant has probably been the best move Pop has made this series. If Durant is going to hit ridiculous contested jumpers, it doesn't matter if it's Danny or Kawhi on him. But Danny pesters him quite a bit

[SpursFan86]

Sorry to ask of this, but did you ever finish this Chinook ? Curious about Westbrook's/KD's numbers when Kawhi was the primary defender.

[Chinook]

Sorry to ask of this, but did you ever finish this Chinook (http://www.spurstalk.com/forums/member.php?u=37557) ? Curious about Westbrook's/KD's numbers when Kawhi was the primary defender.

I haven't yet. I am still working tons of overtime (like seriously haven't had a day off since the second GS game). When that's over (and it could be months yet), I'll do some breakdowns.

[SpursFan86]

I haven't yet. I am still working tons of overtime (like seriously haven't had a day off since the second GS game). When that's over (and it could be months yet), I'll do some breakdowns.

No worries/rush, like I said was mainly just curious to see how KD and Westbrook did when Kawhi was guarding them. I had Finals going on during most of the series so I missed a good amount of the games and wasn't able to watch. Know that Durant had some huge games down the stretch but not sure whether he was torching Kawhi or just doing most of his work when guys got switched or Kawhi was guarding Westbrook.

[Kawhitstorm]

No worries/rush, like I said was mainly just curious to see how KD and Westbrook did when Kawhi was guarding them. I had Finals going on during most of the series so I missed a good amount of the games and wasn't able to watch. Know that Durant had some huge games down the stretch but not sure whether he was torching Kawhi or just doing most of his work when guys got switched or Kawhi was guarding Westbrook.

Kawhi switched onto KD when he was rolling, he was basically asked to put away the fire & it didn't work.

[LittleCriminal]

https://twentysomethingnerd.files.wordpress.com/2014/07/dog-pooping1.gif

Here is my analysis...