Player Pairs - Mar. 7, 2012

[timvp]

For this version of Play Pairs, I did it a little differently. This time, I eliminated lineups that included Cory Joseph, Eric Dawson, Malcolm Thomas or Ike Diogu. I also eliminated any lineup that didn't play together for a total of at least two minutes all season combined. In theory, doing it this way eliminates most of the garbage minutes played so far this season.

And by popular demand, I also added a fourth chart that shows the minutes each pair played together.




Point Differential Per 100 Possessions
http://oi44.tinypic.com/34o4w9w.jpg


Points Scored Per 100 Possessions
http://oi42.tinypic.com/5dwb45.jpg


Points Allowed Per 100 Possessions
http://oi41.tinypic.com/f21m5v.jpg


Minutes Together
http://oi43.tinypic.com/212zqf8.jpg
Green: More than 50% of playing time with that teammate
Yellow: 25-50% of playing time with that teammate
Red: Less than 25% of playing time with that teammate





Observations:

-Anyone wondering if the prime days of Duncan are behind us just has to look at those numbers :(

-While his sample sizes are still small, by all indications Manu Ginobili is an offensive wizard. Ginobili and Splitter have been ungodly together.

-Speaking of Tiago Splitter, his numbers are great. He's only being Manu Ginobili when it comes to offense and, of the players who have played major minutes, only behind Danny Green in defense.

-Danny Green's defense shines. Though he has trouble getting around screens, that apparently isn't as big of a deal as a lot of us think. Can't really argue about Green starting when looking at these numbers.

-Tony Parker's numbers are solid. Taking Joseph out of the equation hurt him color-wise but that will probably change once his current backups increase their sample size.

-TJ Ford looks like he deserves a chance to be the backup point guard. His defensive numbers are really good and the early returns on Ford paired with Ginobili are very positive. That said, there are two big concerns with Ford's numbers: So far he hasn't been a good fit next to Splitter or Neal -- and since those are arguably the two most important bench players, that's a pretty huge deal.

-Gary Neal's offense is great. His defense isn't. In other words, as expected.

-Matt Bonner helps the offense, although his defensive numbers are slipping a little bit. And Bonner next to Blair is, as we all know, a HORRIBLE idea.

-Blair kills the offense. Surprisingly though, he's making pretty big improvements on the defensive end.

-RJ also kills the offense. But he's basically a better version of Blair on both ends.

-Kawhi Leonard's numbers are still bad. The hope remains that if he's partnered with Ginobili, he could become a useful player (statistically speaking).

-James Anderson sucks with the starters but is pretty useful with the bench unit.

[urunobili]

awesome analysis!

[timvp]

With Neal starting for Green tonight, that looks like bad news for anyone who is a fan of defense :depressed

[Dex]

Green and Ginobili :wow

+58.71 Differential, 121.21 Points Scored per 100, 62.50 Points Allowed per 100. Maybe just due to the small sample size....would be interested to see how those trends develop.

Ford & Ginobili are quite an interesting combo as well.

[DPG21920]

Was just about to post that. I don't get what pop is thinking. Maybe pop loves Gino/Green together.

[FuzzyLumpkins]

Was just about to post that. I don't get what pop is thinking. Maybe pop loves Gino/Green together.

I'm with Dex. I can see why he would.

[Cant_Be_Faded]

Maybe its cuz I haven't had my coffee yet, but I don't get green, yellow red.

Care to explain it again?


And I can't wrap my mind around Leonard's numbers being this bad.

[TJastal]

Now we have a "minutes played together" as per several earlier requests? Dang, we are spoiled around here....

[Bruno]

Nice work, timvp. :tu

In a previous thread (http://www.spurstalk.com/forums/showthread.php?t=191194), you had the Duncan/Splitter pair scoring 97.81 and allowing 103.87 points per 100 possessions.
In this thread, they are scoring 102.21 and allowing 100.56 points per 100 possessions.
In both case, the sample seems to be the same (99 minutes).

I guess these differences are because numbers comes from different sources, which one seems to be the most accurate?



-Matt Bonner helps the offense, although his defensive numbers are slipping a little bit.


Bonner helping the offense has been a constant for 4 years.

Points scored per 100 possessions when Bonner is on the court compared to when he isn't:
08-09: +6.2
09-10: +5.2
10-11: +3.4
11-12: +6.5

The whole spacing the floor isn't as laughable as it seems...

[manufan10]

Green: More than 50% of playing time with that teammate
Yellow: 25-50% of playing time with that teammate
Red: Less than 25% of playing time with that teammate

[Fabbs]

Now we have a "minutes played together" as per several earlier requests? Dang, we are spoiled around here....
We do?
The Feb 10th timvp post had
Tim Duncan and Tiago Splitter
98.92 minutes

Todays Mar.7, 2012 post shows 99 minutes for Splittsville and Duncan.

[Cant_Be_Faded]

That's for the third chart. What about first two?

[timvp]

Nice work, timvp. :tu

In a previous thread (http://www.spurstalk.com/forums/showthread.php?t=191194), you had the Duncan/Splitter pair scoring 97.81 and allowing 103.87 points per 100 possessions.
In this thread, they are scoring 102.21 and allowing 100.56 points per 100 possessions.
In both case, the sample seems to be the same (99 minutes).

I guess these differences are because numbers comes from different sources, which one seems to be the most accurate?

Hmmm, good find :tu

As I said in that opening, I took out minutes played with the scrubs and lineups that have played together less than two minutes. I'm sure that's the discrepancy but I'll check.

[DPG21920]

Spacing of the floor is laughable in the playoffs tbh when teams actually have time to game plan.

[TJastal]

We need some fact checking police in this thread tbh. Slackers....

[ChumpDumper]

Spacing of the floor is laughable in the playoffs tbh when teams actually have time to game plan.No, it's still quite important. Criticism of Bonner's playoff performance is warranted, but don't be an idiot.

[ChumpDumper]

We need some fact checking police in this thread tbh. Slackers....We don't need more whining from butthurt haters like you.

[timvp]

Hmmm, good find :tu

As I said in that opening, I took out minutes played with the scrubs and lineups that have played together less than two minutes. I'm sure that's the discrepancy but I'll check.

Yeah, in total, Duncan and Splitter have played together for 111.8 minutes. However, 6.1 of those minutes were eight lineups that played a combined total of less than 1.3 minutes together (it looks like most of these lineups Pop put on the court at the end of quarters for defensive purposes). In addition, Duncan and Splitter played 6.7 minutes together with Joseph on the court.

In those 6.1 minutes of end-of-quarter lineups, the Spurs were outscored 11-9. In the Joseph minutes, the Spurs were outscored 4-2. So that's 12.8 minutes that didn't make the cut, Duncan/Splitter was outscored 15-11.

[Bruno]

Spacing of the floor is laughable in the playoffs tbh when teams actually have time to game plan.

You're right, sorry DPG.
http://assets.sbnation.com/imported_assets/82056/horry7.jpg

[DPG21920]

You're right, sorry DPG.
http://assets.sbnation.com/imported_assets/82056/horry7.jpg

Was talking about Bonner, which is whom you referenced. Also, one could argue defense might have made that guy pretty valuable as well.

[Bruno]

Yeah, in total, Duncan and Splitter have played together for 111.8 minutes. However, 6.1 of those minutes were eight lineups that played a combined total of less than 1.3 minutes together (it looks like most of these lineups Pop put on the court at the end of quarters for defensive purposes). In addition, Duncan and Splitter played 6.7 minutes together with Joseph on the court.

In those 6.1 minutes of end-of-quarter lineups, the Spurs were outscored 11-9. In the Joseph minutes, the Spurs were outscored 4-2. So that's 12.8 minutes that didn't make the cut, Duncan/Splitter was outscored 15-11.

Thanks for looking up.

So the Splitter/Duncan pair isn't as bad offensively as the other thread hinted. This pair isn't far from the Blair/Duncan pair which is somewhat logical. It's encouraging because that's the pair who should end games even if Pop hasn't taken that road for the moment.

[DPG21920]

So if Blair/Tim seems to be similar to Tiago/Tim why do people insist that Blair fits next to Tim while Tiago does not.

[timvp]

Here are the raw Splitter/Duncan stats:

Minutes: 111.8 (203 possessions by both teams)
FG%: 77-186 for 41.4%
3P%: 15-38 for 39.5%
FT%: 35-53 for 66%
Points: 204 (or 100.49 points per 100 possessions or 87.58 points per 48 minutes)

Opp FG%: 83-186 for 44.6%
Opp 3P%: 9-29 for 31%
Opp FT%: 28-38 for 73.7%
Opp Points: 203 (or 100.00 points per 100 possessions or 87.16 points per 48 minutes)





Pretty interesting. Bad two-point and free throw shooting but really good three-point shooting. Also a big advantage at the line. Defensively, doesn't give up many three-point shots or free throws and their field goal percentage defense is decent.

[ChumpDumper]

So if Blair/Tim seems to be similar to Tiago/Tim why do people insist that Blair fits next to Tim while Tiago does not.I'm not sure people are positively saying Blair fits next to Duncan -- more like Blair doesn't fit anywhere else.

[timvp]

If you take out the Joseph minutes and the little used lineup minutes like I did above, here are the adjusted stats of Duncan/Splitter:

Minutes: 99 minutes (181 possessions for the Spurs, 179 possessions for the opp)
FG%: 69-166 for 42.9%
3P%: 15-35 for 42.9%
FT%: 32-47 for 68.1%
Points: 185 (or 102.21 points per 100 possessions or 89.70 points per 48 minutes)

Opp FG%: 74-166 for 44.6%
Opp 3P%: 8-27 for 29.6%
Opp FT%: 24-31 for 77.4%
Opp Points: 180 (or 100.56 points per 100 possessions or 87.27 points per 48 minutes)




Even better three-point shooting and better three-point percentage defense. The low two-point percentage is a concern but other than that, those numbers look promising enough for Pop to give it more of a try.

[TJastal]

If you take out the Joseph minutes and the little used lineup minutes like I did above, here are the adjusted stats of Duncan/Splitter:

Minutes: 99 minutes (181 possessions for the Spurs, 179 possessions for the opp)
FG%: 69-166 for 42.9%
3P%: 15-35 for 42.9%
FT%: 32-47 for 68.1%
Points: 185 (or 102.21 points per 100 possessions or 89.70 points per 48 minutes)

Opp FG%: 74-166 for 44.6%
Opp 3P%: 8-27 for 29.6%
Opp FT%: 24-31 for 77.4%
Opp Points: 180 (or 100.56 points per 100 possessions or 87.27 points per 48 minutes)




Even better three-point shooting and better three-point percentage defense. The low two-point percentage is a concern but other than that, those numbers look promising enough for Pop to give it more of a try.

That makes perfectly good sense. With a strong interior, guys out on the wing can stay home with their assignments. Sounds like a recipe for winning playoff games to me.

[DPG21920]

I'm not sure people are positively saying Blair fits next to Duncan -- more like Blair doesn't fit anywhere else.

I think the "Blair fits next to Tim" crowd is more referencing trades rather than Tiago. They are weary of trades because they say "Blair fits" even though they say Tiago doesn't when in reality it's pretty close. So in the context of discussing trades and bringing in a solid defender with limited offense I think this shows it can be successful.

[ChumpDumper]

I think the "Blair fits next to Tim" crowd is more referencing trades rather than Tiago. They are weary of trades because they say "Blair fits" even though they say Tiago doesn't when in reality it's pretty close. So in the context of discussing trades and bringing in a solid defender with limited offense I think this shows it can be successful.Oh, you just want to work in your trade agenda here.

Carry on.

[Bruno]

Was talking about Bonner, which is whom you referenced.

You were talking about the spacing skill. It's a damn useful skill both in regular season and in playoffs.

[timvp]

In a previous thread (http://www.spurstalk.com/forums/showthread.php?t=191194), you had the Duncan/Splitter pair scoring 97.81 and allowing 103.87 points per 100 possessions.

And the last piece of the puzzle...

In the ten games since that thread you linked to, Duncan and Splitter have played a little bit more together. Here are the results:

Minutes: 12.7 (22 possessions for Spurs, 24 possessions for opp)
FG%: 11-25 for 44%
3P%: 1-2 for 50%
FT%: 2-5 for 40%
Points: 25 (or 113.64 points per 100 possessions or 94.49 points per 48 minutes)

Opp FG%: 7-28 for 25%
Opp 3P%: 0-6 for 0%
Opp FT%: 1-2 for 50%
Opp Points: 15 (or 62.5 points per 100 possessions or 56.69 points per 48 minutes)




:wow (yeah, small sample size but still :wow)

Pop, play them more together. Seriously. What are you waiting for?

I might have to go MultiFabbs up in hurr :pctoss

[DPG21920]

You were talking about the spacing skill. It's a damn useful skill both in regular season and in playoffs.

With regards to bonner.

I've never bashed spacing; I've bashed Matt's playoff performances.

[DPG21920]

Oh, you just want to work in your trade agenda here.

Carry on.

:lol What agenda? I have no influence and I was making an observation about a ST opinion in context of what the stats show. I'm pretty sure in the scheme of things everyone would welcome a trade for a defensive big; is that everyone's agenda?

[ChumpDumper]

I've never bashed spacingYou should rephrase then.
Spacing of the floor is laughable in the playoffs tbh when teams actually have time to game plan.

[ChumpDumper]

:lol What agenda? I have no influence and I was making an observation about a ST opinion in context of what the stats show. I'm pretty sure in the scheme of things everyone would welcome a trade for a defensive big; is that everyone's agenda?:lol so defensive.

It's not a bad thing to have an agenda.

[DPG21920]

You should rephrase then.

Agreed. Although I thought it was implied since the spacing comment I was referencing was all about Bonner.

[timvp]

DPG pulling a TD21. He's changed :lol

[DPG21920]

Dude. No.

[timvp]

Dude. No.


I did not get to watch the live game because I was at the Kings of Leon concert. I am watching the replay and these are my observations for the first half.

People have been way overreacting. The first half was solid overall. I like how they are going to Tim and spacing the floor.

You used to like spacing now you say it's "laughable". :nope

[timvp]

Maybe its cuz I haven't had my coffee yet, but I don't get green, yellow red.

Care to explain it again?

A green cell means the player in the column has played at least 50% of their minutes with the player in the row. A yellow cell means the player in the column has played between 25% and 50% of their minutes with the player in the row. A red cell means the player in the column has played less than 25% of their minutes with the player in the row.

So por ejemplo, Duncan has played only 119.2 minutes with Ginobili and that's just 12.9% of his total of 923.8 minutes played -- so that cell is red. On the other hand, he's played 862 of his minutes with Parker, which is equal to 93.3% of his total minutes and thus the green cell. Duncan's total with Neal (300.3 minutes) is an example of a yellow cell (because it's 32.5% of Duncan's total minutes).

Let me know if it still doesn't make sense. Thanks :tu

[DPG21920]

:lmao. Just because you sold your Spurs soul tbh.

"I can't support that"

"variables have changed aka basketball reasons"

Uncle

[timvp]

^:lol DPG with the subtle goods. Just messin' with ya brosef.

[jag]

timvp, do you by any chance use Excel to crunch numbers? If so, would you be able to upload some of your analysis or post it as code so that it could be re-used (copy/pasted into Excel)?

[timvp]

yeah i dont get the green and red on the first graph

For the first three charts:

A green cell means that the player in the column improved the number for the player in the row. A red cell means the opposite. So basically, the more green in a player's column the better.

[timvp]

timvp, do you by any chance use Excel to crunch numbers? If so, would you be able to upload some of your analysis or post it as code so that it could be re-used (copy/pasted into Excel)?

It's currently not in excel but that's a good idea for the next time I do it. Perhaps I can publish it in Google Docs so others can use the data.

[jag]

It's currently not in excel but that's a good idea for the next time I do it. Perhaps I can publish it in Google Docs so others can use the data.

Thanks, that'd be awesome.

[therealtruth]

That makes perfectly good sense. With a strong interior, guys out on the wing can stay home with their assignments. Sounds like a recipe for winning playoff games to me.

That would have helped against the Grizzlies last year. The Spurs gave them way too many open perimeter shots because of over helping. The teams is not athletic enough to play aggressive help defense against some of the teams in the league.

[TimmehC]

It's nice to see proof of how bad the Bonner/Blair combo is, though there(thankfully) hasn't been much of that this season. The Blair/Kawhi combo is also completely horrible, but Pop doesn't seem to realize that based on the minutes.

[TD 21]

DPG pulling a TD21. He's changed :lol

timvp pulling a timvp, by misconstruing things. To steal a phrase from Mel_13, I don't "move the goal posts". Like many, you just lack reading comprehension.

A couple of observations . . .

- These numbers really don't do Duncan and Leonard justice. Everyone else, they more or less do.

- I knew the 3-point defense would be significantly better with Duncan paired with Splitter. Everyone automatically think rim protection, but they help the 3-point defense too. Because they're both capable of defending the post straight up in most match-ups, the perimeter players don't have to offer as much help and can stay attached to 3-point shooters more often, which inevitably leads to less clean looks.

[jjktkk]

timvp pulling a timvp, by misconstruing things. To steal a phrase from Mel_13, I don't "move the goal posts". Like many, you just lack reading comprehension.



Strange when anyone disagrees with TD 21, we've , all of a sudden, lost the ability to comprehend reading the English language. :lol

[Seventyniner]

My biggest takeaway from this is that Blair and Bonner have only shared the court for 62.2 minutes this season. That's less than 2 per game, though I'm sure there are more in blowouts that included Joseph et al.

Maybe Pop is learning?

Great work as always, timvp.

[Drom John]

Nice of course, but I dislike pulling stats to get rid of some players. Stats are better with bigger samples. Keep the stats with Joseph, et alia, or the 2 minute or less lineups.
Too much clutter? Then don't show Joseph and company while keeping the full numbers for others. Prune after compiling, not before.

Redundant comment, but it's not that Splitter doesn't fit with Duncan, it's that Blair doesn't fit with Bonner.

Those specialty end of quarter defensive lineups count too. So count them. However, as mentioned in another thread (I think by timvp), 1) Pop is often more interested in running down the clock than increasing the differential and 2) those specialty lineups probably have significantly more defensive possessions than offensive possessions, so, a) they should have a negative differential, and b) not counting them means there's a slight overrepresentation of the offensive units at end of quarters.

[ElNono]

So, Manu and <whoever> is always the best combo, right? :lol