Advanced Scouting: San Antonio Spurs at Utah Jazz

by

San Antonio Spurs at Utah Jazz at 9:00 CST November 19, 2010

SRS ranks:

San Antonio: 7.03 (5th)
Utah: 1.28 (13th)

Key Player Statistics (courtesy of 82games.com (effective November 18, 2010) and basketballvalue.com:

Utah Jazz Player Ratings

PlayerFraction of team minutesPER minus Counterpart PER2 Year Net PEROn court +/-per 48 minutesOff court +/-per 48 minutes2 Year Adjusted +/-
Williams0.87.47.71.21.27.92
Millsap0.777.21.33-4.53.27
Jefferson0.74-2.11.4-4.216.8-7.38
Kirilenko0.78.38.311.60.65
Bell0.53-11.4-4.4-5.28.5N/A
Miles0.472.7-4.713.2-9.3-0.64
Hayward0.22-13.7N/A-17.26.3N/A
Price0.21-7.2-6.81.61.1-1.42
Elson0.18-2.9-7.610.6-0.9N/A
Watson0.17-5.1-3.83.80.72.9
Fesenko0.15-10.2-6.816.6-1.59.54

One thing that sticks out is Al Jefferson’s low adjusted plus minus figure. Although adjusted plus minus (APM) is perhaps the most unbiased estimate of value, it is quite noisy over one (or two) seasons. By unbiased, I mean that while measures such as PER, Offensive Rating, etc consistently overvalue certain types of players (and undervalue others), APM doesn’t tend to overvalue any specific type of player. If a player is overvalued by APM, it is largely due to luck. The problem is that luck plays a much larger factor in APM. Less noisy measures of Al Jefferson’s APM use 5 years of data and range from -1.5 to 2.0.

Most valuable/utilized lineups:

Williams, Bell, Kirilenko, Millsap, Jefferson -16 in 205 Minutes
Williams, Miles, Kirilenko, Millsap, Jefferson +46 in 61 Minutes

Preferred method to create shots: Cuts, rolls and passes

Deron Williams is reasonably good at scoring on isolations and Al Jefferson has been among the best post scorers for years, but the Jazz offense is all about cutting, screening and passing to create the easy shot. The Jazz have scored the most points resulting from cut plays, as defined by Synergy, by over 20% more than the next highest team. Last year they were nearly 40% greater than second best! Utah also ranks at the top of the list at scoring from screens. I found a video with a very informative take on how the Jazz offense is run.

Appropriate Spurs counter: stay home

Utah’s homecourt advantage is well-documented, but by suggesting the Spurs “stay home”, I mean that they shouldn’t leave their defensive man assignment as frequently as is typical for them.

The Jazz had the highest percentage of field goals assisted on last year. In their 29 losses, about 36% of field goals were unassisted; compared to a 30% rate in their 53 wins. This decrease by about 20% was among the most significant in the league. It seems that forcing the Jazz to create unassisted shots is a good way to increase your chances of success.

Matt Bonner could actually be a significant defensive upgrade from Dejuan Blair in this respect. Bonner has performed better statistically on help defense plays such as defending the pick and roll and contesting the spot up shooter.

Defensive Weakness: run Al Jefferson off the pick and roll

Al Jefferson has never been know for his defense. In particular, he often struggles with defensive anticipation and taking appropriate angles for cutting off the pick and roll ballhandler when his man sets a pick. He can stand to learn from his injured teammate Mehmet Okur. Over the two previous years, Jefferson has allowed 0.94 points per possesssion (PPP) on pick and roll plays when defending the player setting the pick. Over that same period, Okur has been among the best in the league, allowing only 0.75 PPP.

One thing that is concerning for the Spurs is that Jefferson has only been credited with 4 such plays so far this year. Perhaps the Jazz have figured an effective way to hide him.

Once again, Manu and Parker are the Spurs best options to exploit Jefferson’s defensive weakness against the pick and roll. Bonner’s value might be partially offset when Jefferson is in the game because Bonner has not been utilized as frequently or successfully in the pick and roll as Blair. However, using Bonner on a pick and pop could better help isolate Ginobili/Parker on Jefferson and create mismatches.

Conclusion

The Spurs have out performed the Jazz to this point in the season, but it’s always difficult to go to Salt Lake City and come out victorious. The Jazz have most the home wins since 2006/07.

Other things to watch for

*Since 2006/07, the Spurs are tied with Lakers and Celtics for the most road wins, the Jazz stand alone with the most home wins over the same time period. The Spurs are tied for 5th in home wins over the same period and the Jazz are tied for 11th.

*Tim Duncan has a chance to break David Robinson’s Spurs NBA franchise record for scoring the same day he breaks his record for games played. He is currently 12 points behind the Admiral’s career scoring total. Duncan remains 2,824 behind George Gervin’s ABA/NBA franchise mark of 23,602.

  • Tyler

    Agreed. Run Al Jefferson through PnR’s the entire game. He doesn’t have the lateral quickness to stay with any of our guards. Make him work on that end of the court. When he was with MN, I saw him routinely take possessions off.

    Also, the Jazz really don’t have a shot blocker. Both guards and bigs need to attack the rim, at least early in the game.

    I would stay at home on Deron Williams as stated above - make him a scorer similar to what we’ve done with Chris Paul in the past. If the rest of his team doesn’t get going, I think we have a better shot.

    Should be a good test in a tough environment.

  • AP

    So do these awesome previews come with a Sereday game prediction?

  • Pingback: Tweets that mention Spurs Stats | Advanced Scouting | at Utah Jazz 11/19/2010 | 48 Minutes of Hell -- Topsy.com()

  • Tyler

    Also, not that it should affect a whole lot, but Jerry Sloan will be attending a funeral and won’t be on the bench tonight. Just something to think about.

  • Pingback: Tweets that mention Spurs Stats | Advanced Scouting | at Utah Jazz 11/19/2010 | 48 Minutes of Hell -- Topsy.com()

  • Nick (Italy)

    “Deron Williams is reasonably good at scoring on isolations”

    I am quite curious about this: from the eye test, I would say that Williams is very good in iso plays (but maybe it is because I am still trying to scratch away from my retina certain plays such as the dunk he threw on SA - was it last year?).

    Do you have numbers to put Williams’s ability in ISO in perspective?

  • Dogmer

    I am confused about two aspects of your column. In the table, columns 5 and 6 have the same title: Off court +/-per 48 minutes. Is the first supposed to be “on court +/-per 48 minutes/?”

    Second, what sense are we to make of the sentence “In their 29 losses, about 36% of field goals were unassistedab; compared to 30% of field goals were unassisted in losses?” Presumably, 36% came in wins?

    Being an anal retentive, the first column in your table should be titled “Fraction of teams minutes” and not “Percentage of teams minutes.” Or does Williams play only 0.8% of the time.

  • Espoon

    The Jazz have shown that they can get behind. The Spurs need to get out to a good start and don’t let the Jazz have another comeback. The 3 keys are defense, turnovers, and freethrow shooting. They may get away with a poor performance in one of these areas if they are hitting 3’s or just have a high fg %.

  • BlaseE

    Excellent previews….I haven’t seen anything even remotely comparable in quality for any team online (not that I look at other teams that often).

  • http://www.48minutesofhell.com Scott Sereday

    I apologize for the typos. They should be fixed now. It should be noted that the Jazz won more frequently when they had a higher percentage of assisted scores. I think all but two teams had similar pattern, but the the ratio between the two for the Jazz was among the most significant.

    Last year the Jazz scored 1.063 PPP on Deron Williams Pick and Rolls Including Passes (90th percentile) but “only” 0.945 on such Isolation Plays (74th percentile).

  • Greyberger

    I think the “2-year APM” from Basketballvalue.com is last season and the season-to-date, so it’s only 88 games total in Jefferson’s case. Basically just 1-year APM.

    Looking at APM for 2008 and 2009 might give the right mix of large and recent sample, but I don’t know where to find that specifically.

  • zainn

    Here’s a good advantage for us, Jerry Sloan is going to a funeral, so he won’t be attending the game!

  • Greyberger

    Nvm, found it. Al’s 2-year adjusted +/- for the 2008 and 2009 seasons is +1.17, with a standard error of 2.94. So a more positive and slightly more confident assessment than using 09-11, which gives us -7.38 and a SE of 3.74.

  • Greyberger

    Assist data can be illuminating but also frustrating. With the Jazz and ball movement, what we really want to get at is what percentage of their attempts would have been assisted, whether they actually went in or not. We can only see the ‘makes’ side of the story in the numbers, when what concerns us most directly is forcing ‘unassisted misses’.

    Percentage of buckets assisted is one way to look for ball movement and it certainly seems to be relevant to this Jazz team. If the Spurs can disrupt the Jazz offense to the point where they’re ending possessions with jumpshots vs. a set defense or by turning it over, it will only feed our fast break offense.

  • Jim Henderson

    From the main post:

    “Matt Bonner could actually be a significant defensive upgrade from Dejuan Blair in this respect. Bonner has performed better statistically on help defense plays such as defending the pick and roll and contesting the spot up shooter.”

    A couple of points. Number one, where’s your evidence for this “significant defensive upgrade”? And number two, some of your evidence for Jazz pick & roll precision was based on last year’s Jazz, which had one of the most seasoned, polished, and effective pick and roll combo’s (w/Boozer) since Stockton and Malone. Third, what is your source for the following comment?:

    “Utah also ranks at the top of the list at scoring from screens.”

    “Bonner’s value might be partially offset when Jefferson is in the game because Bonner has not been utilized as frequently or successfully in the pick and roll as Blair.”

    Glad to see at least some balance in the Blair/Bonner assessment.

    “However, using Bonner on a pick and pop could better help isolate Ginobili/Parker on Jefferson and create mismatches.”

    Except that Bonner doesn’t do pick & pop’s?!

  • Tyler

    @Jim

    “Except that Bonner doesn’t do pick & pop’s?!”

    What do you mean? That’s the only type of PnR we use when Bonner sets ball screens. He’s not rolling to the rim….

  • A-Train

    Nice analysis. You are going deep into the numbers man, don’t get lost!

  • Jim Henderson

    Tyler
    November 19th, 2010 at 1:31 pm

    “What do you mean? That’s the only type of PnR we use when Bonner sets ball screens. He’s not rolling to the rim….”

    Bonner get’s the great majority of his looks in a motion offense, and from guard penetration and kick-outs. Yes, he does try to take advantage of running off some “screens”, but at least in my view “pick & pops” really describe a guy that has a mid-range game. Most picks are not initiated out near the three point line, where Bonner is the most effective, and takes the great majority of his shots. Bonner’s picks are not particularly effective in my view, because he’s not a mid-range shooter, and he’s not a good roller.

  • Bankshot21

    I’m looking forward to this game. How about that Milsapp/Blair match up?!…or Duncan/Jefferson…or TP/D.Will?!…or RJ/Kirilenko…or Manu/?….that’s where it stops. Manu goes off. Instead of Bonner in the pick and pop how about McDyess? (Hello there Jim) this will be a good game. Not a measuring stick win or lose. Just a nice game for NBA fans. Wish J.Sloan was there. If we win they will say our 10-1 record is an anomaly. #gospursgo!!!!

  • Greyberger

    Bonner has a role in the screen and roll, but it’s not the screen. (or the roll). Instead Pop will play him with Blair or Dice or Duncan and they will set the screen for the ball handler, while Bonner just draws his defender away from the play.

    Other than that, every now and then Bonner himself will set a screen for Manu or Tony and then take a big step or two away from the action and back towards the 3pt line. You could call this a pick and pop I guess.

  • Dogmer

    Scott,

    The “Advanced Scouting Column” is a phenomenally great idea. While some of us get into the geeky math/stats element of it, it is probably counterintuitive for most. Regardless of your readers’ abilities to understand the underlying mechanics, it is great that you make specific predictions about how the game can be won. It would be fun if you could provide a summary statement of how your predictions fared after the game is over, especially for those of us who cannot watch many games and are relegated to analyzing box scores.

    Yesterday, you made several predictions:
    1st prediction: Al Jefferson is vulnerable on defense, especially to pick and rolls (P&R), based on APM.

    Result: His +/- for last night’s game seems to bear this out (-18 and worst on his team). (Was this attributable to P&R?)

    2nd prediction:. Miles would be a significant upgrade over Bell.

    Result: +/- for Miles was +5; Bell was -14. Yikes, can you get any better?

    3rd prediction: Manu and Tony are the Spurs best options to exploit Jefferson’s defensive weakness.

    Result: This appears to certainly true for both based on the number of made baskets in the paint and Jefferson’s low +/-. Manu’s +/- were relatively low as his missed many long shots that are less likely the direct result of P&R.

    4th prediction: Bonner should be more valuable than Blair with the caveat that Bonner’s value might be offset when Jefferson is in the game because Bonner has not been utilized as frequently or successfully in the pick and roll as Blair.

    Result: Bonner was a bust and Blair had a very good game suggesting this prediction did not come true. This may be partly due to the fact Matt missed a number of open threes. Had those gone in at his regular rate, this game would have been a rout and you would have been half correct. Did Matt’s back spasms from the previous game contribute to his performance? The fact that Blair had such a good game seems to be unpredictable based on any number of reasons beginning with DeJuan’s play of late.

    5th prediction: Forcing the Jazz to create unassisted shots is a good way to increase your chances of success.

    Result: This was your best prediction and came through in spades. The Jazz had a moderated number of assisted baskets.

    Summary: 4 of 5 predictions were spot on and the fifth could have been correct if Matt played an average game. Cheers!

    Lastly, the link to bballbreakdown was great.

  • Jim Henderson

    Dogmer
    November 20th, 2010 at 7:47 am

    “4th prediction: Bonner should be more valuable than Blair with the caveat that Bonner’s value might be offset when Jefferson is in the game because Bonner has not been utilized as frequently or successfully in the pick and roll as Blair.

    Result: Bonner was a bust and Blair had a very good game suggesting this prediction did not come true.”

    Yes, and that’s the prediction that I clearly questioned in a previous comment on this thread.

    “This may be partly due to the fact Matt missed a number of open threes. Had those gone in at his regular rate, this game would have been a rout and you would have been half correct.”

    We can’t make up a bunch of rationalizations and IF’S just because a prediction didn’t come true. It’s more likely he was just wrong in this assessment. What is Bonner’s regular rate? About 40%, his career average, not 80%, his rate this year before Utah. Bonner simply began to “regress to the mean”, and was not one bit surprising to see in last nights game.

    “The fact that Blair had such a good game seems to be unpredictable based on any number of reasons beginning with DeJuan’s play of late.”

    Not true. Blair has been playing about as good as he did last night for the past 6 games, which is now more than half of the season.

  • http://www.48minutesofhell.com Scott Sereday

    Just to clarify, these were not exactly “predictions”, but rather tendencies. (+/- over one game is very difficult to predict with great accuracy)

    Regarding Bonner’s pick and pop. I was simply theorizing that Bonner is the best Spurs big man to open up the floor and create room for a ballhandler to beat Al Jefferson off the dribble, exploiting Jefferson’s biggest weakness.

  • Nick (Italy)

    Thanks Scott for the data about D-Will in ISO: 74th percentile looks definitely worse than I expected (if I am not wrong, it means some 75-90 players in the NBA last year had better numbers than him in ISO, right?).

    I wonder how much of this could be due to the fact that probably most of the times Williams switches to ISO in “emergency situations” (late in the clock or when a play is broken): does the database allow to eliminate a particular category of ISO plays (e.g. ISO which start with less than 7 secs on the shot clock)?

  • Pingback: Spurs Stats | Advanced Scouting | Utah Jazz 1/26/11()