Statistical Analysis. Humor. Knicks.

Wednesday, October 22, 2014

Knicks 2013 Preview: The Carmelo & Amar’e Conundrum

Amar’e Stoudemire has been paired with Carmelo Anthony for 20 months, and over that time period, when the pair have played in the same game the Knicks’ record is an underwhelming 31-40 (1-7 in playoffs) . When Stoudemire, Anthony, and Tyson Chandler were on the floor together, New York scored only 98.5 points per 100 possessions. To put that in perspective, that level of offensive *ahem* efficiency would put the Knicks in the bottom third of the league.

The individual stats available through NBA.com confirm that the pairing of Anthony and Stoudemire are the problem. Anthony posts average to below average numbers in key stats like points, FG%, FTA, and rebounds—when paired with S.T.A.T. And Stoudemire shows a similar drop in production.

A popular option to address this issue, one endorsed by Zack Lowe, Bill Simmons, and others, is moving Stoudemire to the bench, and allowing Anthony to start the game at the power forward position. Lowe noted that of all the players recorded by STATS, LLC’s multi-camera process for tracking every movement in an NBA game, Anthony was the most effective player in the league at driving the ball. Anthony was also the NBA leader in points per possession on elbow touches.

Anthony is highly effective at driving to the basket from the elbow or beyond, and highly effective at scoring when he catches at the elbow. The problem, as Lowe sees it, is that Stoudemire touches the ball at the elbow nearly twice as often as Anthony, and he was a far less effective scorer from that spot. Simply put Stoudemire is getting in the way of what Anthony does best.

In his NBA preview at Grantland.com, [29:30] Bill Simmons said that Amar’e Stoudemire must come off the bench, and that only a fool would argue against that. To quote Gomez Addams “Well, with God as my witness, I am that fool!”

Now, I agree the statistics make a compelling argument that something has to change, but I’d like to offer a few reasons why moving Stoudemire to the bench won’t be as helpful as some seem to think.

1. Making Stoudemire a reserve will only slightly reduce the amount of time he is on the court with Anthony.

Let’s say Carmelo Anthony plays his career average of 36.2 minutes per game. And let’s argue than in a reserve role, Stoudemire plays 32 minutes per game rather than his career average of 34.5. Even if Stoudemire plays every minute that Anthony sits, he will still be on the court with Anthony for 20.2 minutes per game. Last year, they were on the court together 25 minutes per game (976 minutes over 39 games). So bringing Stoudemire off the bench is likely to give Anthony an additional 4 minutes and 48 seconds without Stoudemire. At last year’s pace (93.2), that would mean an extra 9 possessions per game without Stoudemire.

Coach Woodson can tinker with the minutes by increasing Anthony’s or reducing Stoudemire’s, but each of those answers come with their own problems. Anthony’s career high minutes per game was 38.2 while he was in Denver at the age of 25. It is hard to image him playing that many minutes per game or anything in excess of that at 28, especially after a summer of Olympic Basketball. Stoudemire has never played less than 31 minutes per game, and wouldn’t likely respond well to such a sharp drop in minutes. Then again, maybe making Stoudemire sullen and unproductive is just the thing to get the Wizards to inquire about trading for him.

2. It does nothing to address Stoudemire’s productivity issues

Stoudemire’s True Shooting Percentage and Effective Field Goal Percentage last season—.541 and .487 respectively—were well below his career averages of .596 and .534, and below 2010-2011’s .565 and .505. Stoudemire’s drop in scoring is complicated by the fact that he took about .5 fewer shots at the rim in 2011-2012( 5.7) than he did in 2010-2011 (6.2). Furthermore, he took nearly 2 fewer free throws per game in 2011-2012 (5.7 a career low) than he did in 2010-2011 (7.5).

Even if you want to argue that pairing Amar’e with ‘Melo is the proximate cause of the Stoudemire’s shooting woes [Stoudemire’s shot the same 48% while paired with Anthony as he did while Anthony was on the bench, the big change was in the FGAs (17.3 w/o Anthoony up from 14.1 with Anthony)], we’ve already established that Stoudemire will only have about 11.8 minutes per game to take advantage of this. Moving Stoudemire to the bench won’t solve the issue.

Barring a trade [or injury], there really is no way to keep Stoudemire and Anthony from sharing space. Since they are stuck with each other, the Knicks need to make changes to way they play together in order to make this work. Stoudemire seems to have sensed this and has already began adding new “jewels” to his game. I think adding post moves will be a good thing for Stoudemire. Working in the paint rather than at the elbow will free that space for Anthony. Furthermore, if Stoudemire is making moves in the low post, he may see a rise in the number of personal fouls he draws, something he needs to help his scoring efficiency. Catching in the low post and making a quick move may also reduce his turnover volume, which was slightly above average last year. And it is not as though moving to the low post will get in the way of Chandler and his wide assortment of post moves.

In order for this to work the entire team needs to buy-in. The coaching staff needs to incorporate plays for Stoudemire in the low post. The team needs to know when to deliver the ball to Stoudemire. Finally, Stoudemire needs to embrace his new role and not revert to past habits if he struggles early.

With a full training camp and a full season, the Knicks have time to figure out how to make this work before the playoffs start. Hopefully, the team will find a way to get the best out of Stoudemire and Anthony as a pair because there just aren’t enough minutes in a game to keep them apart.

33 comments on “Knicks 2013 Preview: The Carmelo & Amar’e Conundrum

  1. er

    Good article, with well thought out points. Like you said i think the only hope is stat working top of the key pick and pop and post moves, and let melo work from the elbows and three point line

  2. DS

    I would love to see Amar’e play more like Larry Johnson circa 1999; LJ was a much better passer and 3-point shooter.

  3. sisterray

    I think there’s enough positive data out there already to intrigue us about the idea of Melo playing the 4 while STAT is out (which could be several weeks still). But I agree 100% that the team’s problems can only be solved by addressing them head-on: Melo, STAT, and Chandler need to be able to play simultaneously without crowding each other out.

    There are so many ways that you could make this happen, given that both Melo and STAT are versatile enough: they can both hit jumpers and are outstanding at finishing at the basket. But I haven’t seen our coaches — MDA or Woodson — forcing them to make it happen. Instead, we usually see Melo calling the shots from the wing and STAT hiding on the weak side hoping no one notices that he’s not involved in the play. That’s a recipe for failure, given that STAT hardly gives you anything besides scoring. At the very least, isn’t it obvious that it would be more enjoyable to watch Melo and STAT trying to develop a two-man game — even unsuccessfully! — than to just pretend they’re not on the same team?

  4. d-mar

    Well, Mark Stein doesn’t think much of our team (surprise!), ranking us 15th behind such powers as Atlanta and Utah and just ahead of Dallas and Minnesota.

    Hopefully we can prove all the ESPN and TNT naysayers wrong, starting with a beatdown of media darling Brooklyn on Thursday

  5. DS

    d-mar: Well, Mark Stein doesn’t think much of our team (surprise!), ranking us 15th behind such powers as Atlanta and Utah and just ahead of Dallas and Minnesota.

    Hopefully we can prove all the ESPN and TNT naysayers wrong, starting with a beatdown of media darling Brooklyn on Thursday

    I can live with Atlanta: Korver, Lou Williams, and Devin Harris is not a bad substitute for Joe Johnson. Plus a healthy Horford. Plus Utah with Favors and Kanter a year older and having just made the playoffs in the West.

    It’s the Nets that’s so irritating. Johnson over Brooks is a decent upgrade but moving to Brooklyn does not make you automatically better at basketball. Believe me, I know firsthand.

  6. danvt

    This is why stats based analysis bothers me. You take an absolutely loony toons situation in which these guys came to play together. One featuring two coaches, myriad point guards, huge turn over in the supporting cast, lots of games lost to injury both by the players mentioned and other integral parts of the team, a lockout, and a one week training camp and then point to evidence of how these guys skills duplicate each other. The solution of benching one of the best PFs in the league seems reasonable until you realize that they need to play together anyway. So, the conclusion is most likely they’ll lose unless the coach can, in the immortal words of Mike D’Antoni, “figure something out”.

    Well, even the much maligned “eye test” could tell you that much.

  7. ess-dog

    Eddy Curry is likely to start at center on opening night for the Dallas Mavericks. That’s some crazy stuff.

  8. Z-man

    Nice article, Mike. A couple of other points:
    1. It’s not who starts, but who finishes
    Amare is getting paid $20 million per year for the next 3 years. He was brought here to be a main piece. Whether he starts or comes off the bench is much less relevant to whether he will finish games, especially close games. For better or for worse, it is a given that both players are likely be on the floor at the end of games, especially in the playoffs unless one or the other is hurt. So, to entertain any game plan that is intended to minimize the time they play together is counterproductive.
    2. Both players are multidimensional offensive players that can adapt if they really want to.
    I think this is more true for Melo than for Amare, but I also think that there is more onus on Amare to adapt. Amare has to realize that he does not need to average 20+ points a game on this team, and that’s a GREAT thing for him in that it will keep him healthier and more fresh for a long playoff run. He may have to virtually eliminate the part of his game that is better performed by Melo, such as the iso from the elbow area, and focus completely on P&R, spot-ups and post-ups. In other words, play David Lee’s game with less rebounding and more shot-blocking. He’s just not the iso player he was, or in Melo’s class as an iso player. This is not asking a lot.In fact, it’s asking him to follow in Garnett’s footsteps, who left much of the iso work to Pierce and picked his spots. Amare is no Garnett, but I think the analogy fits. Melo, on the other hand, needs to make Amare better by deveoloping a P&R chemistry with Amare and not always looking to score.
    3. Offensive rebounding is the key
    Melo is capable of being a top rebounder in this league. Amare could do some of the things that Chandler does in terms of tip-outs and such. The problem is, neither guy is committed to this critical aspect of the game. Offensive rebounding is a way to have your teammate’s back and Woody needs to preach…

  9. JK47

    Yeah, I don’t understand all the love for the Nets either.

    The Nets starting frontcourt of Brook Lopez and Kris Humphries is pretty lame. Humphries’ strong rebounding makes up a bit for Lopez’ shortcomings on the glass, but that is not a duo that strikes a lot of fear into me. The Nets’ reserve big men are pretty stinky– one-dimensional Reggie Evans, no-dimensional Andray Blatche and the enigmatic mystery that is Mirza Teletovic. They have one big man who is a decent scorer– Lopez. And Lopez himself is not exactly Wilt Chamberlain– in his last full season he rocked a .549 TS%.

    The reserve wings and guards are a who’s who of low TS% players:

    Josh Childress .501
    Marshon Brooks .502
    Jerry Stackhouse .490
    CJ Watson .491

    That team is lame.

  10. er

    lol love how you put that…they are lame. I thought they looked awful against the knicks save for dwill and ty taylor

    JK47:
    Yeah, I don’t understand all the love for the Nets either.

    The Nets starting frontcourt of Brook Lopez and Kris Humphries is pretty lame.Humphries’ strong rebounding makes up a bit for Lopez’ shortcomings on the glass, but that is not a duo that strikes a lot of fear into me.The Nets’ reserve big men are pretty stinky– one-dimensional Reggie Evans, no-dimensional Andray Blatche and the enigmatic mystery that is Mirza Teletovic.They have one big man who is a decent scorer– Lopez.And Lopez himself is not exactly Wilt Chamberlain– in his last full season he rocked a .549 TS%.

    The reserve wings and guards are a who’s who of low TS% players:

    Josh Childress .501
    Marshon Brooks .502
    Jerry Stackhouse .490
    CJ Watson .491

    That team is lame.

  11. njasdjdh

    The “Amar’e doesn’t need to average 20 PPG so he should work on other things” argument makes sense until you consider that our offense sucked last year.

  12. njasdjdh

    Not sure exactly what you’re referencing, but it’s not just “stats based analysis” that says Amar’e and Melo can’t/don’t work well together. There have been numerous pieces with video/scouting evidence of why the combination has been problematic. And Amar’e is not “one of the best” PFs in the game unless your definition of very best is….extremely broad.

    danvt:
    This is why stats based analysis bothers me.You take an absolutely loony toons situation in which these guys came to play together.One featuring two coaches, myriad point guards, huge turn over in the supporting cast, lots of games lost to injury both by the players mentioned and other integral parts of the team, a lockout, and a one week training camp and then point to evidence of how these guys skills duplicate each other.The solution of benching one of the best PFs in the league seems reasonable until you realize that they need to play together anyway.So, the conclusion is most likely they’ll lose unless the coach can, in the immortal words of Mike D’Antoni, “figure something out”.

    Well, even the much maligned “eye test” could tell you that much.

  13. daJudge

    Hey Thomas B. Nice piece. On a side note, I hope all you folks down in the City are safe and doing OK tonight. Here, in far upstate, near VT border, not much action at all. Anyway, loved the article, certainly reasonable points, but I’m one of those who like Stat off the bench. My own view is to “keep ‘em separated”…..To me, we would be well-served if we can give Stat more minutes at the 5 with or without Melo. In fact, I would like (love) to see Chandler getting only 30 minutes per game during the regular season to save his knees or whatever, with Amare filling in for the full 18 minutes at the center position. The rest of the time can be at the 4, or even the 3. 30 minutes for Stat would be fine too. Thomas and Camby can fill in at the 4. When Stat moves to the 5, Melo could play the 4, or even the 3. We have great flexibility and the conflict would be minimized. This probably makes little sense to a lot of you, but my idea is to ensure that Stat plays where he is efective, which I don’t think is the 4. I do not think with or without Melo, Stat has the necessary PF skill set. I do think he is gold at the 5, particularly off the bench. This is even more viable if he in fact developed low post moves, of which I am highly skeptical.

  14. Kurt

    Thomas: good article and I agree. Just one dimension in defense of the benchers:
    Even if Amar’e and Melo overlap for at least 20 minutes, there’s the competition: If Amar’e is in against subs, him and Melo can be in at the same time and it wouldn’t be as horrible defensively against starters. This would enable Melo to still play the 4 and Amar’e the 5, where he’d be quicker than any backup center. Also, no Chandler to clog lane. This would create an advantage for an additional 10 minutes or so.
    Sisterray: I couldn’t agree with you more about a Melo/Stat two man game, like a pick and roll. Another possibility is an Amar’e handoff at the elbow to Melo, where he’s most dangerous. Such a play would also solve the issue of Amar’e always slipping the pick.
    Ruruland: do you have any inside info on whether Melo has been practicing pick and roll with stat?

    STAY SAFE EVERYONE LIVING IN THE NE!!!

  15. ruruland

    Somehow I had a suspicion this issue would be framed in a way that overlooks the biggest reason Amar’e’s efficiency has suffered, and thus, why the Knicks have played poorly with he and Chandler/Melo on the floor.

    It was mostly t do with an anamolous jump shooting season, some to do with the lack of playmaking point guards.

    2009-10 Shot distribution
    Isolation 14.8% (.93 pp, rank 46)
    Post-up 19.2% (.98, 28th)
    PR roll 17.8% (1.21 18th)
    Cut 13.6 (1.43 22nd)
    Transition 6.1% (1.27 7nd)
    Spot-up 8.7% (9.5)

    2010-2011
    Isolation 32.3% (.85 102nd)
    Post-up 12% (1.08 14th)
    PR Roll 9.6% (1.13 13th)
    Cut 9.5 (1.19)
    Transition 7.7% (1.03)
    Spot-up 10.1% (.96)

    2011-2012

    Isolation 19.9% (.65)
    Post-up 10.5 (.77)
    PR ROLL 13% (1.22 12th)
    Cut 12.6% (1.29 56th)
    Transition 8.3% (1.18)
    Spot-up 15.5 % (.81)

    You can figure that out, right?

    Amare remains an elite finisher (despite the weight and back issues)
    With point guards who have a higher amount of assists on penetration and pick and roll, Amare should have a shot distribution closer to 2009-10, less like the last two years.

    He’s a very good isolation/post-up player, and most year’s he is a fantastic spot-up player (for a non 3pt shooting big man)

    Amar’e shooting issues were magnified by the poor pg play, which increased the amount of jump shoots he took relative to other shots.

  16. ruruland

    Kurt:
    Thomas: good article and I agree. Just one dimension in defense of the benchers:
    Even if Amar’e and Melo overlap for at least 20 minutes, there’s the competition: If Amar’e is in against subs, him and Melo can be in at the same time and it wouldn’t be as horrible defensively against starters. This would enable Melo to still play the 4 and Amar’e the 5, where he’d be quicker than any backup center. Also, no Chandler to clog lane. This would create an advantage for an additional 10 minutes or so.
    Sisterray: I couldn’t agree with you more about a Melo/Stat two man game, like a pick and roll. Another possibility is an Amar’e handoff at the elbow to Melo, where he’s most dangerous. Such a play would also solve the issue of Amar’e always slipping the pick.
    Ruruland: do you have any inside info on whether Melo has been practicing pick and roll with stat?

    STAY SAFE EVERYONE LIVING IN THE NE!!!

    Ofcourse they have. Go back to Seth’s blog from the first week to see his notes on it and some video of the two man game they worked on in the first week. You saw it in Toronto.

    Melo has been a very good side pick and roll player. so it should naturally be a good fit. It’s just not a play that happens with high frequency.

    But mixing in some 2 man game with Felton/Kidd/Prigioni/Shump/Smith high pnr should get Amar’es number back up — Not Suns level, but way up.

    Also, when he can pick and choose spots to isolate and pos-up, Amar’e is elite — if both Melo and Amar’e decrease their post-up/isolations, they’ll have a chance to be the best isolation post-up duo in basketball (perhaps behind Wade/Lebron or Howard/Kobe)

  17. ruruland

    Remember, last year, MDA played Melo as te point guard where he worked high pnr with Chandler — that was a highly succesfull play last season. However,it took away side pnr.

    The year prior, the two only had about a few weeks of games together (Amare injury) and were often not even running MDA’ system (because of a new pg)

    In other words, the two haven’t really had a chance to develop a two man game.

    That’s exactly what Woodson was trying to develop early. But it takes reps.

  18. ruruland

    Lin/Harden should be really good isolation as well. But there isn’t another duo who turns isolations into post-ups like Amar’e Melo does.

  19. jon abbey

    Chris Broussard ?@Chris_Broussard

    Amare Stoudemire to miss at least 1st 6 weeks of season with knee injury, according to league sources. Knicks expected to announce tomorrow

  20. knicknyk

    I knew it. I never had a good feeling about his banged knees. There is a lot more to this injury I said it from the beginning. He may end up being out for the entire season. We shall see though.

  21. danvt

    njasdjdh: Not sure exactly what you’re referencing, but it’s not just “stats based analysis” that says Amar’e and Melo can’t/don’t work well together.

    “Lowe noted that of all the players recorded by STATS, LLC’s multi-camera process for tracking every movement in an NBA game, Anthony was the most effective player in the league at driving the ball.”

    That’s statistical analysis and, in the context of the circus that this team has been, I feel, an analysis that’s tangential to why the record together is poor. Lot’s of players have similar skill sets and figure out a way to blend with each other. Meanwhile 36-30 and, most importantly 18-6 down the stretch is trending upward.

  22. Brian Cronin

    At 28, Melo isn’t getting any younger. And although he’s starving to accomplish as much as his friends Dwyane Wade and LeBron James, that appears to be a pipe dream right now.

    Bounced out in the first round eight of his nine years in the league, Anthony continues to face questions about his production, or lack thereof. Is that fair?

    Is it fair, considering that he has limited help offensively? Carrying this franchise on his back. Forced to hope that Felton re-emerges as the point guard he once was, or that Kidd can fight off Father Time, or that Tyson Chandler can display prowess in some category other than defense.

    Chances are, it’s not happening.

    Gone is Jeremy Lin. (Good riddance!) Inserted is an experienced bunch that could beat most of the teams in this league if it were healthy. But age and attrition are inevitable. So is another injury to Stoudemire.

    If it’s frustrating to me, the same can probably be said for others who are closer to the situation. And if that’s the case, one can only imagine what Anthony is thinking.

    Probably something like: “I wish I had pushed to end up in New Jersey, after all.”

    Because then he’d be in Brooklyn right now.

    Stephen Smith….wow…what a dummy.

  23. maxwell_3g

    while i agree with many point of the article, does it even address our biggest personnel conundrum?? i think chandler/ stat may be a bigger problem than melo/ stat. our center (whom i think is fantastic has one way to be of any use on offense, the pick and roll lob. this happens to be the only way Stat is even remotely efficent on offense, cosidering anything that requires him to dribble or make decisions becomes a turnover or a charge. The problem (and it is a Stat problem really) is that for him to be effective on offense requires him to play center, except such a lineup leaves a gaping whole on defense, as Stat is not exactly a rim protector (like saying Swisher is not exactly a post-season preformer.
    The only solution that I can come up with is pairing Stat witha rim protecting center who can score from outside the paint and does not depend on diving to the rim on the 2-man game. Hello Marcus Camby and your moderately effective 13 footer!!! However we work out the rotations (we, as in Woody), we should minimize the time that Stat and Chandler play together and always have Stat out there with Camby

  24. max fisher-cohen

    @maxwell

    Camby hasn’t had an effective 13 footer in a long time. The guy can’t even hit a FT anymore — 42% last season.

  25. Juany8

    maxwell_3g:
    while i agree with many point of the article, does it even address our biggest personnel conundrum??i think chandler/ stat may be a bigger problem than melo/ stat.our center (whom i think is fantastic has one way to be of any use on offense, the pick and roll lob.this happens to be the only way Stat is even remotely efficent on offense, cosidering anything that requires him to dribble or make decisions becomes a turnover or a charge.The problem (and it is a Stat problem really) is that for him to be effective on offense requires him to play center, except such a lineup leaves a gaping whole on defense, as Stat is not exactly a rim protector (like saying Swisher is not exactly a post-season preformer.
    The only solution that I can come up with is pairing Stat witha rim protecting center who can score from outside the paint and does not depend on diving to the rim on the 2-man game.Hello Marcus Camby and your moderately effective 13 footer!!!However we work out the rotations (we, as in Woody), we should minimize the time that Stat and Chandler play together and always have Stat out there with Camby

    Maxwell I’ve been saying this for a while now, Melo and Amar’e led the number 1 offense in the league after they got together their first season and had Billups instead of Chandler. Amare’s numbers suffered a bit, but the real problem was that they had a terrible defense. Chandler might have fixed the defense, but he and Amar’e can’t both roll to the rim at once, and for some reason Amar’e lost the ability to shoot last season.

  26. Juany8

    On that note, the most misleading moment of the entire season last year was Amar’e making 2 smooth pick and pop 3’s in his first game against the Celtics. Just when it looked like he might have developed 3 point range, it turns out he had a horrible shooting year

  27. maxwell_3g

    max fisher-cohen:
    @maxwell

    Camby hasn’t had an effective 13 footer in a long time. The guy can’t even hit a FT anymore — 42% last season.

    maybe so. i cant say that i actually saw camby play a lot last year. the stat catch 22 problem remains though, although maybe we dont have a solution on our roster.

Comments are closed.