Will Shuffling the Pieces Solve the Puzzle?

If you haven’t heard yet, Pistons Coach John Kuester plans to shake One of my all-time favorite Rip picsup the starting lineup beginning tonight at Toronto. The Pistons have deployed multiple starting lineups this season, so why is this newsworthy? This time, Kuester is benching Richard Hamilton, who has been the “masked man” of the franchise for the better part of a decade, in favor of Ben Gordon.

I have a special affection for Rip Hamilton. Some of you may know that I was fortunate enough to play college basketball at a small liberal arts school, and as a freshman, I was struggling to figure out how I could improve my game so that I could contribute to my team. Over a holiday break, I asked my high school coach what he thought – and he told me to watch the way Rip Hamilton used screens to create space, both for himself but also for his teammates. And watch him intently I did, for hour after hour after hour. I modeled the way I tried to play on the court – using screens, moving without the ball, etc. – after the way he did. Naturally through this process, I became a Rip fan.

ben-gordon-detroit-pistons-d5c694c30a5efa37_largeSo like many Pistons fans, this news is bitter sweet. On the one hand, it’s another door closing on the old era that we cherish so dearly. On the other hand, this appears to be the right move for the present and the future. Put simply, at this point in their respective careers, Ben Gordon is the better player and appears to be the best option for the Pistons at starting SG.

This appears to be the consensus of conventional wisdom, and in the case of Hamilton vs. Gordon, the Wins Produced numbers agree. Here are the numbers through 28 games:

2010-12-22 Automated Wins Produced

With Gordon now the starter, what can Pistons fans expect?

Keith Langlois, editor of Pistons.com, offers the following comments:

If Gordon as a starter can provide 18 or 20 points consistently and efficiently, the Pistons will be better than a 9-19 team. Maybe good enough to chip away at .500 and put themselves back in the playoff chase. But if Hamilton is true to his word and embraces his new role, then perhaps the Pistons can be significantly better than that. […]

Twenty points isn’t a magical barometer, I suppose, but both Hamilton and Gordon are primarily scorers. That’s their best asset. If they can figure a way to get each of them to score 15 a game, at least, the Pistons probably could mount a playoff push. […] With neither playing performing at close to capacity, why not reverse their roles?

Best case, both of them start scoring consistently and efficiently. But if it even jolts one of them back on course, the Pistons will be ahead of where they’ve been.

To sum up, there are three things Keith is arguing here. First, both Gordon and Hamilton are underperforming (at least that’s how I read them being “off course” at least). Second, if either one of them is “jolted back on course” by this change, the Pistons will be better moving forward than they’ve been. Third, if this change causes them both to play the way they did on October 28, 2009 on a consistent basis, the Pistons could complete the season at a better than .500 clip.

Wow! The Pistons, who have won just 9 of 28 thus far, might be able to anticipate finishing the season by winning more than 26 of their remaining 54 games simply by swapping the roles of two shooting guards. That is truly remarkable. But do any of the arguments hold up to scrutiny?

Are Gordon and Hamilton underperforming?

Rip’s performance this season hasn’t been good, and there’s no question about that. However, what may be surprising to some Pistons fans is that Rip has never been very productive relative to Wins Produced. In fact, his career season with the Pistons was 2003-2004 (at age 25) when he produced 4.7 wins with a WP48 of .081 (average is .100). Yes, in his best season with the Pistons, Rip was “below average.” Furthermore, Rip is getting old, and in spite of working tirelessly on his conditioning, Rip can’t defy age indefinitely. So while his performance is declining, it’s not clear to me whether this is the natural result of age or the result of being assigned the wrong role in the rotation.

What about the newly-signed, highly-paid Ben Gordon? His minutes, shot attempts, and points per game have all dropped since joining the Pistons. That must be a result of having two good scorers at the same position, right? Well, not necessarily, according to the numbers. Ben Gordon’s career season came in 2008-2009 with the Chicago Bulls (at age 25) where he produced 5.6 wins with a WP40 of .090 – remarkably similar to Rip’s career season, isn’t it? And while he has produced better numbers in other seasons prior to that one as well, this season is not a dramatic departure from what Gordon has offered in the past. Yes, he’s not performing optimally, but it’s not as bad as many people suggest.

So, while it’s true that both players aren’t performing optimally, it’s not clear that the roles of the perspective players are to blame.

Will the Pistons be better if either one thrives in a new role?

This one’s easy. If either player improves, and if the other’s production doesn’t fall of a cliff completely, yes, the Pistons will better. That’s obvious. If your individual players play better, of course the team is better. The question is, how much better? Langlois is clearly thinking playoffs – finishing the season 26-26 for a total of 37 wins is likely to put the team in the Playoff hunt. All due respect to Keith, I don’t think this is very likely.

As a thought experiment, let’s imagine that moving Ben Gordon into the starting lineup does propel him back to his peak performance (.090 WP48). Let’s also imagine that Ben Gordon averages 36 minutes per game at SG (ignoring for the moment how that impacts Rip’s minutes and position). With 54 games left to be played, the Pistons could expect approximately 3.6 wins from Ben Gordon. Given the relatively unproductive cast of players the Pistons employ, I struggle to see how an additional 3-4 wins matters much. Instead of winning 26-27 games as the current numbers project, the Pistons could win 29-31. To my eye, that amounts to fewer lottery balls, not a playoff berth.

A Case Study: October 28, 2009

I recently argued that the Pistons aren’t very good because the Pistons aren’t very good. Yes, I think it’s that simple. We don’t have many productive players; therefore, we don’t win much.

It is true, however, that this collection of players has played very well together at times – and one need look no farther than the first game this team every played together as a group on October 28, 2009. That night, Hamilton scored 25 points on 53% shooting, and Ben Gordon scored 22 points on 58% shooting – a remarkable shooting performance from both, to be sure. Langlois seems to suggest that this should be the case study – both for the fans and perhaps also for the coaching staff as well – that models what Dumars envisioned for this team when he put it together. If Kuester can find a way to replicate this performance, the Pistons problems will all be solved, right?

Alas, replicating that performance has been difficult. As Keith observes,

Remarkably, there has been only one other game since the two became teammates that both scored over 20 points. The other came last Feb. 21 when the Pistons beat San Antonio in overtime.

The good news is the Pistons are undefeated when Hamilton and Gordon score 20-plus. The bad news is it’s happened twice in 62 games when they’ve both been available. The worse news is they’ve had four games already this season in the 25 they’ve both played when each one finished in single digits. Last year, there was only one such game in the 37 when both played.

To put this as simply as possible – Gordon and Hamilton are not going to shoot over 50% on a regular basis, let alone on the same night. And as a result, history will keep repeating itself – the nights that both players score a lot and efficiently will continue to be rare, regardless of who’s starting and who’s coming off the bench.

How do I know this? Because of the statistics that the NBA tracks, I can view their career shooting percentages and see with absolute clarity that they’ve never shot above 50% consistently (and most SGs in the NBA don’t either, not even the best ones). So as fun as that game was to watch back on October 28, 2009 – and yes, I did watch and enjoy it – I think it’s unlikely to expect many games like that in 2010-2011.

In other words, juggling the rotation in hopes of achieving, “If we could just play like we did that one night last season, we could turn this thing around!” doesn’t seem like a very good strategy.

Same Pieces, Same Puzzle

Unfortunately for the Pistons and their fans, I don’t think changing the starting Shooting Guard will have a dramatic impact. I do agree that it’s the right move, and it seems to be the natural next step in the transition from the old guard to the new, but ultimately, not that much is changing. One unproductive SG is being replaced by a slightly less unproductive SG, and the rest of the Pistons problems – specifically, poor shooting, poor defense, and poor rebounding – remain.

Kuester can arrange the pieces however he wants – and I don’t fault him for trying – but the puzzle when completed is going to look very much the same.

About these ads

16 thoughts on “Will Shuffling the Pieces Solve the Puzzle?

  1. If WP indicates that the leading scorer of a dominant team was a below average player, even in his prime, doesn’t that suggest a flaw in the statistic?

  2. If Coach Kuester continues to properly define specific roles for his players in a way which is similar to what he did in yesterday’s game this year’s Pistons will have an opportunity to play much better basketball for the balance of the season.

    The collection of players on this year’s Pistons team is not good enough to challenge for a top 4 position in the Eastern Conference but, as they’vealready shown in brief glimpses to this point in the season there is more than enough talent on-board right now to compete for the #7 or #8 playoff position, if it is coached in a proper way.

    Paraphrasing what the great Bill Walsh said years ago:

    First rate coaching is what takes sporadic dispays of athletic prowess and turns it into a consistent level of performance.”

    What the Pistons have lacked most since the glory days of Chuck Daly & Co. and the solid run of Billups-Hamilton-Prince-Sheed-Diceman-and-Ben is a first-class coach.

      • That looks like a lot of reading to explain why someone with Rip Hamilton’s accomplishments is a below average player. I mean, that’s a pretty big claim considering we’re not talking about gut feelings here. The Pistons were pretty dominant for a while with one championship and another conference championship to show for it and their leading scorer is below average? If you’ve already studied this Occam’s razor-defying metric, perhaps you can offer some analysis? If Rip was below average at his best, perhaps some explanation of what he did so poorly and what other Pistons apparently did so well to make up for it? Or not, but I haven’t seen enough to start reading books for the explanation.

        • Yes, it is a lot of reading, and I assure, I’ve done all of it :)

          But I’m happy to offer a brief summation of why Rip is below average relative to Wins Produced.

          Here’s a spreadsheet detailing averages per 48 minutes for all 5 positions: http://bit.ly/i8sY4H

          In short, Richard Hamilton was “below average” because:

          - In spite of managing to score more points than the average SG, he doesn’t score them all that efficiently, as evidenced by his points per shot, eFG, adjusted FG% or TS% (take your pic, all available here: http://bit.ly/h7lysP)

          - Outside of points per game, Rip is remarkably unexceptional ( grabbing rebounds, dishing out assists, stealing the ball, and avoiding turnovers)

          In sum, Rip shoots the ball a lot so he scores a lot, but he doesn’t do much else other than shoot. And because it takes more than shooting to win in the NBA, Rip’s lack of production elsewhere means he’s not as beneficial to his team as his PPG (or his salary) suggests.

          Several other Pistons players during their glory years, particularly Chauncey Billups, Ben Wallace, Antonio McDyess, and Tayshaun Prince (to an extent), were exceptional relative to non-scoring statistics AND offensive efficiency.

          So to put it simply, the Pistons could afford to let Rip shoot a lot and still manage to win a lot in spite of his deficiencies elsewhere, because he was surrounded by several other players who were both 1) efficient relative to scoring and 2) productive relative to the non-scoring components of the game.

          I understand how “counter-intuitive” my argument is on this blog, but the argument is backed by a mountain of very rigorous, scientific analysis – almost all of which is freely available online. If you’re interested, feel free to check it out. If not, I understand.

          But, I’m sure you’ll understand, it isn’t possible (or desirable) for me to reproduce all of that every time I create a new post or engage a new commenter in conversation. That’s why I created the “Required Reading,” as a resource to which to direct people if interested in learning more.

  3. Jeremy,

    Unfortunately, those who worship unapologetically at “The Church of Statistical Analysis Gone Madd,” in the game of basketball … i.e. where a player’s specific contribution to the success/failure of his team is measured solely by his “production numbers” according to a certain [and generic] metric … will never be able to provide a suitable answer to the question which you’ve asked.

    In general, NBA games consist of somewhere between 160-to-240 individual possessions.

    The actual outcome of a fairly high percentage of these games is determined by what happens on the final few [e.g. 3-4] possessions of the last quarter … based upon a regressive analysis of what is actually most important in determining “a winner” [i.e. the key shot goes in the basket, or a key rebound is obtained, etc.] from “a loser” [i.e. the key shot does not go in the basket, or a key rebound is not obtained, etc.] … and, yet, what did certain “stats-worshippers” would suggest is an accurate depiction of reality is that “average” performance is what counts the most and not “how individual players actually match-up against one another on specific possessions within a specific game.”

    • Unfortunately, those who worship unapologetically at “The Church of Statistical Analysis Gone Madd,”

      khandor,

      I’m open to criticism, but this isn’t criticism. This is an absurd representation of my position. Huff, puff, and blow the strawman down all you want. If you want to actually engage the metric or my arguments in a real conversation, I’m all ears.

      • brgulker,

        There’s been neither huff, nor puff coming from me.

        The metric is flawed because it is based on flawed data and flawed perceptions of what actually determines winners from losers in the NBA game.

        [NOTE: The metric is not flawed, however, in assessing that Rebounds, in general, are grossly under-valued by those who do not fully comprehend how the NBA game actually works, based on Individual and, therefore, Collective Match-ups [e.g. Player vs Player, Position vs Position, Coach vs Coach, Unit vs Unit, Style of Play vs Style of Play, etc.].

        Here’s a simple question for you to consider.

        Was the final field goal scored by Detroit, which took the Pistons’ score from 99 to 101, in a hypotheical 101-100 Pistons’ victory:

        i. Of more importance;
        ii. Of less importance; or,
        iii. Of equal importance;

        compared to the initial field goal scored by Detroit, in the same game, which took the Pistons’ score from 0 to 2?

        Your answer will reveal a great deal about:

        A. How your mind works, from a logical perspective; and,
        B. How you mind works, from a basketball perspective;

        when it comes to distinguishing winning teams from losing teams in the NBA.

  4. brgulker,

    Thanks for the reply. I still don’t think this passes the smell test though. I hear you saying that Rip is a below average shooting guard because, although he may score more than average, he is not efficient enough and doesn’t do enough other things. Some points to consider:

    1. The Pistons teams of the mid-2000s were touted as being true teams, where everyone shared responsibility and worked together. Even in an environment with numerous scoring options, Rip was above average for points scored because he was the primary option. It was his job to score. Other people could do it too, because that’s how the pistons were, but that was his primary role. How do his shooting percentages compare to other shooting guards who are their team’s first option?
    2. Because the Pistons were such a balanced team, Rip wasn’t required to excel in numerous other statistical categories. He didn’t need to rebound a lot or block shots because he had Ben Wallace in the middle along with varying other worthy inside presences. He did show in 04-05 that he could make plays for others when he averaged 4.9 assists, but he wasn’t really needed for that either with Chauncey as the floor leader. Finally, it might not show in statistics, but I can’t remember Rip ever being less than an average defender, and let’s not forget he was a part of a (literally) game-changing defensive team. The Pistons were known for their balanced, team play, not for chasing statistics. But they still won a lot of games.
    3. How you score affects your percentages. You don’t expect the same shooting percentages from a post player as you do from a 3-point shooter. Rip’s specialty is the mid-range shot, ideally catch and release off a screen or six. This is an inherently more difficult shot than many, such as lay-ups (driving guards) and tip-ins (hustle-y big men) and probably more than the open corner three, a league favorite for shooting guards and small forwards. Though it is more difficult, that shot is very helpful for initiating a half-court offense without a dominant post presence (see point two, Ben Wallace:). This style of play is particularly effective in the post season, when it counts, so even though the percentage may be lower, do you believe the value of Hamilton’s scoring style is properly gauged byt his system?

    There are probably other points to be made, but I think that’s enough.

    I’m open to persuasion though. If you ever feel like doing a post that breaks down WP by analyzing the success of the mid-2000s Pistons (perhaps even as compared other teams, or to the failure of the current squad) I would be interested in reading it. :)

  5. Hmmm …

    Dec 19 W vs New Orleans [111-10 OT]
    Dec 22 W @ Toronto [115-93]
    Dec 26 L vs Chicago [92-95]
    Dec 27 L @ Charlotte [100-105]
    Dec 29 W vs Boston [104-92]

    It has certainly been interesting to see the Pistons collection of supposedly “below average players” beat the Hornets, Raptors and Celtics in 3 of their last 5 games with a set of “re-shuffled puzzle pieces”.

    As I’ve said, in advance of actual events, for the past 2+ years …

    When the Pistons collection of “average-to-above average” players have actually been used in one of several different but always “proper” types of ways – i.e. combining the individual strengths and weaknesses of their players in a correct way, re: Rebounding, Team Defense and Team Offense, for their 1st and 2nd units – they have been more than capable of competing effectively for a lower tier playoff position in the Eastern Conference, as Joe D. attempts to re-build the core of their franchise without ever sinking to the very bottom of the league standings.

    Unfortunately, from a Pistons’ perspective … primarily because John Kuester has not yet been able to display the ability to coach his troops in a proper way ON A CONSISTENT BASIS … the tier structure in the EC now looks like this:

    1st Tier – Miami/1, Boston/2 and Orlando/3
    2nd Tier – Chicago/4 and Atlanta/5
    3rd Tier – New York/6, Milwaukee/7 and Indiana/8
    4th Tier – Charlotte/9 [good shift to Coach Silas], Detroit, Philadelphia [Coach Collins is making gradual headway], Toronto and Washington [good trade with Orlando]
    5th Tier – New Jersey [Coach Johnson is still butting heads with certain players] and Cleveland [Coach Scott is going back to square one]

    and the Pistons are unlikely to finish with a plus .500 W-L Record and/or qualify for the playoffs once again this season.

    How Detroit will fare in the immediate future … the rest of this year and beyond … will depend on the work of John Kuester, or whoever else it is that Joe D. decides to have as the head coach for the Pistons.

    Have a Happy New Year’s Eve!

  6. Pingback: Have the Pistons found a winning rotation? | Pistons by the Numbers

  7. Pingback: Pistons roundtable: Joe Dumars’ culpability for the state of the Pistons « PistonPowered

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s