Sunday, May 11, 2014

The Misunderstood Success of the Detroit Pistons

For many basketball purists, the dominant Detroit Pistons teams from 2001-2006 served as a metaphorical hat tip to the halcyon days or yore, when it was allegedly teams, not superstars, who won dominated the league by sharing the load, prioritizing “we over me” and generally playing the “right way.” When the 2004 Pistons bested the heavily favored and star-studded Lakers, it served as a coronation for the grouchy pundits who had long lamented the (alleged) new shift of emphasizing individual superstars over teams. For a generation of hoops fan, these Pistons were living proof that you didn’t need a superstar to win it all, let alone a “Big 3,” so long as you had a strong “team.” 

All of this philosophical pontificating is well and good and serves as excellent fodder for Thanksgiving debates with your uncles. Unfortunately, for those old-school hermits who are adamant that you can (and should?) win without a superstar and that good teams beat great individuals, the 2000-2006 Pistons teams happen to provide excellent support…for their opposition.

You see, dear reader, the Pistons featured one of the best players in the league. In order to demonstrate this point, I built off of the excellent research produced by Mr. Arturo Galletti. A few years back, Arturo crunched the numbers and developed a chart delineating the average distribution of wins among a given team. I opted to compare this average distribution of wins to that of the Detroit Pistons to determine whether the Pistons really were successful because of their depth. If the Pistons really were a starless team who won because they featured 5 (or more) good players, one would assume their breakdown of wins to be somewhat equitable among the starters (~16% each) or at least for the contributions to not skew heavily towards one player. The data, on the other hand, suggests something completely different:

As you can see, the Pistons relied heavily one afroed man, who contributed an absurd 36% percent of their wins: Mr. Ben Wallace.

So what have we learned from this exercise? Not only that the Pistons were actually more top heavy than than those frequently cited 2004 Lakers, who produced 62% of their 51.55 wins from their top three slots (Shaq, Kobe, Payton), but that the Pistons did indeed feature a superstar; his name was Ben Wallace and he had excellent hair and a proclivity for airballing free throws. Don’t believe me that Ben Wallace was not only the Pistons best player, but also among the handful of top players in the league during that time? Check out where Wallace ranked in the NBA in terms of Wins Produced each year from 2001-2006: 1, 1, 1, 2, 6 and 2.

In reviewing the success of the Pistons, it’s important to remember them as an excellent team who owes much of their success to team defense, a facet of basketball that is more attributable to the efforts of the entire team than any individual (1). However, the prevailing narrative of the Pistons as the quintessential example of how “teams beat stars” is downright inaccurate. The story of the Pistons isn’t one of ultimate teamwork or the whole being greater than the sum of its parts. Rather, it’s about a team dominating its opposition by relying on great team defense as well as the otherworldly production of Ben Wallace and the occasional brilliance/consistent contribution of Chauncey Billups and Tayshaun Prince.


(1) For the record, Ben Wallace was their best individual defensive player, too...'

Thanks to Xquach for pointing out that while Ben Wallace was a superstar and the Pistons were more top-heavy than average, they were not an exceptionally top-heavy roster.

Sunday, May 4, 2014

Beyond the Big 3: the Spurs Real Secret to Success

Since the Spurs began their incredible run of success under Coach Gregg Popovich, fans and pundits alike have sought to define the root of San Antonio’s unmatched consistency within the paradigm of a familiar basketball narrative. When the Spurs won their first champion ship in 1999, experts attributed their accomplishments to the popular “Twin Tower” model, which featured the historic duo of David Robinson and Tim Duncan. When the Spurs won again in 2003 with an older (but extremely productive) version of Robinson, the narrative shifted to emphasize the individual brilliance of Tim Duncan, while still highlighting the emergence of Tony Parker and Manu Ginobili as key contributors. When the Spurs won again in 2005 and 2007, the narrative shifted once more, this time attributing San Antonio’s success to the all too familiar “Big 3” model (Duncan, Ginobili and Parker).

All of this is well and good and makes life easy for lazy sportswriters and prognosticators who would rather recycle a recognizable storyline than uncover a new one. Unfortunately however, any explanation that attributes the Spurs success to simply having the best player in the league or the best "Big 3" is incomplete at best and more likely downright inaccurate. In truth, the Spurs success is owed to a number of factors, two of which are criminally underemphasized, likely because they lack the sex appeal we all covet.

So what are these two hidden factors that enable the Spurs to consistently outclass the competition? Simple: the Spurs don’t allocate any minutes to bad players and consistently surround their stars with stronger supporting casts than those of their opponents’.

Since I’ve already written about the way in which the Spurs avoid giving away wins via playing unproductive players, this post will focus on the Spurs consistently excellent supporting casts.

Is it really the Big 3 or the Big 12?

In order to demonstrate that the Spurs differentiator for success is more attributable to their remarkable depth than their top players simply being superior to those of other teams, I sought to show that the Spurs didn’t typically boast the best “Big 3” in the league, even during their championship years.

As a proxy for demonstrating this theory, I investigated the Spurs three most recent championship rosters (2003, 2005 and 2007) and compared those rosters to the rest of the league. Specifically, I calculated the aggregate average per-minute production (1) of each team’s top 3 win producers, the goal of which was to objectively isolate the teams in the league who boasted at least three highly productive players. I then investigated the production these teams received from their "other" players (e.g. non-top 3 players) in order to determine if those teams couldn’t match the Spurs success on account of employing lesser top talent or an inferior supporting cast. The results show that the Spurs model for success is much more nuanced than simply attaining better top talent than their competitors (2):


According to the data, San Antonio has indeed featured one of the league’s better “Big 3’s” in each of its championship years. In fact, in 2007, San Antonio’s “Big 3” of Duncan, Ginobili and Barry (3) boasted the highest aggregate production rate in the league. That said, in both 2003 and 2005, numerous teams employed more productive “Big 3’s” than San Antonio, both in terms of per minute effectiveness and total wins produced.

However, it should also be noted that, compared to its competition, San Antonio has received more contributions from the rest of its roster relative to other teams who boast three strong performers. For instance, in 2005, San Antonio's "Non-Big 3" players produced nearly 12 more wins than the average of their counterparts on the other teams investigated. In fact, San Antonio’s “Non-Big 3” players contributed more wins in 2005 and 2007 than any of the other teams reviewed.

So what has this little exercise taught us about the Spurs model for success?

In the world of sports punditry, there is no room for humble agnosticism. Each victory or defeat is traced to precise variables, some of which may be as unquantifiable and nebulous as “intangibles,” “team chemistry” or simply “wanting it more.” In the case of the San Antonio Spurs, however, a more nuanced explanation for success is required, and no explanation is complete so long as it overlooks the significant contributions of San Antonio’s “other” players.


1. In this case, average production is measured via Wins Produced per 48 minutes WP48 (WP48).

2. I’ll be the first to acknowledge that is hardly a foolproof methodology: for example, it doesn’t account for instances when a top player may not have produced a top three win total for his team due to injuries. Moreover, defining "top talent" by looking at top 3 players (as opposed to top 1 or 2 players, for instance) is pretty arbitrary, and it should be noted that in each of the Spurs championship years, Duncan was among the best players in the league. Caveats aside, this exercise provides a decent proxy for determining if San Antonio comparative advantage is on account of its top tier talent or its depth.

3. If you substitute Tony Parker (the assumed 3rd member of San Antonio’s “Big 3”) the aggregate WP48 drops to .238 although the absolute wins produced stays roughly the same as Parker played many more minutes (albeit less effectively) than Barry in 2007.