Roster Balance and the Outlier Atlanta Hawks

We’re proud to present this guest post from Alex Lee, a native of Canandaigua, New York. His passion, aside from basketball, is ultimate frisbee and he currently writes at the blog https://upcall.wordpress.com/. He can be followed on twitter at @superfluoushype.

Too often basketball analysts rely on anachronistic clichés; one of the most cited ones is “you need a superstar to win a championship.” Like clockwork, every time the playoffs roll around, teams that don’t have “someone who can take over” are always left off the list of contenders.

This looked like it could be the year where the platitude would be put to rest. There was the 60-win Atlanta Hawks—a team with a beautiful offense where unselfish passes whipped around the court, a team that had four players selected to the All-Star team, a team so balanced that when they tore off their historic 17-0 January, all five starters were named the Conference Player(s) of the Month. They had been anointed the “Spurs of the East”. Atlanta had captured the hearts of many basketball purists.

Of course, the Golden State Warriors were the championship favorites heading into the playoffs, but if you had picked Atlanta to win, no one would have laughed. Vegas had them as the fourth highest, with 12-1 odds, behind Golden State, Cleveland and San Antonio. When coach Mike Budenholzer received his Coach of the Year Award, the Hawks were up 1-0 on Brooklyn and everything looked well. However, a pesky Brooklyn team would push the Hawks to sixgames  and in what was likely the best series of the Eastern Conference playoffs, the Wizards were a Horford box-out and a millisecond-quicker Paul Pierce release away from winning the series.

But Atlanta had survived all the scares and only Cleveland stood between them and a Finals visit. This was it: the clashing of polar opposite team constructions. There was a stark juxtaposition between the Hawks’ balanced roster and LeBron and the injury-ravaged Cavaliers team. How would it turn out?

Four games later and we’re waiting a week for the Cleveland vs. Golden State finals.

History caught up with the Hawks. Sorry Atlanta.

It is possible to debate for days what exactly quantifies a “superstar,” but this is not the point of this piece. The point of this piece is to look at the proportional part of a team’s success each player produces. The premise here is that if you have a good playoff team, a player who is responsible for a large portion of your wins, then that player is somewhere near or in that “star” player category. There are many ways to measure a player’s contribution, but I chose to look at total Win Shares—where the proportional share of each of the top-five players is:

This is not a perfect metric by any means but it should give us a fair idea of how “important” each of the top-five players is, how much they contributed with respect to each other. I also chose to look at the top-five players as opposed to whole rosters because in the playoffs, the rotations get much tighter and weaker reserve players don’t see the court as much. Note, these are not the starters; it is the players who ranked in the top-five on each roster for regular season Win Shares.

If we just look at the breakdown of the average championship team as compared to the Atlanta Hawks, we see the differences:

The Hawks are an almost perfectly balanced quintet, while the average championship team has had their largest contributor at 31.3 percent and second largest at 23.2 percent. The Hawks’ largest contributor was Al Horford who accounted for 22.2 percent of the starterss win shares. The standard deviation among those shares was 3.43 percentage points, leaving Horford 2.65 standard deviations away from the average—certainly something we might consider an extreme value, if not an outlier. In fact, if we look at the highest contributors over the past 38 championship teams, we see that the Hawks would have set an all-time low:

Only four championship teams have had their highest contributor at even less than a 25 percent share of their total Win Shares: 2014 Spurs, 1989 & 1990 Pistons and the 1981 Celtics.  Assuming future Hall of Fame inductions for Tim Duncan, Manu Ginobili and either Tony Parker or Kawhi Leonard, all four of those teams had at least three hall of famers on their team. In all likelihood, this Hawks roster will have had no hall of famers.

It is fitting that we labeled the Hawks as the Spurs of the East when we see how closely they resemble the 2014 Spurs distribution:

Of course, we already know that Gregg Popovich and company are always the exception, not the rule. A slight asterisk probably belongs on the Win Share proportions for the 2014 Spurs because their highest Win Share contributor was Kawhi Leonard who was injured and only played 64 games. If we extrapolate to him playing 10 more games, his contribution goes up to 25.9 percent.

Some fun asides that we can see by looking at the chart of historical championship teams: the second and third highest proportional contributions belong to LeBron and ironically, the highest proportional contribution for all time belongs to Tim Duncan on the 2003 Spurs. Eleven years later, and Timmy’s still going. Both of those men are robots.

Having a player who contributes a large proportion of win shares is not going to lead your team to a championship. That’s not the conclusion reached here. The point is that it’s really, really hard to win without one of those players. It would’ve been a truly once in a lifetime thing to watch such a balanced team with no hall of famers take home the championship. Unless you’re sharing the ball between multiple hall of famers, there’s nothing to suggest that you can be a championship team without an “alpha.” Empirically, we grasp the idea that it is easier to slow down five very good players with detailed scouting and playoff level effort versus having a great player(s) who is so good that you can’t slow them down or you have to cater their entire defense to stopping them.

Finally, if you’re wondering where the Cavs and Warriors fall:

With the Warriors we see how much Curry’s MVP season has contributed to the team as he contributed a staggering 35 percent of top-five win shares. The Cavaliers rating is a little inaccurate because Irving and James tied for most win shares on the team and James’s win share contribution is a little lower than his usual due to his two-week hiatus.