Thinking Outside the (Salary) Cap: Basketball, Superstars, and the Lack of Parity in the NBA

13 rings. 12 seasons. ‘Nuff said. It’s a superstar’s league!

Much of the talk surrounding the NBA’s lockout, CBA, and the future trajectory of the NBA has centered on the NBA’s relative lack of parity. To quote J.A. Adande’s recent update on the lockout on

“The NBA’s salary cap — the first in American professional sports — was instituted in the 1984-85 season. Since then only nine teams have won a championship. The NFL and NHL have crowned 14 different champions each in that span. Baseball has had 17 different teams win a championship. The NBA has been the least successful at legislating balance.”

In Henry Abbott’s TrueHoop piece on the hard cap, he comments on the value of parity going forward in the NBA and on whether the hard cap would help level the playing field:

“Television viewers like close games. The idea is that more close games make the league a more valuable property, which lead to higher TV ratings, which lead to more revenues for the entire league. There is research about the effect of parity on TV ratings, they say, to prove this. I have talked to economists about this, and they say that there is not strong evidence that parity will have a big effect on TV ratings. Some even question whether it’s possible for the league to do anything to change how wins are distributed. Wins mostly go to teams that combine players drafted incredibly high with excellent management. And nothing about a salary cap can make more really good players, or more really good GMs.”

A level playing field would presumably boost the success and popularity of struggling teams in small markets such as the Kings, Bucks, and Bobcats. But Henry’s point that wins go to teams with high draft picks and excellent management hints at a bigger point about  basketball and the NBA: The NBA is a league where a team’s overall popularity and success (read: winning) are driven by superstars. Unfortunately for some teams, there is a limited number of superstars, and therefore, success and popularity will generally be limited to the teams with true superstars.

Thinking Outside the Cap

As Adande pointed out, there is a select crop of NBA teams and players that have represented their franchises in the Finals since the introduction of the salary cap in 1984-85. Although Adande said nine, I only counted eight teams (Celtics, Lakers, Spurs, Bulls, Pistons, Rockets, Heat, Mavericks) that have hung up championship banners in those 27 seasons since the start of the salary cap. Of those eight teams, six of them won multiple titles (all but the Heat and Mavs), taking home 25 out of 27 possible Larry O’Brien trophies. You can point to draft picks, management, or coaching, but to me, the overarching trend is the success of superstars and dynasties in the NBA. Recent NBA champions such as Magic, Jordan, Hakeem, Duncan, Kobe, Shaq, Wade, Garnett, and Nowitzki were all among the top five or so players when they won championships. And these players won big and won consistently throughout their careers. With a little bit of thinking beyond the specifics of CBAs and salary caps, other reasons related to the inherent nature of the sport of basketball and to the culture of the NBA can explain this lack of parity.

Reason #1: Individual Player Opportunity for Impact (Minutes x Usage)

Reason #1 focuses on why NBA superstars can tip the scales of a match outcome more than in other sports. As you’ll see, it boils down to two reasons: 1) NBA superstars play a higher proportion of their team’s minutes relative to superstars in other sports. 2) NBA superstars have fewer teammates on the floor and therefore get more opportunities to contribute while they are on the floor (e.g. shots, play-making opportunities, etc.). Broadly speaking, when I say “usage,” I’m referring to the rate at which a given player in any sport attempts to contribute in a meaningful way–the rate at which they “use” a play. This rate exists regardless of minutes played by any player/team, but interacts with minutes played to paint the picture of total potential for individual player impact.

1) In the NBA, each team plays 3936 minutes per season (48 min/game for 82 games). Superstars will routinely play at least 85% of all possible minutes in a given game and, barring injury, they play a substantial number of games per season. Monta Ellis of the Golden State Warriors led the league at 40.3 min/game and played 3227 minutes in 80 games last season. This means on average he played in 84% of the minutes each game he played in and 82% of all minutes (including the games he was injured for). That’s a lot of minutes, and by playing such a high proportion of his team’s minutes, Monta Ellis could constantly be there for his team (except on defense, where he technically plays but does not contribute at a superstar level). The bigger the star, the more pronounced the effect. A good example: Dwight Howard propelled his team of shoddy defenders to a top-3 defense for three straight years. On offense, his inside presence forced double-teams and opened up threes for his teammates. If Dwight were limited to a third of his team’s minutes, the Magic would not be close to as formidable, thus proving the value of a superstar in the NBA by way of the potential proportion of minutes played.

By contrast, the NHL leader in Time on Ice was Duncan Keith at almost 27 minutes a game, good for 44.7% percent of the 60 minutes his team plays. In hockey, lines are subbed in and out constantly–superstars such as Alexander Ovechkin and Sidney Crosby have relatively fewer opportunities to make an impact. In the NFL, most players only play one side of the ball (offense or defense) and therefore at most they play half the game. Quarterbacks get ample opportunity, so a superstar QB can make a substantial impact on par with that of NBA superstars. (And they have–Tom Brady, Peyton Manning, Drew Brees, and Aaron Rodgers all have recent Super Bowls to their name.) But if the superstar plays another position, his team may be out of luck. In the MLB, even the best hitters have only a handful of offensive opportunities (i.e. plate appearances) per game. Regardless of talent, starting pitchers play once every five days at best, and for only a fraction of innings in the games that they do end up playing. All in all, it’s easy to see that based on minutes alone, NBA superstars have more opportunity to make an impact than superstars in the other major pro sports.

2) Now that we’ve covered minutes, the next item to discuss is the concept of usage. Broadly speaking across sports, usage is the percentage of overall plays that a player uses while playing. describes usage as “an estimate of the percentage of team plays used by a player while he was on the floor.” Note that usage is a statistic/concept describing an on-court percentage independent of minutes played. Going back to intersport comparisons, basketball superstars not only play many minutes, but share the floor with only four teammates at a time. This means that an NBA superstar can at least touch the ball every time down the floor, and he can frequently “use” the play as the designated star on a court with only four other teammates. In addition, NBA superstars make offensive and defensive impacts, allowing for another route for an individual force to dominate a game. Couple the high usage rate and offensive+defensive impact with the high number of minutes and you get the highest individual opportunity for impact (by far) among the four major professional sports.

Hockey superstars actually have some of the same usage potential as NBA superstars, but this is attenuated by their lower level of minutes played. NFL quarterbacks and running backs also have the opportunities to take over games, but as discussed earlier, they can’t play defense and their success is often predicated on offensive lines (which are rarely made elite by a single superstar). Baseball might not have a salary cap at all, but parity will probably always exist because superstars, both offensive and defensive, cannot carry a team alone. By nature of the rules of baseball batting orders, offensive superstars (i.e Barry Bonds) have their rate of offensive opportunity maxed out at 1/9th of all team offensive opportunities (i.e. plate appearances).  The most important defensive superstars, pitchers, make substantial impacts in the games they do play, but their value is also attenuated by the fact that they only pitch once every five days.

All in all, NBA superstars have more power for consistent domination because of their ample opportunity to make an imprint on the game. This is a function of both their ability to play many minutes and their inherently high usage percentage (which is due to sharing the floor with relatively few teammates). NBA superstars do not depend on pitchers or offensive lines and they barely have to worry about what their teammates do while resting because they are virtually always on the floor. In addition, NBA superstars, more so than stars in other sports, are frequently superstars on both offense and defense, giving even further opportunity for impact. In the end, it means that although basketball is a team sport, in the NBA, a superstar has a greater impact on a game; in a league with a limited number of superstars, the superstars will reign supreme year in and year out.

Reason #2: Predictability

As Dave Berri noted on Wages of Wins, NBA player statistics are more consistent from year to year than are the player statistics in the other major professional sports. To quote Berri from one of his original posts on the Wages of Wins blog:

“In football payroll explains less than 5% of wins. But in football we also see very little consistency in player performance…A similar problem, albeit to a lesser extent, exists in baseball. In basketball, though, players are much more consistent across time. The correlation between a player’s per-minute Win Score from this season to last season is 0.84. As we detail in The Wages of Wins, the consistency we observe in basketball exceeds what we observe in either baseball or football….In sum, decision-makers have a greater ability to predict the future in the NBA…”

This is another way of saying there is less year to year random variation in NBA player stats than there is other major sports. And when there is less randomness, you can expect the cream to rise to the top; that is to say, in the NBA, because there is less randomness, the best players/teams are less likely to be derailed by randomness en route to a championship than players/teams in other sports. Why there is less random variation in the NBA is a different question, although I have some theories:

1) In comparing the NBA to the NFL, the NBA is more predictable (and therefore is less susceptible to random chance) for a host of reasons. For starters, the NBA has 30 teams while the NFL has 32. The NFL has always had more teams since 1984-85, and the bigger the field, the less predictable the outcome (you have a better chance of predicting Heads vs. Tails than a six-sided die). The second reason is because of the NBA’s significantly longer regular season schedule (82 games versus 16 games). The longer the season, the better the chance the best teams rise to the top, claim home-court advantage, and exploit this advantage en route to a championship. Even in the postseason, the NBA deals with greater sample sizes–NBA teams play best-of-seven series while the NFL teams advance after a single victory. Looking back at the 2008 Super Bowl, if the Patriots and the Giants played a best of seven series, who would you bet on, even with the knowledge that the Giants stole that game at the time?

2) In comparing the NBA to the MLB, it is more difficult to come up with reasons why the MLB is less predictable. The MLB has a 162 game regular season, so we have to rule out regular season sample-size as an issue. I would argue that on a theoretical level, hitting pitches of varying spin and speed with a thin wooden bat into a field of nine defenders produces more random results than does shooting on a stationary basket with a standard ball. Beyond that, the relative inconsistency of baseball player-statistics is beyond me, but regardless, it contributes to less predictable team performance. Another reason that the MLB sees more post-season parity is due to a point mentioned earlier–starting pitchers can only pitch once every five days. When a team does not field the same lineup each game, it becomes less predictabe, more susceptible to randomenss, and as a consequence, we see more parity. Four wild-card teams have won in the World Series (Marlins in ’97 and ’03, Angels in ’02, Red Sox in ’04). Could you imagine an 8-seed winning the NBA championship four times in an eight-year span? Could you even imagine it happening once? Didn’t think so.

3) In comparing the NBA to the NHL, I can’t say I know much about hockey or the NHL, but every playoffs all I hear about are the “hot” goalies. The concept of “hotness” is not unrelated to “streakiness,” be it random or not. By this logic, a hot goalie can upset a top seed by virtue of being hot for a few games. Because goalies can heat up at any time, hot goalies lend themselves to upsets and more parity in the playoffs than in basketball.


The owners have legitimate concerns with the top-heavy dynamics of NBA basketball. While the jury is still out as to whether more parity would increase the league’s popularity, it is easy to understand why small-market teams in the dumps would like a fighting chance at big dogs like the Lakers and Celtics. For all of their concerns, however, it is worth noting that the NBA has been like this for years. In the NBA, superstars have more opportunity to will their team to victory via the high volume of minutes and plays they use, and this fact  coupled with the NBA’s relative player-predictability translates to a lack of parity in the NBA. Because the limited number of NBA superstars can play at a high level for a long time, dynasties form and dominate the league. If the owners and the NBA want more parity, the sport of basketball–not the CBA–will have to undergo a revision. And I must say, even as a fan of the lowly Wizards, that massive changes to the sports rules and mechanics would hinder the league’s popularity rather than help it.

Comment on the article or e-mail us at AGRbasketball (at) gmail (dot) com. Don’t forget to follow @AGRbasketball on Twitter and to like us on Facebook.

This entry was posted in NBA analysis. Bookmark the permalink.

4 Responses to Thinking Outside the (Salary) Cap: Basketball, Superstars, and the Lack of Parity in the NBA

  1. Pingback: Court Vision: The latest from around the league | DAILY NEWS

  2. nolop says:

    extremely logical and well argued article. i was very impressed, and the comparison between these other sports has opened my eyes to the differences in the nature of average player vs superstar dynamics. not to mention the fact that I now am more convinced that these debates over salaries in the NBA will do little to nothing for the struggling, smaller franchise teams…

  3. Pingback: Should tired stars rest or play for their fans? An alternative, team-focused solution | Alone In The Green Room

  4. An insightful post right there mate . Thank you for it !

Leave a Reply

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

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

Google+ photo

You are commenting using your Google+ 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 )

Connecting to %s