Age, Success, and the NBA: The Young Produce Wins but the Old Win Games. Wait, What?

The veteran teams, like the Spurs and Mavs, just won’t stop ’till they drop

One of the interesting surprises of the 2010-11 NBA season has been the prominence of the aging (and historically successful) teams. In the Western Conference, the Spurs, Mavericks, and Lakers hold the top three spots; in the East, the Celtics, Heat, and Magic are three of the top teams. Orlando and LA have struggled of late, but I don’t think anyone is counting them out, yet. Drawing from an old blog post from hoopism.com, we can see that, by the numbers, all of these teams are actually old–they make up six of the seven oldest teams in the NBA. When properly adjusting for the playing time of the differently-aged individual players, the Mavs emerge as the elderly statesmen of the league, boasting a team that is on average 31.33 years-old; the Magic are the youngest of the aforementioned teams, but are still the seventh oldest among all teams, at 28.32 years-old.

While no one will deny that these teams benefit from some of the league’s premier superstars, it is certainly curious that they are showing so few signs of regression. Yes, some of the teams employ the likes of younger, productive stars such as LeBron, Rondo, and Bosh, but they also employ older stars such as Kobe, Garnett, Duncan, Nowitzki, Kidd, Ginobili, Gasol, and Pierce, among others. The trend of the successful older teams is especially interesting when considering that Dave Berri’s research (which can be read at his blog, The Wages of Wins Journal) indicates that NBA players typically start declining at 27 or 28 years old. Berri measures performance using his somewhat controversial metric Wins Produced, a statistic that is predicated on its strong correlation with winning.* If each of the teams highlighted in the first paragraph are older than 28 on average, and thus full of players past their primes (based on a composite-statistic derived from that which correlates most strongly with winning), then how are these same teams sitting atop the league standings? Armed with Berri’s research on peak performance, we observe the dissonant reality of the 2010-2011 NBA: although the statistic most strongly correlated with winning concludes that, on average, a player produces the most wins from 24-28, the older NBA teams (with average ages of 28+) are producing the most wins as teams. What gives?

I cannot give a definitive answer to this question–for that I will e-mail Berri himself–but for now I can certainly theorize. For one, it is my understanding that when Berri looks at player peaks, he is not comparing the performance of the average 24 year-old to the average 30 year-old; rather, he looks at the trends of each individual player’s career and sees that an individual player’s statistical prime occurs from 24-28. Why is this difference in measurement important? Well, if Player X is 28, 29, or 30, he is only inferior to his 25 year-old self, not to the average 25 year-old. In fact, given the fact that Player X remained in the league until age 30 indicates that he is likely above average, even if he is still worse than he was from ages 24-28. What does all this mean? If a team is full of 31 year-olds, like the Mavs, then they still may win, even if the older players are past their primes. A player can be over the hill at age 28, but there’s a good chance he’s still better than than your average third-year 24-year old. The moral of the story for fans, coaches, and GMs is this: more important to winning than either energetic youth or veteran savvy is sheer production. Sometimes it comes from a youthful Magic Johnson or Dwyane Wade, and at other times from an aging Jordan or Kobe. Winners win. Mind-blowing, right?

Have a different answer to this conundrum? Which veterans or teams are surprising you this season? Concerns? Comment on the article or e-mail us at agrbasketball@yahoo.com. Don’t forget to follow AGRbasketball on Twitter.

*=Berri essentially regresses the most accurate predictor of winning (efficiency differential) onto individual box-score statistics, assigns values to the individual statistics depending on their importance to offensive and defensive efficiency, and then rates players using their box-score statistics/assigned values. (In addition, Berri controls for position and team influences on individual statistics.) The end results are the statistics Wins Produced and Wins Produced per 48 minutes. For a more complete and accurate description of Wins Produced, read this. I also encourage you to read his books (Stumbling on Wins and The Wages of Wins) and his blog.

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.