Can advanced basketball statistics really bridge 50 years of NBA history?

David Friedman Columnist

By David Friedman, David Friedman is a Roar Expert

 , ,

4 Have your say

    Say a person can do 20 pushups in a minute. Does that mean he can do 60 pushups in three minutes? Possibly. Does that mean he can do 2000 pushups in 100 minutes? Almost certainly not.

    Russell Westbrook is performing at a very high level this season, averaging 27.5 points per game, 8.6 assists per game and 7.2 rebounds per game. He could possibly lead the league in scoring and steals while also ranking in the top five in assists.

    ESPN’s Tom Haberstroh recently compared Westbrook’s numbers with the numbers that Oscar Robertson posted in 1961-62 when Robertson became the first (and still only) player to average a triple double for an entire season (30.8 ppg, 12.5 rpg, 11.4 apg).

    Haberstroh argued that if one adjusts for pace of play (much faster in Robertson’s era than today) and minutes played (Robertson averaged well over 40 minutes per game, while Westbrook averages about 34 minutes per game) then Westbrook’s averages would be 46.9 ppg, 14.6 apg and 12.2 rpg.

    The problem with this kind of analysis is that there are many more factors involved when comparing the 1962 NBA with the 2015 NBA than just adjusting for pace and minutes played, and it’s difficult to say how those factors would affect Robertson or Westbrook if those players were transported to a different era.

    Playing over 40 minutes per game requires more stamina – and a greater ability to avoid foul trouble – than playing 34 minutes per game. It is far from certain that a player could keep up his 34-minute pace for an extra 10 minutes per game.

    Some would argue that the 1960s NBA had less talent and featured less sophisticated defences than today’s NBA. Maybe that’s true – but maybe today’s NBA is watered down; after all, making the cut in a league with nine teams is more difficult than making the cut in a league with 30 teams and expanded rosters.

    It is interesting to speculate about what Westbrook might have done in 1962 or what a young Robertson might have been capable of doing today, but the confidence with which some writers make assertions about such matters is off putting.

    The reality is that ‘advanced basketball statistics’ are not quite as advanced as their proponents suggest. Sabermetrics has changed the way that baseball is analysed but there is a major difference between baseball and basketball. A baseball game consists of a series of actions that can be separated and measured: pitcher throws the ball, batter hits the ball, fielder catches the ball and so forth; a basketball game consists of 10 players simultaneously doing a variety of things, many of which cannot be easily quantified.

    What is the value of a player drawing extra defensive attention compared to the value of making an accurate swing pass compared to the value of alertly cutting to the hoop, catching the ball without fumbling it, and then finishing at the rim? No formula can precisely determine this.

    Not only is there no way of saying for sure how Robertson and Westbrook would have fared in different eras, there is also no way of saying for sure just by crunching numbers which player actually had more of an impact on his team’s results.

    Several years ago, sabermetrician Phil Birnbaum wrote about a detailed study that showed that the basic counting statistic minutes played per game applied to each player on a roster predicted a team’s success from one season to the next more accurately than so-called advanced basketball statistics such as player efficiency rating and wins produced (alternate win score was more accurate than minutes played by a very slight amount).

    Minutes played reflects the coach’s opinion about which players contribute the most to winning; those who contribute the most play the most, while those who contribute the least play the least.

    Regarding the Robertson-Westbrook comparison, it is fair to say that – all things being equal – Westbrook’s 2015 production is equivalent to certain numbers in 1962. However, it is not fair to conclude that this proves that Westbrook is having a better season than Robertson did, any more than it would be fair to conclude that someone who can do 20 pushups in a minute can do 2000 pushups in 100 minutes.

    David Friedman
    David Friedman

    David Friedman has covered the NBA for more than a decade, and in doing so, has interviewed nearly two dozen members of the NBA's 50 Greatest Players List. You can find his work at 20SecondTimeout.

    Oldest | Newest | Most Recent

    The Crowd Says (4)

    • April 9th 2015 @ 2:48am
      Jrod said | April 9th 2015 @ 2:48am | ! Report

      Good article

    • Roar Guru

      April 9th 2015 @ 9:38am
      Justin Ware said | April 9th 2015 @ 9:38am | ! Report

      Good article, but I think people need to cool it with the Robertson westbrook comparisons. It’s kind of insulting to the big O.

      • April 9th 2015 @ 11:45am
        express34texas said | April 9th 2015 @ 11:45am | ! Report

        Why is that? Oscar only won more than 48 games once before joining Kareem, and he was playing with another HOFer for most of those years plus other good players. Westbrook could’ve and probably should’ve been MVP this year if he didn’t miss 15 games. Oscar played more mpg like most other good players in his era. But, Westbrook will likely end up playing more years and games, and has a chance at playing more minutes if he remains as healthy as he has been for most of his career. If Haberstroh doesn’t realize his pace calculator isn’t to be considered absolute, then he needs to wake up. However, his premise of comparing each is still relevant. There’s no absolutely no chance Oscar could approach his averages if he played today, while Westbrook’s averages would all certainly increase. How much? No way to tell for sure.

    • October 16th 2015 @ 5:37pm
      pr holguin said | October 16th 2015 @ 5:37pm | ! Report

      What about Larry Bird; is there absolutely no chance that larry could approach his averages if he played today, while russ Westbrook would all certainly increase?

    Have Your Say

    If not logged in, please enter your name and email before submitting your comment. Please review our comments policy before posting on the Roar.

    , ,