[This article was first published on More or Less Numbers, and kindly contributed to R-bloggers]. (You can report issue about the content on this page here)
Want to share your content on R-bloggers? click here if you have a blog, or here if you don't.

We are in the midst of what many are calling a “golden age” of the NBA.  Being in the midst of a time where attention to the sport has seemingly increased is difficult to quantify.  For most people who have had an interest in the NBA over a long period of time, the current state of the game just “feels” like a time unlike recent years.  Awareness of the calibre of game we are witnessing is important to more fully appreciating the games and the players we get to see perform.

In the last couple of years we have seen two giants in the game emerge as contenders (Durant/Lebron) that reminds many of the Bird/Magic years.  Friendship coupled with competition in the kind of way that keeps you glued to the screen when they play.  The most valuable player (MVP) distinction, I would argue is a reasonable way to see where the game is at in terms of the quality of play in the league.  The site basketball-reference provides an enormous amount of data on the sport and is a great place to begin looking at MVP as a metric for determining the “era” of current play in the NBA.

Below is a heatmap showing different statistics gotten from the website for players that were awarded the MVP in different years.  The colors show a distribution of how the players ranked compared to eachother based on these yearly stats (Red>Blue).  On the left side of the heat map is a dendrogram showing how players could be grouped based on these stats.

The stats are total games (Games), field goal % per game (FGoalPercen), free throw % (FTPercen), assists per game (Assists), rebounds per game (Rebounds), minutes per game (Minutes), average points per game (Pts), and player age (Age).  Next we take this same dendrogram and divide players into clusters using a method (kmeans) based on the above statistics.  The red lines outline the different clusters we get when creating 5 of them.  Again these groups are based on the similarity in these stats between players.

What results is a set of data where we can see how MVPs could be grouped based on the stats in the heatmap above.  Kevin Durant hasn’t been awarded the MVP yet, but let’s just assume he does, and his current stats don’t change at all after these 81 regular season games (this is all on a per game basis).

Clearly, cluster/group 2 or what I will call the “Golden Era Group” is the largest.  Even though some players arguable shouldn’t be in this group, it’s mostly comprised of players that reflective NBA watchers can agree were apart of what many have called “golden age(s)” in the NBA.  Also interesting to note are the Bill Russell and Wilt Chamberlain clusters.  In the case of Wilt Chamberlain his rebound and shooting numbers were much higher than his peers, whereas Bill Russell is placed into his own group because of his free-throw % being in the 50%-60% range…or much lower than his MVP peers.

Here are other players in the “Golden Era Group” with their Points per game against the year.  Notice how comparable Durant is to other giants in the “Golden Era”, and how amazing Jordan was compared to his MVP peers.

In general, we can see that recent years’ MVP awards are grouped with Bird, Magic, and Jordan.  As a proxy for measuring each players’ performance in the league, measuring the performance of MVPs seems to indicate that the current level of play of the best in the NBA could be associated with these by-gone eras of greatness.  In many ways knowing that the current level of play is comparable is intuitive without looking at the numbers, just by watching the game.  In support of feeling like it’s a “golden age” of the NBA there are numbers to support it.

Time for the playoffs…..

To leave a comment for the author, please follow the link and comment on their blog: More or Less Numbers.

R-bloggers.com offers daily e-mail updates about R news and tutorials about learning R and many other topics. Click here if you're looking to post or find an R/data-science job.
Want to share your content on R-bloggers? click here if you have a blog, or here if you don't.

# Never miss an update! Subscribe to R-bloggers to receive e-mails with the latest R posts.(You will not see this message again.)

Click here to close (This popup will not appear again)