Welcome to Vizual Statistix! My name is Seth Kadish. I live in Portland, OR, where I work as a scientist. To learn more about me, visit my LinkedIn profile, and send an invitation to connect.
This blog is a product of my passion for data visualization. The data shown here are sourced from other websites, but all statistical operations on these data and the resulting graphics are original.
If you would like to use one of my graphics on your website or in a publication, please email me. I also take requests and am available for freelance work. Contact me if you have a suggestion for a graphic or need support on a project.
It’s been a while since I posted a correlogram (see link for explanation on how to read these), so I feel this is overdue. This graphic shows the correlations between basketball stats from the 2012-2013 NBA season. I’ve included data from 436 players (everyone who played in at least 10 games).
Because the players who play more minutes will usually accumulate higher stats – more rebounds, assists, steals, etc. – I generated two correlograms. The top uses all the player averages per game. For the bottom graph, I divided each player’s non-shooting percentage stats – any stats that would likely increase as the player spent more time on the court – by the average number of minutes he played each game. This normalization gives a better sense of what statistics are actually correlated based on players’ skills, rather than the time they get to play.
There are so many interesting results that I won’t even try to summarize all of them. Here are a few highlights:
There is no correlation between field goal and free throw percentage.
The only moderately strong negative correlation on a per game basis is between three point shooting and offensive rebounds.
On a per minute basis, players who dish out a lot of assists don’t get many rebounds or blocks (not surprising), but they also pick up fewer fouls. Personal fouls are most strongly correlated with offensive rebounds, likely due to posting in the paint and getting loose-ball (over-the-back) fouls.
Players who play in more games, and average more minutes per game, get fewer fouls per minute.
On a per game basis, average points scored is most strongly correlated with minutes played, and then with turnovers. On a per minute basis, points scored is most strongly correlated with minutes played, and then free throw percentage.