The Sloan Sports Analytics conference was held at MIT last weekend. The event is, in part, a place to discuss new ways to look at statistics and it always leads to an explosion in discussion of the use of advanced analytics in the assessment of basketball.
I preface this post with that because Stats.com has an interesting ranking of passers in the NBA (via Zach Lowe), by the field goal percentage of the players who caught that guy's passes. Here's the breakdown.
- Paul Pierce leads the NBA, having thrown 705 passes, leading to 241 shots being taken and 137 made field goals (but only 110 of those classified as assists) for 56.8% shooting.
- Pierce is in the top 10 of top combinations by field goal percentage. Pierce passes to Ray Allen amount to 65% shooting. Brandon Bass is shooting 61.1% off Pierce passes and Rajon Rondo is shooting 60% off Pierce's passes.
Interesting stuff. Even more interesting is the top 10 passers by FG%:
- Paul Pierce
- Tiago Splitter
- Derrick Rose
- Tyreke Evans
- LeBron James
- Stephen Jackson
- Danny Green
- James Johnson
- Kyle Lowrie
- DeJuan Blair
Rondo isn't on the list. Neither is Chris Paul. Or Steve Nash. Or Tony Parker. Or any "elite" point guard besides Derrick Rose. Tiago Splitter's passes are leading to a higher percentage of shots made then theirs. But of the 475 passes he's thrown, only 31 were for assists. Kyle Lowry has passed the ball 1466 passes. 461 led to shots. 174 were assists.
What does all this mean?
It means we're getting a little out of control with the advanced analytics. You can watch a game and know Paul Pierce is a good passer who will find open teammtes in good spots. The conclusion drawn in the paper is that Pierce has "shown a remarkable ability to create quality shots for his teammates this year," which your eyes will tell you makes some sense. But can the same be said for Tiago Splitter and DeJuan Blair? Is James Johnson REALLY one of the 10 best players in the NBA at creating quality shots for teammates?
I see value in advanced stats. Breaking the game down a little further is good. But things like this, they go too far… even if a Celtic is at the top of the list. This confuses the issue too much to be useful. You'll see more of this stuff flying around the internet as the Sloan conference is still fresh and the hardcore advanced stats guys push its limits. But certain things should be taken with a grain of salt.