Nov 9, 2014; Phoenix, AZ, USA; Golden State Warriors guard Stephen Curry (30) makes a behind a back pass to Golden State Warriors center Andrew Bogut (12) against the Phoenix Suns during the second half at US Airways Center. The Suns won 107-95. Mandatory Credit: Joe Camporeale-USA TODAY Sports
If you’ve read anything at Nylon Calculus over the past few weeks, you know we’ve been immersed in the new publicly released data from SportVU. Things like rebounding logs and shot logs allow us to answer some questions at a much more granular level than has ever been possible before. However, there were things in the data that was released last year that were never fully explored.
With assist opportunities[1. All shots where an assist would have been recorded, regardless of whether the shot was made or missed, and whether or not the shooter was fouled.] finally being tracked, last season Tim Sartori did some work for Hickory-High, looking at field goal percentage on potentially assisted shots. With assist opportunities and actual assists we can see what percentage of assist opportunities are converted into made baskets. Sartori characterized this as field goal percentage on assist opportunities, but I like the verbiage “assist conversion percentage” because it implies that responsibility for making the basket is not solely with the shooter. This percentage is hinting at both the quality of scoring opportunities a player is passing into and the ability of his teammates to his those shots.
The graph below shows all players who are averaging at least five assist opportunities per game this season and you can see the implication—not all assist opportunities are created equal.
D.J. Augustin is averaging slightly more assist opportunities per 36 minutes than Stephen Curry, but those assist opportunities are being converted at a far lower rate. Augustin’s teammates are shooting just under 35 percent on his assist opportunities, Curry’s teammates are shooting just under 55 percent on his. In fact, despite creating more of them, Augustin’s assist opportunities are creating nearly seven fewer points per game than Curry’s[2. Some of this includes the breakdown between two and three-pointers, which we aren’t accounting for here].
There is obviously a wide disparity in talent between Curry and Augustin, and we can assume that some of the difference is the added defensive attention Curry demands creating a ripple effect and more open space for him to pass into. But if we expand things to a team level, we can see that the contrast runs much deeper than just these two players.
The Pistons are generating far fewer assist opportunities per game than the Warriors and the difference in field goal percentage on those opportunities is near catastrophic levels. As we mentioned above, several factors are in play here and the biggest is probably that the Pistons simply don’t have the outside shooters that the Warriors have. SportVU data also allows us to see the shots each team has been taking, broken down by defender distance.
56.6 percent of the Warriors’ shots this season have come with a defender four feet or closer to the shooter. For the Pistons that number is 63 percent.
I picked the Warriors and Pistons, Augustin and Curry, simply because the disparity jumped out at me when I was first looking through the numbers. This sort of difference is not limited to them, or these two teams, by any means. John Wall and Jeff Teague have been creating a nearly identical number of assist opportunities, but Wall’s teammates are converting at a rate nearly 10 percentage points higher. There are too many factors at play to use these numbers as a measure of a player’s passing ability but they help reveal the interplay between all the factors that make up offensive efficiency—shooting, passing and moving the defense to create space.