Assists are an oddity in the statistical lexicon.
They are the only counted statistic that, without exception, require a pair of players to complete a task. One player must make the pass, another must make the basket. While those two actions must be present, the level of difficulty for each can vary greatly depending on the situation.
That fact alone makes it incredibly difficult to generalize how the credit should be partitioned for an assist.
In some cases the shooter is doing the bulk of the work to create that scoring opportunity, be it running off screens, hustling in transition or recognizing the opening to make a smart cut to the basket. But there are also times where an assist is literally the reflection of a shot created by the passer penetrating and breaking down a defense, or finding an unrealistic angle to sneak the ball through.
To decide exactly who is responsible for each made basket is a project that just isn't entirely and objectively feasible at this point. But one of the things we can look at to add some context is the location of shots that were assisted on by each player.
I looked at the 40 NBA players who racked up the most total assists last season and separated their assists into those that resulted in shots at the rim or shots from behind the three-point line (typically high-value shots) and those assists that led to mid-range jumpers and long two-pointers (typically low-value shots).
I split the two categories into percentages and created this handy graph to sort the results. The blue bar represents the portion of assists which went to shots from typically low-value areas of the floor. The red bar represents the portion of assists which went to shots from typically high-value areas of the floor.
Before we dive in, it's important to note that this conversation is somewhat academic. We're looking at assists, which means that every shot included here went in. When I say high-value and low-value I'm talking about the general value provided by all shots (makes and misses) from those locations, but here we're only looking at makes.
The idea of splitting assists into these two categories just implies a little about the level of difficulty. Defenses (the good ones at least) focus the majority of their attention into limiting three-pointers and shots at the rim. In general we would assume that assisting on those types of shots would be more challenging and require more work from the offense as a whole to stretch the defense and probably from the passer to create the opening. But of course, this is an assumption, certainly not true in all cases and probably not reliably testable until data from the NBA's new SportVU cameras is made public. But it's what we have in the interim.
It's really interesting that the non-point guards in this group (Kevin Durant, LeBron James, Kobe Bryant, Monta Ellis, Dwyane Wade, Nicolas Batum) tend to fall more on the high-efficiency side. It seems reasonable to assume that with scoring being offensive priority No. 1 these players are much less likely to pass up a shot unless the resulting potential assist has an extremely high likelihood of going in. With such solid individual scoring talent, even forcing a less than ideal shot for one of these players is probably a net positive compared to passing to just about anything but a wide-open teammate.
I also found it really interesting that the players on the extreme ends of the spectrum came from teams with strong stylistic tendencies when it comes to shot selection. At the high-value assist end, six of the first seven players (James Harden, Jeremy Lin, Raymond Felton, Andre Miller, Andre Iguodala and Ty Lawson) came from teams that ranked in the top five in the overall efficiency of their shot selection last season.
At the other end of the spectrum we find players like Paul Pierce, Rajon Rondo and Nate Robinson, who played for teams that relied heavily on inefficient, long two-pointers. It begs the question about how much a player's assists reflect the nature of their own abilities, and how much they are a reflection of the system and talent around them.
Only two players, Paul Pierce and Mike Conley, assisted on more shots from low-value areas than high-value areas. But again, I think a lot of that can be explained by context.
The Grizzlies had a tremendous lack of outside shooting last season, and many of the shots their offense created at the rim came off post-ups and were therefore unassisted. Pierce was often paired with Brandon Bass and Kevin Garnett, players who vastly preferred the space they found at the elbows and rarely made themselves available at the rim.
Ultimately this information tells us less about the quality of a player's passing or shot creation abilities and more about how they are actually expected to implement those abilities in their team's offense. However, when entering a discussion about the overall talent of a player, it's important to remember the context of their teams and teammates and how those circumstances can drastically affect their numbers.