How to Evaluate Playmaking like a Scout

Assists Don't Tell the Whole Story...

In basketball, playmaking is a really hard thing to quantify. At its core, the term “playmaking” tries to describe a player’s ability to set up their teammates with good shots. A playmaker is someone who does that well when the opportunity presents itself; A really good playmaker creates those opportunities themselves, whether it be with their own scoring gravity (a la LeBron James) or their elite vision (think Steve Nash). Some do it within their limited role in the offense, like my hero Boris Diaw on the Spurs’ 2014 title run.

Everyone’s after playmaking, because open shots mean high percentages of scoring. Higher percentages of scoring mean more points, and more points mean winning basketball games. So in order to win basketball games, you need as much playmaking as you can get…And the primary metric for quantifying playmaking is none other the assist.

I remember back to setting up a Fantasy basketball league a while ago, and the most contentious issue by far was how many points an assist should be worth. On one hand, a good pass can quite literally lead to a wide-open layup. Like this one:

If we were to break down the “responsibility” for those two points, virtually all of it should land on the passer, not the shooter. But then there’s those assists where a player makes a basic pass that just so happens to end in a bucket. Sometimes they don’t do a lick of work, like this one, which was logged as an assist for Beal:

On top of all of these, the three pointer itself complicates the issue. Should a tough pass that frees up an elite three-point shooter for a wide-open look be worth the same as someone shuffling off a shot clock buzzer beater to a teammate? Maybe it all evens out, but it doesn’t seem exactly prudent to just assume that fact.

What I want to do here is distinguish the “assist” as a statistic from “playmaking” as a concept. There’s plenty of reasons to do this, but I have one reason in particular. As someone deeply interested in scouting, playmaking remains the skeleton key to unlocking many players’ ability to function in high octane NBA offenses. It’s at the professional level, not college, where players will be confronted with the absolute necessity to make quick decisions on every possession. Ball-stoppers kill the modern NBA offense.

For the sake of teambuilding, I simply want to understand how we determine great playmakers. The conversation weighs assists too heavily to me, and I want to change that by determining how players get their assists. How do we work toward a comprehensive understanding of “Playmaking” with this in mind?

I don’t expect to answer every aspect of this question right now. But maybe with the help of some numbers we can start to evaluate playmaking like an NBA GM might do when looking for pieces to put together on an NBA Championship team.

Playmaking Archetypes

As I said in the introduction, playmaking is a term we use often to describe how well someone can set up their teammates for easy shots. Likewise, when we talk about the best playmakers in the league, we consistently list those at the top of the leaderboard for assists. But those two things are not necessarily the same, are they?

Draymond Green, for example, is a good passer… but is he as good of a passer as Steph Curry’s shot-making allows him to be? What about Russell Westbrook, who led the league in assists three times, yet is still considered selfish in many circles?

I think that this issue can be resolved in part by describing the types of playmakers that are out there. Here I will break them down into two categories: Gravitational and Contextual Playmakers.

Gravitational Playmakers are players that open up teammates because of, well, their gravity. The archetype that people think of in this case is someone like LeBron in his prime, or Giannis. Both are freight trains who demand three people in the lane to stop them in transition That gravity opens up tons of passing lanes, like here:

Gravitational Playmakers always have a high usage rate and lots of assists. However, there’s more than one way to attract attention from the defense. While someone like Russell Westbrook attacks the paint with the threat to score, a player like Chris Paul seems to let the passes come to him. How do we differentiate this? Let’s look at some numbers.

Let’s check out Westbrook’s MVP season in 2016-17 as opposed to, say, Chris Paul’s season that same year. Russell Westbrook finished the season averaging 10.4 assists as opposed to Chris Paul’s 9.2 that very same year. Was Westbrook a better playmaker? By assists, sure, but when you look at the deeper numbers, it actually doesn’t seem quite so obvious.

Russell Westbrook’s ungodly 57.3 AST% (one of my favorite stats for scouting a prospect’s passing ability) is paired with a league high 41.7 USG%. That second figure means that Westbrook is in control of nearly 42% of all possible possessions when he’s on the floor. For reference, that’s almost 4% higher than Luka Doncic’s busiest season, and 3% higher than Kobe Bryant’s 2005-06 campaign where he averaged over 35 points a game. In fact, it’s more than James Harden, Michael Jordan, or anyone: Westbrook’s usage in 2016-17 was the highest ever recorded in the NBA.

What does that mean for his playmaking? Well NBA.com’s handy Advanced Stats page can help us out there. Despite being 3rd in assists, he ranked 13th in hockey assists (passes that lead directly to assists). Even more shocking, he ranked 19th in passes per game.

Chris Paul, on the other hand, ranked 4th on hockey assists, and 3rd in passes per game. All this despite having a 24.4 USG%, almost half as many opportunities per game.

So it would seem as though these two players gather assists in very different ways. Westbrook has the ball in his hands a lot, drives the ball looking to score, and passes out once he's garnered enough attention for his potential to score. A bit like this:

Chris Paul, on the other hand, usually gets his assists by focusing on the passing itself. Sure he can score, but he’s passing the ball around more (hence the top ranking in passes per game), and taking what’s given to him. He has a high usage rate (anything over 20% is substantial), but he gets his assists by asking "What get's my teammate the easiest shot?" A true "pass first" player, like this:

Whereas Russell Westbrook would probably attack the outside foot of Jaxon Hayes and exploit the big man's lack of quickness and leverage, Chris Paul looks to pass first. He doesn't even look to score here in fact; look at where he's looking the whole time. Hayes had to respect CP3's gravity, but it was never going to be Chris shooting this shot... it was always going to be the roll man.

So despite both being Gravitational Playmakers, Russ is certainly not passing the ball in the same way Paul is. Perhaps here we can distinguish the two by describing one as a Gravitational Passer (Chris Paul) and a Gravitational Scorer (Westbrook). Both terms describe playmaking talent, but one operates as the facilitator of the offense while the other operates as the offense itself due to the massive usage rate.

It's tough, admittedly, to do these sorts of analysis, because one statistic cannot capture the entirety of playmaking talent. Passes per game, AST%, USG%, hockey assists, and total assists all play a part in determining these things, but the overarching takeaway is that assists do not necessarily determine the best playmakers in the league.

I want to be clear about something: This statistical reality does not mean Westbrook was a worse player than Chris Paul during the 2016-17 season. We are evaluating playmaking talent here, not overall player talent. LeBron James qualifies more as a Gravitational Scorer than Passer, yet he took his team to the finals basically every year for a decade. Whether you believe in the heliocentric model of offense is not the question (albeit one I would love to explore soon). What we can reasonably conclude, though, is that offenses revolving around one player produce assist numbers that may or may not actually reflect a player’s passing ability, but their ability to attract defenders.

That concept becomes important, say, as players get older. If player’s gravity dissipates with age, will he still be able to pass well without as much gravity? Or will he continue to play as if he still has that gravity? That seems to be the issue the Lakers ran into when they traded for Russell Westbrook not too long ago – they expected his playmaking to hold up even as his scoring prowess took an obvious downturn. This is not to mention how Westbrook has always been a high usage player – much like LeBron and Anthony Davis themselves. His playmaking reputation may have been, at least in part, tied up in his usage and gravity, but once those things had left Russ, he wasn't nearly as effective anymore. They didn’t need Russ to be that sort of player for them; He needed to be a compliment to the Lakers’ stars. He needed to adjust from a Gravitational Scorer to a Contextual Playmaker.

Contextual Playmakers are players who get their assists not as a result of any high usage, but because, in their limited touches, they pass quite well. These players will rarely top the assist leaderboards, but nevertheless can oftentimes be better passers than the league leaders.

A perfect architype of this sort of player is Draymond Green. Green has never eclipsed 19 USG% in his ten-year career, yet he remains one of the best passers in all of basketball because of how efficiently he uses his touches. During the 2016-17 season, Draymond ranked 9th in the league in assists, at 7.0 per game. That’s below Paul and Westbrook, which makes sense considering Green’s 16.2 USG%. However, despite his getting significantly less touches than either Paul or Westbrook, Draymond still ranked 6th in passes per game! He was working within the confines of the Warriors’ offense with less opportunities (24th in the league in touches per game), but still producing a massive quantity of playmaking.

Contextual Playmakers are usually good, quick passers in situations like short rolls or exploiting slow rotations. Look at how Draymond makes the right decision time and time again on the short roll:

Players like Green serve as the connective tissue of a functional offense. They make things flow – sort of like Chris Paul’s passing, but in much lower doses. Contextual Playmakers will usually have a high AST%, which is one of the reasons I love the statistic for scouting matters – most guys will not be gravitational forces, but they will be asked to make second-side decisions, and AST% tracks that better than most stats because it scales the percentage of their possessions rather than the aggregate number.

People think that you need one of those Gravitational forces in order to win championships, but that's actually not true. The 2014 Spurs, my favorite team of all time, had a few Gravitational players, but none to the level of someone like Lebron James or Dwayne Wade. But they were all very, very good contextual passers. Here's the top 11 minute getters for the Spurs:

I narrowed the stats down to usage and Assist %. Those usage numbers may seem high, but when you account for the fact that no one played over 29 minutes a game, the volume is actually pretty small. Ginobili, for example, only played about 22 minutes a game. But then you look at the AST%, and only one player was in single digits, Danny Green. The result? The ball was flying around the court.

Now to the left are the Heat's numbers, who the Spurs played in the Finals that same year. LeBron and Wade had massive usage rates to go with massive assist rates (not pictured is their combined 71 minutes per game). They had 2 great playmakers, one good one (Mario Chalmers), and not much else. The result? The ball stagnated constantly on offense when it mattered most, and the Spurs blew them out of the water in the NBA Finals.

So it should not be too hard to see what the value of contextual playmakers is to winning. If you're still not convinced, look at what those statistics look like in action:

So to reiterate: There’s three types of playmakers that we can reasonably call “good” playmakers. The first two, Gravitational Playmakers, attract a lot of attention from the defense and parlay that into open looks; And while they may possess similar assist totals, evaluators must contextualize that number with their particular role in the offense to see if they are elite passers or not. The third type of playmaker, Contextual Playmakers, possess high passing marks in their more limited opportunities. They do not possess a ton of attention from the defense, but nevertheless do a good job of finding open men to give their teams good shots.

In a rather ironic way, it's usually easier to see how good of a passer they are with less usage, not more, because they are not demanding as much gravity, and so need to rely on their actual passing to be effective.

Scouting

Okay, so I’ve come up with some fake labels to categorize different players by their playmaking ability. So what? Well, maybe we can return to my reasons for writing at all: Teambuilding.

In regards to scouting: When evaluating players’ playmaking ability, perhaps we can use these archetypal models to project their numbers into their role in the offense. In this upcoming cycle, for example, South Carolina’s GG Jackson possesses a 6.4 AST% as of my writing this (February 14th 2023). That number is downright terrible. So what GMs are banking on with GG Jackson is that he will, in all likelihood, not be a contextual playmaker; You will make him the focal point of your offense. From there, a GM will ask themselves the following: Why is his AST% so low? Is this something we can work on? Does he have flashes of passing? Is he good enough at scoring that his passing does not matter? This is a guy who will go in the first round… what if he simply refuses to pass the ball? Look around the league. Are there any key players that work on title contenders with a 6.4 AST%? These are the sorts of questions you need to be able to answer if you’re picking in the lottery.

Okay, sure. What about if I’m picking 25th in the draft?

Another scouting scenario: A GM is drafting for a contender, and they want a player who can bolster their defense. The only problem is that this player, while elite on the defensive end, can’t really score too well – no shooting, nothing. Will they be playable, or a complete detriment to your offense? In this scenario, looking at playmaking ability can be very helpful. Maybe the player can work as a Contextual Playmaker, as he passes well out of short rolls. Maybe he becomes the connective tissue on your offense that sets other guys up even if he himself isn’t putting the ball in the basket. These are important questions for all front offices to answer, but it requires a ton of attention in contextualizing each player’s playmaking ability.

Now, we have to realize that there’s no singular statistic that predicts NBA success. I’m not saying that GMs should go out and take every low usage player with a high AST% (although it wouldn’t be the worst thing in the world with a late second round pick). No, more than anything I want to simply show how NBA front offices can look at this aspect of scouting. Given these simple archetypes, perhaps we can more seamlessly predict a prospect's role in an NBA offense with their role in mind.

So then evaluating someone’s playmaking talent becomes absolutely crucial to putting together a winning team. Players don’t have to get 10 assists a game for their passing ability to be valuable, in fact, usually titles are full of really solid contextual playmakers. It's something to keep in mind as you try, like every GM out there, to see how each player would fit within your team's system.