In the NBA game, passing the basketball, and more specifically, the ability to create easy shots for one’s teammates, is a very important skill. The problem with identifying this skill, at least when we look at traditional stats, is that we don’t have yet have an adequate proxy that is both widely used and relatively easy to calculate. Which is why want to bring your attention to this ratio, Assist Percentage : Usage Percentage. And specifically, I’d like to use it to elucidate how it can be useful in evaluating college prospects.
To understand why Assist Percentage : Usage Percentage is a potentially important metric, we first have to understand why Assist Percentage in this case fails us. Specifically, it’s because Assists to some degree just track how often a particular player has the ball in his hands. The player’s who possess the ball the most are the most likely to make passes and thus also the most likely to rack up assists, even if their passes aren’t particularly special.
If we look at college basketball this year, Jaylen Brown is a reasonably good example of this effect. He racked up 4.2 Assists per 100 possessions, which seems a reasonable amount, until we look and see that he also racked up 23.6 shots per 100 possessions and 6.6 turnovers. Which means that in only something like 12% of the times a play Jaylen Brown made to end a possession (Shot Attempt+Assist+Turnovers) did it come by way of helping a teammate to score. Hence we add Usage Rate (a ratio that tracks the percent of the time a player ends a possession with a shot attempt or a turnover) to the equation as a proxy for how often a player had the ball in their hand.
Then we can better see which players rack of high Assist totals and decent Assist Percentages because they have the ball in theirs hands a lot, and which players, guys like Chris Paul and John Wall in the NBA, are particularly great at creating opportunities for your teammates. To give you a sense of how it works, I’ve made a table tracking NBA Players who have put up a Top 20 ORPM season in one of the last three years.
What do we see here?
1. We see a number of players who have been Top 20 ORPM guys multiple times in the last three seasons. Chris Paul, Steph Curry, Lebron James, James Harden, Russell Westbrook, Damian Lillard, Kevin Durant, Isaiah Thomas, Carmelo Anthony, Kyle Lowry, Mike Conley, Jimmy Butler and Ty Lawson. All of them are their team’s primary offensive ball-handler. Almost all of them are point guards.
2. That’s 35 of the 60 player-seasons on the list comprised by 12 players. 21 of the 30 Top 10 seasons. And ALL 15 of top 5 ORPM Seasons over the last 3 seasons.
3. We also see that almost all of these player-seasons carry an Assist Percentage : Usage Ratio right at over 1:1. The exceptions are of course the three seasons by Kevin Durant, the three seasons by Carmelo Anthony and the two seasons by Jimmy Butler, three wings besides Lebron to be part of this exclusive club. However, we also notice that these seasons tend on average to be less valuable than those of their counterparts.
Which is to say Kevin Durant has managed two +5 ORPM seasons the last three years, and Carmelo was close in 2014-2015. But the players with a ratio at over 1:1, Chris Paul, Steph Curry, Lebron James, James Harden, Russell Westbrook, Damian Lillard, Kyle Lowry have managed 19 such seasons. Which is to say, there have been 21 such seasons the last three years. Of those, 19 were by players who are among the best in the league at creating for themselves and others. The two others are by Kevin Durant, who is probably the best passing Wing outside of Lebron. It is almost impossible to be a great offensive player in the NBA without creating offense for your teammates.
4. Furthering this realization, we find indeed that 37 of these 60 seasons are by players nearing or considerably above a 1:1 Ratio of Assist Percentage to Usage Rate. Beyond even that, that 29 of the 36 Top 12 seasons were comprised by such players. In addition to the players listed above, Manu Ginobili, Deron Williams, Goran Dragic, George Hill, Kyrie Irving and Reggie Jackson all had qualifying seasons. The 7 exceptions to this group are again Kevin Durant (Twice, +6.27 ORPM with a ratio of .809 and + 5.5 ORPM with a ratio of .785), Carmelo Anthony (Once, +4.90 ORPM with a ratio of .559), Klay Thompson (once, +4.36 ORPM with a ratio of .529) Dirk Nowitzki (once, +4.56 ORPM with a ratio of .528), Louis Williams (once, +4.49 ORPM with a ratio of .504) Kawhi Leonard (once, +4.66 ORPM with a ratio this season of .48).
So again, the best offensive players are in general the ones who are most creative. Though it is possible to have a great offensive season just as a finisher of plays. However, this pathway to greatness does seem to be less sustainable form season-to-season, since it relies so much on offensive efficiency. And that finding also mirrors what we would find if we looked back through NBA history with very few offensive players (DR. J, Dominique Wilkins, Reggie Miller as examples) able to sustain absolutely elite offensive performance from year-t0-year without elite, or approaching elite ability to create for their teammates.
5. Lastly, we find another corollary to this argument. That is, if we look at John Wall, we find that elite ability to create for one’s teammates (A 46.3% Assist Percentage this year vs. 28.8% Usage) is enough to provide considerable value on the court (+3.05 ORPM) but if the player has holes in his game, in John Wall’s case, turnovers (5.5 per 100) mixed with an inability to score efficiently (51% True Shooting), it still won’t take the player to an absolutely elite level. For that you have to do everything.
Unfortunately, we can’t look at how all these players did in college, since the data isn’t in the Sports-Reference database. However, we can look at a number of them as well as a number of this year’s Top 40 offensive players to see how good they were in college, and if improvement is likely or possible.
1. One of the first things we see is that improvement is possible at the NBA level. Damian Lillard is not one of the league’s most intuitive passers, but with his athletic ability and shot-making ability he is able to create a number of good opportunities for his teammates. Indeed Lillard has improved from a decent ratio, albeit mediocre for a PG, ratio of around 0.8 to one in both his Junior and Senior season to regularly being over 1:1 in the NBA. Perhaps also because he has better teammates.
2. Jimmy Butler is the player with perhaps the most noticeable improvements. He’s up to a .806 to one ratio this year, elite for a wing, after being at 0.62 to one his senior year in college. Gordon Hayward is another such player who was relatively mediocre for his position in college (0.49 : 1) and improved a lot (above 0.7 : 1 this season). DeMarcus Cousins, who is only at .302 to one, is another player who improved a lot. (Close to a .5 : 1 ratio in most seasons, which is pretty good for a Big, even one with such high usage.) And we notice in general most college Power Forwards and Centers are worse at passing and creating for their teammates.
What do these players have in common? Besides a necessary basketball intelligence is that they all they can dribble and they all have athletic advantages over their competition. Jimmy Butler and Gordon Hayward are both 6’8″ wings. DeMarcus Cousins is a freak athlete center at 6’11” or 7’0″. Which is to say, these players, because they create mismatches, can change the shape of a defense. That’s something you would not reasonably project most current NCAA players to be able to do, especially if they can’t play the point in college.
The other thing that all three players have in common is that they were on relatively loaded college teams that didn’t need to rely on just one guy to create opportunities for himself and others. Cousins had John Wall, Eric Bledsoe, Patrick Patterson, Darius Miller to name a few. Jimmy Butler was flanked by Jae Crowder and Darius Johnson-Odom, who all took similar amounts of shots for Marquette. And Gordon Hayward was clearly an Off-Ball player at Butler a good portion of the time while Shelvin Mack, also in the NBA, held a lot of the traditional ball-handling and creation responsibilities.
So if we’re looking for current players who might fit into this group in the future, we’re likely looking for a player who has 1) Athleticism/Size for his position, 2) Dribbling ability, and 3) Is being under-utilized on his college team. I’d also add 4) Has reasonable amounts of turnovers per his role, since as we learned in my previous post, turnovers do often get worse when a player’s role increases.
2. We also notice that those who were very at this skill in college, John Wall, Kemba Walker, Reggie Jackson, Isaiah Thomas often remain so in the pros. Draymond Green, as well, who had a very rare passing ability in college for a player of his size and defensive position, and has maintained it at the NBA level.
3. This is also the case for players like C. J. McCollum and Khris Middleton, both of whom basically ran point for their team’s in college. Before this year we might have thought their college passing skills hadn’t translated to the pro game, but both players were good dribblers in college who remain so in the pros and both players create very well in their roles. C. J. McCollum has a 21.3% Assist Percentage against a 27.2% Usage Rate. A .777 to 1 ratio nearly identical to the .756 to one ratio of his junior year. Whereas Khris Middleton currently carries an 18.6% Assist Percentage this year as compared to a 22.9% percent Usage rate. A .812 to one ratio nearly identical to his sophomore year. And it should not be surprising to us, that these two college seasons, also carry another important indicator with them in that both players consistently made unassisted shots from all areas of the floor. (Though we’ll get to the importance of Unassisted Makes in a future post.)
Rodney Hood is another example of this class of players.
4. That of course doesn’t hold for everyone. Kyrie Irving for instance showed more ability and willingness to create for his teammates in his small sample at Duke than he has ever showed in the pros. (He’s also still reasonably okay at creating Assist Opportunities.) And most players do indeed get worse at creating opportunities for others as they move up a level. Paul George, for instance, is very good for an NBA Wing but worse now (.679 : 1) than he was in college (.803 : 1). But George is also the relatively rare case in which he’s the Pacers’ primary playmaker from a Wing position.
The reason most players get worse from level to the next is because their games aren’t strong enough to warrant that type of responsibility. A player like Klay Thompson is a great example of this. Another guy who often played Point in his junior year. He’s a fine dribbler and a fine passer, but he’s more a ball mover than a creator. I’d bet most players who have good but not excellent passing seasons in college are more likely than not to see their passing success at the next level decline to some extent. In part because many players will move from an On-Ball Role to an Off-Ball one. And the reason why there aren’t more of them on this list is because I’ve selected from a group of the 40 best offensive players in the NBA. Which means I’m mostly selecting from a group of players who can pass the basketball as a plus level.
5. Lastly, there are the players like Kawhi Leonard and Jae Crowder, both who excel at being ball movers for the Spurs and Celtics respectively. But neither of whom creates that many direct assist opportunities for their teammates. Thus their NBA ratios, which are around 0.5 to one closely resemble those of their college selves.
And it would probably be good moment to note, Kawhi Leonard is pretty much a total outlier as a perimeter player who can put up a near +5 offensive season without this skill. It should not be expected.
What Next?
So now we know not only that most Top Offensive Players excel at creating scoring opportunities for their teammates. And also that all of them were at least average at this skill in college (relative to position). Now I want to just hint at a few players who were terrible at creating for their teammates in college, noted draft busts Michael Beasley, Anthony Bennett, Derrick Williams and Al Thornton. Unfortunately, Sports-Reference.com doesn’t carry the Usage Rate data for all players, but it’s pretty easy to see, even looking at the raw Assist numbers as compared to the FGA, FTA and Turnovers these players generated they were in nethers of all potential NBA players when judging by this metric.
Michael Beasley had 38 Assists as compared to 577 FGA, 279 FTA and 95 turnovers. Which is to say, Michael Beasley never passed the ball to his teammates, nor, perhaps unfortunately for him, did he have to do so to be successful. And since we know, almost all truly great NBA offensive seasons are by players who not only pass the ball a large amount of the time, but create easy scoring opportunities for their teammates, it’s very likely we should have been more suspicious.
Anthony Bennett has similar numbers: 34 Assists as compared to 379 FGA, 177 FTA and 65 turnovers. Which is to say, it’s not just that these players have poor Assist to turnover ratios, it’s that they don’t really understand basketball as a team sport. (Anthony Bennett also has other terrible indicators, like that basically all of his made field goals from all areas are assisted, ie. created by others.
Derrick Williams: 43 Assists, 380 FGA, 331 FTA, 100 turnovers.
Al Thornton: 24 Assists, 244 FGA and 210 FTA and we don’t have the turnover numbers. Nor do we need them. NCAA perimeter players who get 1 Assist out of every 10 Field Goal Attempts they take are likely destined to fail, not matter how amazing their athleticism. And that’s a pretty standard ratio for these four players. Whereas you compare them to a guy like Draymond Green who had 141 assists against 345 Field Goal Attempts his junior year and it’s not hard to see the difference.
So What’s That Mean For This Year’s Prospects?
1. At the top of this table we see a number of players with a plus passing skill for college. We will get to it in a future post, but most of these players have a number of significant red flags elsewhere in their games:
Fred Van Vleet and Kay Felder are likely too small to defend at the NBA level. Van Vleet is also relatively unathletic. Felder doesn’t even defend well against college players.
Monte Morris, Demetrius Jackson, Cat Barber, Melo Trimble are all likely to be negative defenders at the NBA level. Morris, my favorite of the four, has games where he defends very well, but he also generally is a non-factor offensively when that’s the case.
Then there’s Dejounte Murray has major turnover and scoring efficiency issues.
2. That leaves us with five players who figure to get a real chance to play the point: Ben Simmons, who not surprisingly has exceptional numbers for a 6’10” Freshman with relatively high Usage numbers, Wade Baldwin, Kris Dunn, Gary Payton II, and Josh Adams. As players who can get to the rim and pass the basketball, these are all high upside Offensive players. We’ll discuss other aspects of their offensive games in more depth in the near future. But Baldwin, Payton II and Adams are among the most underrated players in the draft.
3. It also leaves us with a number of Wing Prospects who are excellent passers: Denzel Valentine, DeAndre Bembry, Caris LeVert, Thomas Walkup and Patrick McCaw. I think you could make an argument for all of them in the first round. And for Denzel Valentine and DeAndre Bembry in the lottery.
You may say, I’m not being consistent with marking down the Point Guards for defense but not Denzel Valentine. And there’s two simple reasons for that:
Firstly, it’s much easier to hide poor wing defense than poor PG defense. Virtually every team plays several wings who aren’t real threats to drive the basketball. Guys like Danny Green on the Spurs, Wesley Johnson on the Clippers, Mike Dunleavy on the Bulls, etc . . . And even a team like Golden State, for whom both Klay Thompson and Harrison Barnes have some dribbling talent, you would much prefer the offense run through one of them than through Steph Curry.
Secondly, Denzel Valentine has a lot of positive indicators on defense which we will get to in time. For one, he player for an excellent defensive team. For two, Valentine is an excellent defensive rebounder. And for three, the reason why Valentine is such an excellent defensive rebounder is because he never takes anything for granted. He’s precisely the type of player who you’d think at least has a chance to make up for athletic short comings by guarding Off-Ball players and using his intelligence.
4. Even though these four wings are excellent passers for the college game, my guess is that they will all be somewhat less effective at the pro level, since none of the four excel at getting to the rim off the dribble. That doesn’t mean they still won’t grade as pluses, perhaps significant pluses. It just means it’s hard to rely on players who might struggle to alter the shape of a defense at the point of attack continuing to be this good.
5. We can’t leave out Gary Clark. (A .9 to 1 ratio for a college power forward is very rare.) Clark probably projects as a pro Wing due to his height. But he’s exactly the type of player that becomes a big-time plus defender at the position. Indeed, many of the best NBA Wing defenders excelled, or at least held their own at the college PF position: Kawhi Leonard, Jimmy Butler, Jae Crowder, DeMarre Carroll, Kevin Durant, Andre Roberson, Justise Winslow. It’s a long list of plus defenders. Ass to that the fact that Clark is a max effort player, a player who does little things to help the offense, that he’s a decent dribbler and has shown some signs of a jump shot coming around, and he’s way underrated. And in the unlikely case that Clark could stick at the pro PF position, even that is saying too little.
(He slumped to end the season on offense, but I find him much more interesting on Upside than a number of more well-known guys. Given that relatively unknown college freshman with elite defensive skills, guys like OG Anunoby and Mikal Bridges, who was amazing last night basically guarding and shutting down everyone on Kansas, probably aren’t going to leave, I’d pick Clark in the Top 15 or 20. Much prefer the upside over supposedly safer guys like Hield and Allen, who probably aren’t any safer, since neither has a defensive game that will play at the NBA level.)
6. We see a couple of players who have huge Red Flag Scores from the PF and C: Skal Labissiere and Chris Boucher. A ratio of Assist Percentage to Usage Rate below .2 to 1 is a huge red flag. I actually probably like Boucher better, Both of these guys seem like they have upside. Both seem very unlikely to ever realize it.
7. We also see another player, Jarrod Uthoff, who has some potential to be a steal in the draft, due to his shooting and defense, both potentially plus, who may find himself without a position since he doesn’t have a really well honed passing skill at a .261 rate this year. (To be fair, he was much better as a junior.) Since Uthoff may be too small to play PF in the pros, this number has to really concern you. As it should with both Chriss (an identical .261) and James Webb III. For the most part, these are players who are only attractive NBA prospects if you think they can stick at PF on defense. A position from which their potential shooting ability becomes much more valuable. And I have huge rebounding concerns about both Chriss and Uthoff.
8. We see a number of Wings that rate at best as ball movers. Taurean Prince, Troy Williams (whose handle I don’t trust at all), Josh Hart, Jaylen Brown, Damion Lee, Jamal Murray (I don’t buy that he can become a top-notch passer without also becoming a top-notch turnover player), Buddy Hield, Malik Beasley and Allonzo Trier. (I might also consider Malik Newman here, whose resume might as well be a suitcase filled with red flags.)
These players are all going to have to shoot the living daylights out of the ball to bring value on Offense. (Except Brown, who’s going to have to drive it without creating turnovers.) And even then, there’s probably a cap on that value at around +4, a number which almost no NBA shooters get to. Then on top of that, these players are going to have to be even or plus defenders to become elite. Though only Prince, Williams, Hart and Brown really seem to have either the defensive ability, or at the very least potential to get there. I personally wouldn’t think of drafting any of these players before the end of the first round. And most of them I wouldn’t draft at all. Except for perhaps Brown, they all lack any kind of upside to be Top 10 or even Top 15 players at the NBA level. And even in Brown’s case it’s pretty unlikely.
9. Then we a number of guys who are pretty good passers in college who will probably have less success at the pro level, since they lack the athletic ability and/or skills to put a pro defense at a disadvantage: Grayson Allen, Dillon Brooks, Nigel Hayes, Dedric Lawson, Dorian Finney-Smith (who has real value if he can stick at the PF position and is probably underrated regardless), Malcolm Brogdon. I’d consider Brogdon to be the potential exception here because of his potential to handle the ball at the next level. Perhaps Brooks too. Dribbling is a really important skill, especially on the perimeter.
And again, for several of these players, especially Allen and Brooks, defense is going to be an issue at the NBA level.
10. We get Brandon Ingram, whose probably slightly underrated because of his age and his position as second fiddle in the Duke offense. If you were to ask yourself which prospects in this draft most likely are being underrated by this metric as Gordon Hayward and Jimmy Butler were, I think you only find two players: 1) Brandon Ingram. And 2) Wade Baldwin. And Baldwin is one of the top scorers by this method. The reason Baldwin may be underrated is because Vanderbilt ran a high-low offense through Damian Jones a lot of the time, which took away driving opportunities from Baldwin, where a player is most likely to create an assist opportunity for his teammates.
And that’s it. Huge improvement in a passing skill is really not something that should be expected, or even hoped for, unless the prospect has real dribbling ability and athleticism.
11. We see a number of adequately skilled PF and C, at least in terms of passing. Pascal Siakam, Henry Ellenson, Deyonta Davis, Brice Johnson, Thomas Bryant and Ivan Rabb. All between 0.3 and 0.5 to one. None of them are going to blow your socks off with their passing ability, but they’ve all shown enough to suggest that they could fit into an NBA offense given that they could stick at the PF position.
12. We see a number of uniquely skilled PF and C, at least in terms of passing the basketball. Daniel Ochefu, Robert Carter Jr., Chinanu Onuaku, Jakob Poeltl and Jameel Warney. To highlight just how so, here’s first a table of players in the Sports-Reference database who fit these critera: Forward, Defensive Rebounding Percentage over 20%, Assist Percentage over 14%, Steal Percentage over 1.5%, Block Percentage 4.0% or higher. BPM over 9:
Five players. Two big-time NBA successes in Draymond Green and Kelly Olynyk. Two players in this draft, Daniel Ochefu and Robert Carter Jr. And Javon McCrea, who was a pretty decent college player in his own right. That’s it.
Or another table, this time all College Forwards and Centers. Defensive Rebounding Percentage over 20.95%, Assist Percentage over 12%, Steal Percentage at or over 1.0%. Block Percentage at over 5.0%. BPM over 8:
15 player-seasons. Gorgui Dieng, Mike Muscala, Larry Nance, all NBA players of varying success. Aaric Murray isn’t the type of passer these five players are. It’s just a small sample size fluke. You can probably see it on a highlight video if there is any. And he was actually pretty selfish as a senior.
What we unfortunately don’t see is any great players. The reason for that is that the database only goes back to 2009 or 2010, and since that time, most any Big with talent is identified early and leaves for the NBA. So that’s why we only have 15 player-seasons on this list, and a number of them from Bigs in this class: Markus Kennedy who I didn’t include. It’s one of his best skills. Daniel Ochefu is on the list three times. (He can really pass, and it’s one place where Villanova’s offense has options that most other college offenses do not.) Robert Carter Jr. made several highlight real type passes the another night vs. Kansas. Ellis had a great game. It wasn’t all Carter’s fault, whose responsibilities shift when he traps screens. This is one of the reasons Carter Jr.’s game resembles that of Paul Millsap.
Jameel Warney has shown great aptitude for interior passing since he was a sophomore.
And then there’s the two sophomores in Chinanu Onuaku and Jakob Poeltl, who’ve made pretty remarkable improvements as passers since being kind of non-entities there as Freshman. It’s one reason both players should be in the lottery.
Though Onuaku is the better prospect if you ask me, since he’s the better current defensive player, the better overall defensive prospect, is tougher and more physical as a player, has a longer wingspan (7’2.5″ for Onuaku to 7’1″ for Poeltl), has the livelier body, and is basically a year younger. If you were to look at the Top 30 NBA players in almost any year, between 7 and 13 of them would be centers who bring a vast majority of their value on the defensive end. Shouldn’t that be what we are looking for in the draft, players who have some real potential to eventually become Top 10, Top 20, or Top 30 NBA players. In that light, I don’t understand how Onuaku isn’t being considered in the Top 14 of this year’s draft, since there are only so many players with the chance to be Primary On-Ball Initiators or Legit Two-Way Power Forwards in the vein of a Millsap, Green or Marion, or Off-Ball Wings who have the potential to provide value on both the defensive and offensive end.
