Minnesota Sports Forum

OK, an upfront warning. Those here who put little or no faith in on/off metrics should ignore this thread.

A couple months ago there was a discussion about the value of on/off as a means of evaluating a players' value, and I posted that it was one of my 5 or 6 favorite stats when applied over a large sample size and in conjunction with other metrics. I wrote that if you take 20 current players who would be considered by most smart NBA pundits to be the consensus top 20 best in the league, all except one have shown consistently excellent on/off numbers over their careers...that is, their team performs much better when they are on the court than when they are off. And that one exception is Luka Doncic. Luka is in his 5th season now, and by the end of the season he will have a negative career on/off statistic according to cleaning the glass. Even worse, only Dwight Powell (who was terrific last night by the way) has a worse on/off this season among Dallas rotation players. And further, Luka's -7 last night was the worst on Dallas...except for Max Kleber who I'm sure we were all encouraging to keep shooting!

So what is it with Luka? Why is he the only consensus excellent NBA player I can find who doesn't have an impressive on/off career record? I love to watch him play, because similar to a Jokic or a Harden, he controls the offensive play in a manner seldom seen in a young player...almost like a man playing with boys (in fact, he reminds me of myself when I used to dominate my 8-year-old son and his friends ;-) ). But for some reason, and unlike any other superstar (and I believe he is a superstar), the Mavs aren't any better when he's on the court than when he's sitting.

For those who are also puzzled by this, I'm interested in why you think Luka is a rare outlier in the on/off stat. I have my thoughts. While he is an elite distributor (how many open threes did he create last night?), his tendency to turn the ball over hurts the Mavs...8 turnovers last night, including one where he unnecessarily tried a behind the back pass that ended up in the backcourt. And he averages 4.6 turnovers per game this season. Coaches hate turnovers, because they mean both an offensive opportunity lost plus too often an easy basket at the other end...a potential 4 point swing. I think his 4.6 turnovers per game is the main reason his team doesn't dominate when he is on the court, despite his indisputable elite talent.

My conclusion is that while Luka is an almost certain future HOFer in my book, his carelessness with the basketball (which was on display last night and almost led to the Wolves' stealing a big win) is something he needs to correct to realize his enormous potential...almost 5 TOs a game is not a prescription for success, and he's too nifty with the basketball to have that many. My only hope is that he doesn't learn how to be more careful with the ball this week, and that we get another 8 TOs on Friday night! That would really help us beat a very good team.

Someday when FNG and I get to run our respective NBA teams... we can get to the bottom of this debate as I'll have Morant, Doncic and DeRozan on one side vs. his squad led by Tyus Jones, Josh Green and Troy Brown Jr.

Luka Doncic sucks. The Mavericks are better without him. They should trade him while they still can.

AbeVigodaLive wrote:Someday when FNG and I get to run our respective NBA teams... we can get to the bottom of this debate as I'll have Morant, Doncic and DeRozan on one side vs. his squad led by Tyus Jones, Josh Green and Troy Brown Jr.

Ha, I'll take your team in a pickup game, Abe!

(And by the way, I got to watch you on the big screen last night in the Godfather...terrific movie, and even better in the theatre. I thought your performance was excellent, but you got what you deserved for selling out the Corleoni's to the Barzini's).

But seriously, as an NBA stat-head, I'm always interested in measures that clearly are outside the norm, and this is one of them. Like I said, I love Luka's game and see him as a top 20 player even at a young age, but when he is the only superstar (and I'm assuming everyone here agrees he is a bona fide superstar) whose team does not outscore the other team when he is on the court (and often by a huge margin), a curious NBA fan is going to wonder why. I mean he was the most fun player to watch on Dallas last night (outside of Dinwiddie perhaps...he was brilliant), but Dallas being outscored by 7 points last night when he was on the court almost gave us a huge win. So what is it that sets him apart from other superstars? Are there any Hall of Famers whose team were outscored during their careers when they were on the court (I can't think of any)? Luka is a great player on a very good team, so this one stat just seems odd to me.

1 extra TO or so a game over what a really good PG or star player would have isn't nearly enough to start to explain this quandary that keeps FNG up all night and pondering all day. :) Luka isn't quite doing the triple double thing that Westbrook did for years but he has the ball a lot. Westbrook averaged at least what Luka is averaging in terms of TO a game. What were his on/off metrics those OKC years?

To clarify, he has a career net rating that is in positive territory, meaning that his team generally outscores his opponents when he is on the floor. And since he is a starter, plays big minutes, and carries a massive usage rate, he must be doing something right for his team to consistently outscore the opponent while he's on the floor.

His net on/off rating is barely in the negative over his career and if you look at it season by season, it's been positive as much as its been negative. So I do not put him in the Zach LaVine category in this metric (Zach has had a negative net on/off rating in 6 out of 8 seasons, including this one!).

I think one of the hallmarks of Dallas over these last few years is that they tend to have a lot of really solid role players and they seem very well coached, whether it was Carlisle or now Kidd. I mean, they just seem to overachieve their collective talent level on paper. So it doesn't surprise me that their bench generally outperforms other team's benches, which in turn helps drive Luka's "meh" net on/off rating.

As for Luka himself, I don't quite put him in the same category of star as Jokic, Giannis, or Embiid to give three examples. You're right that he tends to turn the ball over quite a bit, but he also is only average when it comes to scoring efficiency. His 3-point shot and Free throw accuracy are just OK-ish. You combine that with the turnovers and the result is a lot of possessions from him that end in a miss or a turnover.

That being said, you can build an entire offensive system around him because he is capable of carrying such a massive usage rate and get more offensively limited players easy buckets. Now with Dinwiddie on board, he has some more help in shot creating for the team....such a better fit as compared to KP. I think Dallas could make some noise in the playoffs.

Q, I also think Dallas could do some damage in the playoffs...Dinwiddie is a nice piece to have coming off the bench. And Luka has an uncanny ability to find the open man beyond the arc, so you are going to see a lot of uncontested three's when you play Dallas. I was quite impressed with them last night, and also impressed that the Wolves stayed with them.

I agree with the reasons you added for Luka's relatively poor advanced numbers compared to other teams' best players (his net rating on basketball reference is a modest +2 both this year and for his career, while most "best players" are positive by double digits): He's not as efficient as scorer as other stars...his TS% is only average. And you make a good point about him not playing as much with a Dallas bench that is often dominant. But still, with his ability to find the open 3-point shooter, I would think playing next to very good shooting starters like Finney-Smith, Brunson and Bullock would make the Dallas starting five very potent. His numbers are still somewhat of a mystery to me.

Anyway, how Luke performs may not matter to us. He's been ruled out for tomorrow's game against Houston with an ankle, so there's a chance we won't see him Friday night. I have mixed emotions about that. If he's out, it helps our chances to move up from the 7th seed. But I'm a basketball fan, and I would rather see us playing a team not missing its best player. I suspect Carlyle has concluded they don't need him against Houston, and he plays Friday.

As for turnovers... meh.

Doncic leads the league in usage rate, so the ball is in his hands A LOT. And yet, Dallas is 3rd in the league for fewest TOs. Some of us might remember "stop-and-pop" Terrell Brandon. And some of us might have fallen for the ruse that was his ridiculously low TO rate. But it came with a downside... he was too cautious. The NBA rewards talented players who push the pace and even take chances from time to time.

Another angle. Doncic's TO rate is 15.5%. Jordan McLaughlin went on that crazy run where he had 50+ assists and only one TO. He only averages 2.0 per 36. But he's a caretaker, not a creator. His TO rate is only slightly better at 14.3%. Simply put, he has fewer chances when he's in the game to screw up.



[Note: James Harden came in 2nd for MVP while averaging 5.7 TOs per game one year.]

AbeVigodaLive wrote:As for turnovers... meh.

Doncic leads the league in usage rate, so the ball is in his hands A LOT. And yet, Dallas is 3rd in the league for fewest TOs. Some of us might remember "stop-and-pop" Terrell Brandon. And some of us might have fallen for the ruse that was his ridiculously low TO rate. But it came with a downside... he was too cautious. The NBA rewards talented players who push the pace and even take chances from time to time.

Another angle. Doncic's TO rate is 15.5%. Jordan McLaughlin went on that crazy run where he had 50+ assists and only one TO. He only averages 2.0 per 36. But he's a caretaker, not a creator. His TO rate is only slightly better at 14.3%. Simply put, he has fewer chances when he's in the game to screw up.



[Note: James Harden came in 2nd for MVP while averaging 5.7 TOs per game one year.]

Understood Abe on the TOs vs. Usage. Harden was also ridiculously efficient as a scorer, whereas Doncic is kind of "meh" on that front. I'm just saying that his shooting efficiency AND TOs puts him a notch below the Tier I stars in my opinion. But he is definitely a star.

AbeVigodaLive wrote:As for turnovers... meh.

Doncic leads the league in usage rate, so the ball is in his hands A LOT. And yet, Dallas is 3rd in the league for fewest TOs. Some of us might remember "stop-and-pop" Terrell Brandon. And some of us might have fallen for the ruse that was his ridiculously low TO rate. But it came with a downside... he was too cautious. The NBA rewards talented players who push the pace and even take chances from time to time.

Another angle. Doncic's TO rate is 15.5%. Jordan McLaughlin went on that crazy run where he had 50+ assists and only one TO. He only averages 2.0 per 36. But he's a caretaker, not a creator. His TO rate is only slightly better at 14.3%. Simply put, he has fewer chances when he's in the game to screw up.



[Note: James Harden came in 2nd for MVP while averaging 5.7 TOs per game one year.]

Yep, he did. And as others have said, it's not surprising with him having the ball in his hands all the time...like Luka. And yet Harden still finished the season with a double digit net positive rating for the yea,r while Luka is closer to breakeven in net rating. I think Q nailed it when he mentioned Luka's shooting efficiency. Harden's annoying talent of drawing fouls allows him to regularly finish above 60% in TS%, while Luka hovers around the league average. I see the key to Luka reaching his enormously high ceiling and joining the Top Ten ranks is finding a way to improve his mediocre shooting efficiency while also limiting silly passes like we saw Monday night. He's young and immensely talented, so I wouldn't bet against him.

And I wouldn't mind if Carlyle decided to rest his ankle one more game Friday...