MavsBoard

The Athletic posted their player ranks/tiers recently.  In their piece on 8/9 regarding players 80-125, they included this nugget about what it takes to be a "Championship Team".  BTW, Seth Partnow is the one who chose VORP, not me.  If you feel strongly that some other catch-all measurement is better (none of them are perfect), please feel free to do this analysis using your favorite metric instead (and maybe spare us the posts that say VORP sucks because Player X...who obviously sucks, is ranked higher than Player Y):


"The median championship team from 2010 to 2019 had players who summed to just over 15 “Championship Value Over Replacement.”  Looking at end-of-season rosters, the five most likely teams to win a title based just on their “star power” would have been

1. Brooklyn – 21.5 points
   
   
2. Milwaukee 14.75 points
   
   
3. LA Clippers – 12.5 points
   
   
4. LA Lakers – 12 points
   
   
5. Denver – 11.75 points
   

That feels about right, especially considering the degree to which one or two re-tierings of players who are on the border between two tiers up or down might rearrange the list completely. Phoenix was seventh at 10.5 points, suggesting they were a legitimate contender regardless of the good fortune from which they ultimately benefited en route to the Finals."


In case you are wondering, Luka is 4th in VORP at 5.0, KP is ranked 71st, Brunson is 89th, THJ is 90th, Bullock is 107th and DFS is 110th.  Of course, the moment I saw that, my mind immediately went to "I wonder where Dallas ranks...especially after the off-season.  Below are the VORP's for our top 8 players.  The list also shows each players all-time best in case you are wondering what happens "if Kidd brings out the best in Player X".

            2021    Best
Luka     5.0       5.0
KP        1.4       1.8
THJ       1.1       1.4
JB         1.1       1.1
Bullock  0.9       1.2
DFS       0.9       0.9
DP        0.7       2.1
Maxi      0.5      1.1

Tot:      11.6     14.6

Well, right off the bat, I notice 11.6 is pretty close to Denver at 11.75 (and Denver doesn't have its second best player right now...and LAC doesn't have it's first best player right now).  11.6 is also better than NBA finalist Phoenix at 10.5 (sometimes a little luck goes a long way).  We were 10.7 with JRich in place of Bullock.  If you could directly correlate those numbers, Bullock makes us 8.4% better.  Our record in 20/21 would have equaled 48 wins in a normal 82 game season.  8.4% improvement takes that number to 52 wins (take the over).

I did the "Best" column mainly to see what we might expect if KP was a better version of himself.  Frankly, I was hoping for more.  Even "Peak KP" is only ranked as the 52nd best player in the league.  You have to be careful with the "Best" column with many players.  It is a bit of a slippery slope.  Could KP return to 1.8...maybe.  It was just a year ago.  Might he become a 2.5 (which would move him to 31st in the league)?  There is nothing in his history to suggest that.  I do think KP at 1.8 is more likely than THJ at his 1.4 peak.  That is because that 1.4 was several years ago and is clearly an outlier compared to a very consistent level of about 1.1 since.  The same is true of Bullock.

Probably the biggest room for improvement comes from Maxi and Powell.  Maxi's 0.5 is an outlier.  The prior two seasons he averaged 1.0.  Note that "bad Maxi" returning to "average Maxi" of the two seasons prior is actually more potential improvement than "last season KP" moving up to "Peak KP".  Powell is a tougher call.  That 2.1 peak is definitely an outlier (Luka's rookie season where the top guys in minutes were Luka, DFS, Powell, Brunson, Barnes and Jordan in that order).  Someone has to put up numbers, even if you are the 20th ranked O in the league.  His other four VORP's in the last five years have been 1.0, 1.5, 0.9 and 0.7.  I suspect his post-ASG number last season is more like 1.0-1.2 (B-Ref doesn't give splits on VORP, but there is a huge jump in all of Powell's post-ASG numbers where they do provide splits).  

I don't think it unreasonable at all that we might get another 1.0 of improvement between Powell and Maxi normalizing around 1.0 and KP moving up even halfway back to what he was in 19/20.  That is another four game improvement to a hair over a 56 win pace (I'm telling you, take the over).  Much of that is health dependent.  Even rest games hurt as the drop off from KP to WCS is huge.  Same thing with all of the 9-15 bench guys.  They are all basically replacement players at best.  Are there clear paths to improvement?  LM doesn't help compared to Maxi/Powell except in terms of age.  Dragic doesn't help compared to Brunson or even S. Brown except for chemistry and a missing skill-set.  Those are nice things to have, but they don't move the needle any closer to a Championship VORP of 15.  Conley, DeRozan and Lowry would have helped as would Collins or Lonzo (but those last three not as much as you'd think).  I talk a lot about Batum, Crowder and Thaddeus.  All of those guys would have provided similar VORP improvement to some of the more highly paid headline guys.

If we are to get some improvement this season, it is probably more about making our ninth guy more than just a replacement player.  Right now our ninth guy is really three guys:

WCS behind KP/DP/Maxi

S. Brown behind DFS/Bullock

Burke behind Luka/THJ/Brunson

When you look at it that way, it is easy to see why they took a shot with Sterling and are rumored to have interest in LM and Dragic (but not in a way that it costs a top 8 guy to get it done).  BTW, another path to improvement would be Green passing Bullock as his game catches up with his athleticism.  Might be why Bullock is only partially guaranteed in year three.  Sterling is fully guaranteed for all three years and becomes the defensive version of Burke if Green passes him in a couple of years and makes Bullock expendable.

Exceptional analysis. I read the Athletic piece and wondered where Dallas stands post-FA. Thank you for putting this together. The Bullock addition looks to be sneaky good and I think a lot of fans will be really surprised when they see him play. I really like the DFS-Bullock-THJ rotation on the wing and there won't be such a huge drop off in shot making when THJ is off. Bullock has a similar quick release and he's not afraid to let it fly.

Keen observation regarding the 9th man. I like S Brown. That's a big addition if he shoots 40%. 

Will be interesting to see how Kidd handles the guard rotation and Burke's minutes. I think they're grooming Green and Hinton here but if Green plays he'll probably just be asked to stand in the corner and cut, at least until he can make a shot consistently. I don't think Hinton has NBA talent. We'll see. 

Will also be interesting to see what Kidd does with WCS and to a lesser extent Boban and Brown. Boban will likely have a similar role and one can only hope Kidd is as creative in situational substitutions as Carlisle was. One thing about Carlisle is he knew how to maximize positive match ups for role players. Brown may not make the team. 

I do wonder who starts next to KP. Does Kidd go back to Powell or Kleber? Does he start DFS at PF? That may be the biggest question of the offseason.

Another interesting point about the Bullock acquisition relates to Brunson. A lot of folks around here seem to want to trade him which in my view is nuts unless we're packaging him to get an All Star. People forget he was often the Mavs second best player last year. He was a sixth man candidate, he's young and improving. Brunson's upside is Fred VanVleet, but I digress. 

My point is Brunson - clearly one of Dallas' best five players - was often played off the floor when the Clippers went small. (You can legitimately argue with Carlisle's substitution patterns. Go back and look at the play-by-play and you'll see Carlisle repeatedly putting Brunson back in the game at critical moments, only to pull him minutes later to put more size on the floor). Unfortunately, at least according to VORP, the guy often replacing Brunson was contributing far less to winning basketball.

Bullock on the other had provides similar output (in different ways) while also bringing enough size to matchup with the Clippers small-ball lineup. If Bullock were on last year's roster, the Mavs likely beat the Clippers handily.

Wow! What a pleasant good morning reading this is. Thanks Dan. 

That Luka guy is really good. If he is 5th, how much can he improve our overall number? How much can he improve KP’s number? How much can Luka and KP led chemistry improve say 3 of our top 9’s numbers?
I agree on the take the over on wins too.

Amazing stuff DS. And it's immediately clear it looks like a team without the second star. That would something in between your superstar and role players. Something around 2.5. KP is very far from that.

It's interesting way to look at it. It might be much more effective to focus to get that second star and improve the team by 1-1.5. Rather than invest in trying to improve several role players where those margins are much less, and one would need to really do well in each of them and do the correct decision.

Jokic was at 8.6 VORP.  2nd place Giannis at 5.5.  That's nuts.


Devin Booker 88th at 1.1 VORP.  Montrez Harrell 49th at 1.9.


(I realize most advances stats will have noise like this.  Just fun to look at.  I wonder if each NBA team has their own formula/metric or how many organizations are lagging far behind.  It seems easier to tell among MLB teams and the way they're covered who is really strong/weak in analytics.)

(08-13-2021, 03:22 PM)jesusshuttlesworth82 Wrote: [ -> ](I realize most advances stats will have noise like this.  Just fun to look at.  I wonder if each NBA team has their own formula/metric or how many organizations are lagging far behind.  It seems easier to tell among MLB teams and the way they're covered who is really strong/weak in analytics.)

I always found fantasy basketball and rotisserie baseball easier to win than fantasy football.  Alas, no time for any of that for years now.  There is just a greater consistency to B-Ball players (both B's) than football players.  I used to do really well for years in a league with mostly TV and newspaper sports types.  It is easy to see where a growth curve is likely to peak and see where a trade might open up opportunities for an emerging player.  I never found a formulaic approach to football.  A change at left tackle can kill a wide receiver.  Just too much to track.

Back to your point.  Several years ago on another site I used to predict playoff series by minutes weighted PER.  Even as flawed as PER is, it was a great guide to predicting playoff winners...even lower seed winners.  It has been 10 years, but I think I remember it correctly called the Mav series up to the finals.  Dallas and Miami were essentially a toss up (caution:  it is possible I'm remembering 06...I'm not as young as I used to be).

Really nice post!

VORP regards Jalen Brunson a lot better than a lot of us have been giving him credit for. (me included) I felt he was being mis-valued due to the Clippers series, but I think I need to give him even more credit for what he means to the team overall.

I hadn’t read your thread when I posted mine about the 7+ starters. Your analysis is appropriately detailed and much more thorough, and I probably wouldn’t have posted had I read this. High-five!

So interesting. Love the focus on finding room for improvement. My 2 cents:

1. Luka, despite his obscene numbers, will continue to improve. Age, experience, conditioning, defense, turnovers, shot selection, free throws. All of these things say he will grow into an even more dominant force. 

2. Brunson is also likely to improve. Just getting through year 2 and his first playoffs is likely to make some difference. Add no longer getting scrub treatment from the refs and we’ll see an immediate impact on both his offensive and defensive efficiency. And with Timmy moving into the starting 5, I expect Jalen’s role to only increase.

Luka needs a top-30 teammate IMO. Ideally KP would make a big jump this year but if he's not capable of that(most likely), the Mavs will have to pivot to a strategy to bring that guy in.

I think in the second half of last year after all the COVID drama Dallas was probably a top 4 team, so that’s my expectation for next year.   This excellent data/analysis just confirms it in my mind.  Except. . . Jason Kidd has been one of the worst coaches I’ve ever seen.  The stuff he’s pulled is ridiculous.  He’ll have to be a lot better.  I think the fact that for the first time he’ll be coaching an influential player who won’t take his crap will help keep him grounded, but a total flame out is absolutely possible.

(08-13-2021, 05:17 PM)DallasBasketball Wrote: [ -> ]Really nice post!

VORP regards Jalen Brunson a lot better than a lot of us have been giving him credit for. (me included) I felt he was being mis-valued due to the Clippers series, but I think I need to give him even more credit for what he means to the team overall.

I hadn’t read your thread when I posted mine about the 7+ starters. Your analysis is appropriately detailed and much more thorough, and I probably wouldn’t have posted had I read this. High-five!

I mean last year the best lineup (2nd most used behind the starters) by net rating had no Luka. And it was the 6th best lineup in the entire NBA for lineups with 100+ min.

Brunson - JRich - THJ - DFS - KP

127.6 Off Rtg
107.9 Def Rtg
-----
19.8 net

Anyway I think this thread is kind of confusing the symptoms with the cause- I don't think teams win titles because they have over 15 VORP, I think teams have 15 VORP because they have 2 or 3 GREAT players plus good role players. Luka is doing too much heavy lifting to be sustainable in the playoffs, he needs a legitimate sidekick.

(08-14-2021, 02:15 AM)sefant Wrote: [ -> ]I mean last year the best lineup (2nd most used behind the starters) by net rating had no Luka. And it was the 6th best lineup in the entire NBA for lineups with 100+ min.

Brunson - JRich - THJ - DFS - KP

127.6 Off Rtg
107.9 Def Rtg
-----
19.8 net

That’s incredible. How do other Brunson line ups fare? Is there a line up that is almost exactly like this but without J Rich?

(08-14-2021, 02:47 AM)Branduil Wrote: [ -> ]Anyway I think this thread is kind of confusing the symptoms with the cause- I don't think teams win titles because they have over 15 VORP, I think teams have 15 VORP because they have 2 or 3 GREAT players plus good role players. Luka is doing too much heavy lifting to be sustainable in the playoffs, he needs a legitimate sidekick.

Maybe.  But in basketball, your success is often determined by your weaknesses.  In the playoffs, teams hunt your worst defender.  Sometimes you can't even play your best guys.  Two of the Laker's highest rated players were Harrell and Gasol and neither couldn't get any burn in the playoffs.

We probably need to define what a GREAT player is.  In VORP, a 3.0 is the 22nd ranked player in the league.  2.5 is the 32nd ranked player.  It is probably hard to be GREAT if you aren't in the top 32, right?  And what is a "good" rotation player.  There are 106 players with a rating of 1.0 or better (about 3.5 per team).  There are 172 players with a 0.5 or better (about 6 per team).  Which is good enough to be good?  

Let's look at four teams:

Team A's best players are rated 3.2 and 2.8
Team B's best players are rated 3.6 and 2.5
Team C's best players are 4.2 and 3.1
Team D's best players are 3.7 and 3.0

Only three teams have a second best player at 3.0 and I've listed two of them here.  One of these teams was 34-38.  One of these teams made the finals.  Another had the best record in the league and another lost in the first round.  I would propose that these teams are indistinguishable by the differences between their GREAT players, yet two were highly successful and two weren't.

The difference was depth.  The team with the best record in the league had seven guys rated 1.0 or better.  The finalist had six.  The sub-500 team and the first round flame out had two.  There were five teams with five or more players rated 1.0 or higher.  Utah, Portland, Phoenix, Philly and Milwaukee.  Only the champs had a player ranked more highly than Luka.

I don't disagree that it would be nice for Luka to have a buddy who is also in the top 30.  Hell, even the top 50.  But if there isn't a clear path, then shoring up weaknesses is a valuable exercise.

(08-14-2021, 07:21 AM)DallasBasketball Wrote: [ -> ]That’s incredible. How do other Brunson line ups fare? Is there a line up that is almost exactly like this but without J Rich?

Decent i guess. Some lineups are hard to explain in general. Maybe because KP was pretty bad last season or because THJ straight up didn't work in some lineups. The next 5 Brunson lineups minutes wise:



Luka - Brunson - THJ - DFS - WCS

145.4
108.0
-----
37.4 net

62 min


Luka - Brunson - THJ - DFS - KP

115.5
147.5
----
-31.9 net

47 min


Luka - Brunson - THJ - Kleber - Powell

117.5
104.0
----
13.5 net

45 min


Brunson - Jrich - THJ - DFS - Powell

107.1
103.4
----
3.7 net

42 min


Luka - Brunson - JRich - DFS - KP

121.9
104.2
---
17.8 net

34 min

(08-14-2021, 08:18 AM)sefant Wrote: [ -> ]Decent i guess. Some lineups are hard to explain in general. Maybe because KP was pretty bad last season or because THJ straight up didn't work in some lineups. The next 5 Brunson lineups minutes wise:



Luka - Brunson - THJ - DFS - WCS

145.4
108.0
-----
37.4 net

62 min


Luka - Brunson - THJ - DFS - KP

115.5
147.5
----
-31.9 net

47 min


Luka - Brunson - THJ - Kleber - Powell

117.5
104.0
----
13.5 net

45 min


Brunson - Jrich - THJ - DFS - Powell

107.1
103.4
----
3.7 net

42 min


Luka - Brunson - JRich - DFS - KP

121.9
104.2
---
17.8 net

34 min

I think one problem with this kind of lineup-data is, you can't see whom they played against.

And you just can't ignore luck in such small sample sizes.

Said this, I'm always confused, when I see how bad Luka-Brunson-THJ-DFS-KP looks here. This makes absolutly no sense to me.

Just like TPA. VORP is another BPM adjusted for minutes/games stat. Not sure if I am the biggest fan of the methode but it is probably enough to give us a frame to work with. Don´t think the choosen catch all metric really matters in this case. They all have strengths and weaknesses.

Overall it brings us back to a discussion about different ways of roster building. Dan highlighted different ways to reach contender level in a salary cap driven league.

To simplify it with a few examples:
Top heavy contenders like the big 3 Heat. With three guys combining for a 15+ VORP and mediocre depth.
A duo with a more balanced supporting cast. For example the Lakers or Clippers in the last few seasons. Two star players combing for a 10+ VORP.
A superstar that is surrounded by good role players. Best example would be the 2011 Mavs team. Maybe the Raptors team around Kawhi.
Or the rare team with incredible depth but no clear star player with great boxscore metrics. Like the 00s Pistons or the 2020 Heat team that made the finals.


What is the best option for the Mavs?
They already have Luka. And okay depth. As currently constructed they lack the necessary 2nd or 3rd best player to reach the next level. Maybe KP can be that guy. Maybe a trade can land the Mavs that kind of player. Or a trade turns a struggling 2nd/3rd option into even better rotation/depth pieces that can make up for the lack of a clear 2nd best player.

(08-14-2021, 08:35 AM)Mapka Wrote: [ -> ]I think one problem with this kind of lineup-data is, you can't see whom they played against.

And you just can't ignore luck in such small sample sizes.

I did an analysis of this a while ago. That "Brunson" unit was feasting on opponents second units in late first and second quarter. It was the usuall substitution pattern with Luka and Maxi out and THJ/Brunson in. However, that same unit was mostly ineffective when tried against best opponent lineups, like in fourth quarter. After a while they stopped using it in those situations.

(08-14-2021, 09:03 AM)omahen Wrote: [ -> ]I did an analysis of this a while ago. That "Brunson" unit was feasting on opponents second units in late first and second quarter. It was the usuall substitution pattern with Luka and Maxi out and THJ/Brunson in. However, that same unit was mostly ineffective when tried against best opponent lineups, like in fourth quarter. After a while they stopped using it in those situations.

This sounds like it's a bad thing, but I really like having a backup who is able to abuse opposite backups.