RealGM

Okay, here is my theory on championships.


Okay, I'll explain the chart(s). Each chart illustrates a different offensive flow but either will serve for explanation purposes.

http://img258.imageshack.us/img258/7289/currentheat.jpg
http://img844.imageshack.us/img844/2402/kobeheat.jpg
http://img25.imageshack.us/img25/8738/bronbal.jpg


Basically though, the chart postulates that a team must be good at outside shooting, have a good post game and also a dedicated wing scorer or scorers that can get them go to baskets.



The chart only focuses on the top three players, as unless you have two extraordinary talents that compliment one another very well in most cases you need three good players to win. It also serves the purposes of dissecting the Miami Heat well and illustrating the problems with Bosh quite nicely (if all three of these players USG's from last year were combined they would comprise something like 95+% of a teams offense so this really does show the issues they are encountering right now).



That said, the chart could theoretically expand to cover an entire basketball team if need be.



Again the basic idea is as follows:



On the offensive end: (The defensive end is an entirely different story and one not covered by this chart, although a great defense can cover for weaknesses in other offensive areas e.g. 04' Detroit Pistons)



Outside shooting, a good post game and a great wing/ISO presence is required on some level to generate points for ones team. The hypothesis is that in regard to weight in today's NBA a focus of 50% ISO/star based plays, 30% post presence and 20% outside shooting is required. The weights explain why a team with no great star on the perimeter ever wins (with a few exceptions).



The perfect use of each player on a team would have each player 100% specializing in the skill they are most good at, offensively.



However, because there are only so many touches available and only so many plays can be dedicated to ________ while maintaining offensive efficiency, in practice a player ends up delving in a multitude of roles if they are a star player.



Let's look at chart one:



http://img258.imageshack.us/img258/7289/currentheat.jpg



^^ Above we have the current big three of the Miami Heat. Their ability level in a given skill is the first number in a cell, and their current role in the offense would be represented in parenthesis.



These numbers are NOT meant to be exact numbers of a players ability level or role in the present Heat offense. Rather, they are meant to illustrate how even if you have three trans-formative talents (all three of those players would rate out as top 5-7 talents offensively based on the criteria listed above) if their skill sets are redundant they may still have a weakness in their team.



As you can see from the chart, Lebron, Wade and Bosh are all very good iso/go to scorers. Their rankings are 9/9/7.5, respectively. However, since they can only dedicate so much of the offense to iso based plays, they must divide the players between the three of them (40/40/20). The players must then get the rest of their touches in areas they don't specialize. The net affect is guys like Bosh predominantly jump shooting and posting up (even if they're bad at it) because they are the best option on the team to do so. And so a team with three top offensive talents can't make use of them.



Now let's take a look at what happens when we put Lebron in Kobe's place:



http://img844.imageshack.us/img844/2402/kobeheat.jpg



Here, the offensive efficiency sky rockets. Even though Kobe's absolute efficiency is only a fraction of Lebron's (45 for Lebron which would happen if he was 100% ISO/Bronball versus 40 for Kobe in that same scenario), he is a much better fit for this team. By adding Kobe to the team, Bosh can leave the post and take less jumpers (Kobe will do those things instead). In fact, Kobe will spend LESS time doing his specialty (iso based play) than anything else in favor of a less efficient player doing it. The effect?



Overall ISO/go to scoring declines while the team's net efficiency goes way up. Even though it would be most efficient for the offense in some ways for the best ISO player to get all of the ISO touches (a Kobe/Wade split) and the worst player to take less of the ISO offense (Bosh ala current Heat), the offense as a whole would suffer because then that player is forced to do things he isn't good at.



By using comparative advantage, however, and in some ways (ISO offense) acting at a loss the team ultimately gets at a net gain.



Now let's look at the dangers of Bronball:



http://img25.imageshack.us/img25/8738/bronbal.jpg



^^ Here, Lebron dominates the ball and go to scoring is at an absolute premium, as are his numbers. However, as a result not only are the other star players unhappy (resulting in a net decrease in their efficacy), but they are forced to spend most of their time and ability doing things they have comparatively little skill at (resulting in a huge waste of their talents).



Theoretically, Lebron could just take on ALL of the offensive load, but at that point he would experience a net decrease in efficiency (i.e. the ability score) due to fatigue and being the absolute focus of the defense. I didn't chart this, but basically imagine that as the % of the three offensive categories Lebron shoulders increases his ability level in each decreases due to fatigue.



That said, Bronball interestingly becomes much more effective if we take Wade and Bosh out and replace them with two shooters with 8's or 9's in the category (even if they have no iso or post ability). I didn't do a chart for this, but you can do the math.



In the playoffs, however, when defenses at the upper levels lower Lebron's offensive efficacy by honing in on him he will need another star player to take the pressure off. This is why Bronball always fails even though during the regular season it has incredible efficiency.



There, you see is the catch 22.



In order for Lebron to be able to play with star players they need to complement each other, but his one dimensional ball dominant style makes this impossible.



Final note: The total sum can be looked at as a W/L record/championship score, with a 65 being the cutoff point for a championship shoo in. As you can see, trading Lebron for Kobe would lead to a championship shoo in despite Lebron's absolute higher efficiency, due to the fact he (Lebron) doesn't fit with the other star players.



Obviously this is rough as I came up with in it like an hour, but I think it represents my thoughts.



I think I'll refine this, copyright it and call it championship theory or something and send it to ESPN. ^_^References:



http://img258.imageshack.us/img258/7289/currentheat.jpg
http://img844.imageshack.us/img844/2402/kobeheat.jpg
http://img25.imageshack.us/img25/8738/bronbal.jpg

P.S. You always hear people saying "Even though player A has a higher PER player B is better because he has a better all around game." This explains that thinking.

The concept is interesting, but you don't explain where these ratings come from. If they're based on your opinion, then none of this means much. If they're based on research, using shot charts, etc. then this is potentially worth something. Also, what's the basis for the claim that the output from the top 3 needs be over 65 in your index?

I'm hoping there's some research behind all this and that it's not just a veiled way of arguing Kobe > Lebron.

Nivek wrote:The concept is interesting, but you don't explain where these ratings come from. If they're based on your opinion, then none of this means much. If they're based on research, using shot charts, etc. then this is potentially worth something. Also, what's the basis for the claim that the output from the top 3 needs be over 65 in your index?

I'm hoping there's some research behind all this and that it's not just a veiled way of arguing Kobe > Lebron.

I only used Kobe vs. Lebron because it's a common example of better efficiency output vs. more diverse offensive game which fit the theory nicely. You could do this for several players.

65 is the basis because I tried it out with several trios and the ones that won championships almost always had ratings over a 65.

As far as rating players based on charts etc., does anyone have suggestions on the best shot data to use? 82games is incomplete and you have to assume too much from using NBA.com hotspots.

I still don't understand what you're doing here. When you talk about a player's "ability" are you referring to something drawn from data, or are you basing it on ratings you're assigning based on your own knowledge? By what criteria is Kobe's inside game and "8" while Lebron's is a "3"? How do you know each player's % of role in that aspect of the offense? What are you including in "inside game" (for example)? Is this post-up stuff? Does it include dribble penetration? If it doesn't include dribble penetration, why not? Why is an inside basket from a post-up set any different from an inside basket scored on dribble penetration?

You can find a shot distance breakdown at www.hoopdata.com.

I feel like a math teacher here. Show your work, young man!

Yeah, there's a lot of sketchy stuff here.

I'm agreeing that the ratings aren't quantitatively backed up. What I'm trying to show is a correlation between skill sets meshing and W/L records. If that end of it makes sense to you guys then I'll take the theory further (i.e. quantifying ability ratings).

jwilde86 wrote:I'm agreeing that the ratings aren't quantitatively backed up. What I'm trying to show is a correlation between skill sets meshing and W/L records. If that end of it makes sense to you guys then I'll take the theory further (i.e. quantifying ability ratings).

The idea of skill sets meshing is a good one. Roster design/construction (whatever term you want to use) is extremely important. It's something teams are constantly analyzing. If you have Player X, then you need Players Y and Z. But, if you have Player A, then Y and Z might not work. The critical thing in this kind of analysis is that next step -- quantifying abilities. That might not be easy because the records on shot distance (for example) don't go back very far.

This is an underrated topic. Though I wish it had more reliable references.

Well offensive success comes down to efficient shots, so at the rim and from 3. The 3pt shooter/3pt shooters/Lebron perimeter was perfect cause scores would either come on those open 3s for Lebron inside

On the Heat teams leave Wade or Lebron when the other is driving, then leave Arroyo all the team. Much easier to protect the paint without getting punished

The way this Heat offense works is if Wade commits to spotting up at the 3pt line when he's playing beside Lebron, beside another shooter at PG. Wade might not be a great shooter, but he can be a 33-37% one off wide open 3s which is enough.

And then they're advantage would be having an absurd player to run the offense when Lebron is gone

House
Spot up 3 shooting Wade
Lebron in 30/8/8 mode
Bosh
Ilgauskas

Is awesome offensively

But can Wade's ego take being a 15-20ppg guy? probably not

DONT need a theory bro...

Ainge said it best........you need a top 5 player to win a championship....

That, and/or the best defense in the league.

ry3303 wrote:DONT need a theory bro...

Ainge said it best........you need a top 5 player to win a championship....

That, and/or the best defense in the league.

No you don't, look at the 2003 Detroit Pistons, and you could probably make the same argument for the 1989/1990 Pistons as well. The Sonics team in the 70's didn't have a top 5 player either. Dynasties have all had top 5 players, but outside of that it's more about balance.

Ming Kong! wrote:

ry3303 wrote:DONT need a theory bro...

Ainge said it best........you need a top 5 player to win a championship....

That, and/or the best defense in the league.

No you don't, look at the 2003 Detroit Pistons, and you could probably make the same argument for the 1989/1990 Pistons as well. The Sonics team in the 70's didn't have a top 5 player either. Dynasties have all had top 5 players, but outside of that it's more about balance.

Those are anomalies, and thus don't disprove the notion that you NEED a top 5 player to win a championship.

This is a good article to read in that regard:

http://www.realgm.com/src_feature_piece ... _&_part_2/

nonemus wrote:

Ming Kong! wrote:

ry3303 wrote:DONT need a theory bro...

Ainge said it best........you need a top 5 player to win a championship....

That, and/or the best defense in the league.

No you don't, look at the 2003 Detroit Pistons, and you could probably make the same argument for the 1989/1990 Pistons as well. The Sonics team in the 70's didn't have a top 5 player either. Dynasties have all had top 5 players, but outside of that it's more about balance.

Those are anomalies, and thus don't disprove the notion that you NEED a top 5 player to win a championship.

This is a good article to read in that regard:

http://www.realgm.com/src_feature_piece ... _&_part_2/

I was thinking about that article when this thread was made. It was a valiant effort, not as embarrasing in retrospect as that MLB article before the NLCS saying the Giants team building strategy was flawed and doomed to fail, but ultimately, short sighted

What that author failed to realize is stars are judged by leading winners as much as their stats. I said at the the time there's probably between 7 and 10 players in the league with good enough stats to be called superstars if they led a title winner and thus fit into his graph. I pointed out Deron Williams and Yao as examples. Top 7-10 guys by most standards in their primes. But Deron puts up 20/10 on Utah's 60 win contender and Yao puts up 20/10/2.5 on Houston's? Now they're superstars. Players like Drexler and Dirk are examples of guys who weren't A list stars and then became them once their teams topped the league. Perhaps the best example is Duncan. I love Duncan. And his advanced stats are more superstar like than most would consider. But I feel pretty confident saying if his 20/12/2.5 was coming on 50 W pretenders, people wouldn't consider him a superstar like Shaq and Lebron, even if his impact was that high. Like I said Yao and Elvin Hayes put up similar stats and got nowhere near the MVP. You need either flash or HUGE winning to get in the MVP consideration. Duncan is a guy who's 'superstar rep' got a gigantic ass bump by how much he won compared to his stats, and he deserved every bit of it. The flipside is say Adrian Dantley. 30/6/4 on 58% shooting (.65 TS%+), but it doesn't matter cause he didn't win and people knew he was an empty impact guy. You imagine Dantley putting up those stats on constant 55 W+ teams and he's one of the all time offensive superstars.

Because superstars are defined by winning, you can't say you need one to win. You just need one who'd get enough of a boost by a best player on a title team run to be called a superstar.

The other problem with that article is this. Everything in the Russell era should be thrown out of the graphs. In an 8 team league, half the league had one of Russell, Wilt, Oscar, or West/Baylor, and early on Pettit. Regardless of Russ winning them all, it doesn't tell us anything that all the titles in the era went to these guys cause half the league had superstars. They were the equivalent of all-star caliber players today. Saying you need an all-star caliber guy to win a title doesn't say anything. So after that you're dealing with the 70s with a lot of non superstar led titles, and then the 80s to now where Magic, Bird, Jordan, Shaq, Duncan dominate, but all had great teams around them as well. Is it enough of a sample size to say superstars are required? Or should we say when players that good have great teams around them, it's hard to beat them... well until now. Also I'd point out how the Magic, Bird, Jordan dynasties were built on stockpiling future draft picks, a team building strategy that has since gone defunct cause of the lottery and GMs not being stupid. So it's harder to build teams around superstars now as Cleveland and New Orleans among other teams have found out. Again it just shrinks the historical trend for his claim

My opinion is people vastly, vastly underestimate culture and mentality when it comes to competing for a championship. It's not a sidebar. It's what the Lakers, Spurs, Celtics have and what the Heat don't. It's what the 90s Bulls, Jazz, and Rockets had, it's what the 80s Celtics, Lakers, and Pistons had. They all had it. Almost nobody has won without that foundation. Shaq and Kobe had some problems but for the most part got it together when it mattered. For the most part you need to play the right way to win and a lot of teams don't do it

You can even see it on lesser examples. Right now the Indiana Pacers are 9-7 with a 4.0 SRS (which by pythagorean is 50-55 Ws), while the Kings are 4-12 with a -8.16 SRS (which would lead to less than 20 Ws). I screwed up evaluating the Kings before this year. I saw two guys who drew double teams in Tyreke and Cousins, 3pt shooters like Garcia and Casspi, a decent PG in Udrih, useful big men like Thompson and Dalembert. I thought a guard and big man double team punch surrounded by a PG, shooting, and rebounding was a recipe for success and they'd look like Portland during Roy's soph year

The problem is the Kings are not a basketball team. They don't play the right way at all. Then look at the Pacers. Now there's a team who plays basketball. Like Sac they have nice pieces - a PG, a scorere/defender wing, a center, and some role players. But they actually fit into their roles, pass the ball, and compliment each other. They play basketball. They're a team.

The Pacers play like Portland and OKC and the Kings play like Memphis and the Clippers have the last 15 years. It doesn't matter how they built their talent level, Indiana's going to be more succesful than them going forward because they're a real basketball team. Portland/OKC's success is not about accumulating talent, it's about getting young players to play the right way.

Dr. Mufasa, I feel as if YOU are the mediator between Science and Religion, or in this case Stat geeks and Non Stat geeks

SideshowBob wrote:Dr. Mufasa, I feel as if YOU are the mediator between Science and Religion, or in this case Stat geeks and Non Stat geeks

Dr Mufasa wrote:My opinion is people vastly, vastly underestimate culture and mentality when it comes to competing for a championship. It's not a sidebar. It's what the Lakers, Spurs, Celtics have and what the Heat don't. It's what the 90s Bulls, Jazz, and Rockets had, it's what the 80s Celtics, Lakers, and Pistons had. They all had it. Almost nobody has won without that foundation. Shaq and Kobe had some problems but for the most part got it together when it mattered. For the most part you need to play the right way to win and a lot of teams don't do it

You can even see it on lesser examples. Right now the Indiana Pacers are 9-7 with a 4.0 SRS (which by pythagorean is 50-55 Ws), while the Kings are 4-12 with a -8.16 SRS (which would lead to less than 20 Ws). I screwed up evaluating the Kings before this year. I saw two guys who drew double teams in Tyreke and Cousins, 3pt shooters like Garcia and Casspi, a decent PG in Udrih, useful big men like Thompson and Dalembert. I thought a guard and big man double team punch surrounded by a PG, shooting, and rebounding was a recipe for success and they'd look like Portland during Roy's soph year

The problem is the Kings are not a basketball team. They don't play the right way at all. Then look at the Pacers. Now there's a team who plays basketball. Like Sac they have nice pieces - a PG, a scorere/defender wing, a center, and some role players. But they actually fit into their roles, pass the ball, and compliment each other. They play basketball. They're a team.

The Pacers play like Portland and OKC and the Kings play like Memphis and the Clippers have the last 15 years. It doesn't matter how they built their talent level, Indiana's going to be more succesful than them going forward because they're a real basketball team. Portland/OKC's success is not about accumulating talent, it's about getting young players to play the right way.

Meh, those are really just meaningless platitudes assigned in retrospect. Where was the winning culture in Boston prior to the KG and Ray Allen trades? Everyone gobbled up Larry Brown's rhetoric about playing the right way when he won with the Pistons in 2004, but the epic failure of his season with the Knicks shows that playing the right way means a whole lot less than having the right players.

I'm really working to flesh out the numbers behind this theory and hope to have something really substantiative in a few months.

Great original post.

This is what I've been arguing all along when it comes to Kobe vs Lebron. You just did it with numbers backing the theory up.

Basically, my argument all along was:
More diversified offensive game = a higher pool of players that you can select from to play with because your game can complement each other's

One dimensional offensive game = lower pool of players that you can select from to play with because you only want to pick players that can complement your game

Kobe vs Lebron

Dr Mufasa wrote:My opinion is people vastly, vastly underestimate culture and mentality when it comes to competing for a championship. It's not a sidebar. It's what the Lakers, Spurs, Celtics have and what the Heat don't.

Apart from the rest of your post, which is great, what leads you to make this conclusion? Culture I can understand, they do not have the history Boston and LA have, but why would this competitive, winning, championship mentality be lacking on the current Heat team?