MAVS 108, PELS 92

[fifteenth]

Are you talking about losing every game, or accumulating a negative point differential over a season? 

If we are talking about losing every game, sure. I would think they were either tanking, which involves intentionally performing badly, or just performing badly enough as a group to lose every game. 

I was talking about being outscored in every game. I went to the ultra extreme example to find the place where you and Kam can agree. :-) If a team gets outscored in each and every game, then +/- does say something about performance. 

Also, performance, in its usual sense, takes account of expectations. A team that came up from last in the league to having a -1 total point differential over the season might be considered to have performed very well, and vice versa.  

I completely agree with this. I had thought about performance based on expectations while sorting through this discussion, but hadn't brought it up. 

However, in either case I would not at all think that a losing record or a negative point differential meant that every player on the team played badly. It is pretty common for young bad teams to have outstanding young players, and no one dismisses them because of their team's season-long point differential or record.

Yes to this as well. 

Sports inherently involve an element of randomness, which means that the better team, or even the team that plays better that night, doesn't always win the game. It's part of what makes sports exciting. If we wanted to see a competition where randomness isn't a factor, we could watch a chess game. 

Here I agree as long as the game of chess is played between two decent players 

I have never taken the position that +/- is a worthless statistical tool. In fact, I think it is a very valuable tool when it is used properly. I also think it can be misleading, and it is easy for a fan to think it proves more than it does. I think it can be related to performance, but doesn't tell you anything specific about performance, unless you simply define performance to mean having a good +/-.

Agree here too!

I could be completely wrong on this, but I think Kam would probably agree with all these points as well.

[Kammrath]

Here is a metaphor to try to explain how I see performance:

I measure a car's performance based on it achieving its purpose and goal: moving me from point A to B. I can have a car whose engine is not "performing" very well or whose transmission "performance" is poor and whose wheel alignment "performance" is off, still get me from point A to B. So if it still accomplishes its main purpose then I will grade it as having performed well in an overall sense. At the same time I could have a car whose engine is "performing" perfectly, with a smoothly "performing" transmission, and wheels perfectly aligned that doesn't get me from point A to point B because its radiator did not "perform" and the car overheated and the engine was killed.

Likewise, I see the Mavs as a TEAM "performing" well with a 7-3 record because THAT is the objective (they have had a positive point differential in 7 of their 10 games). But when I look under the hood, I am concerned that the "performance" of certain parts might need addressing or they might end up sabotaging the whole thing. Team shooting on open shots is one. The team D when Luka is playing is another. 

At the end of the day, I only care about the team "performing" in the sense of getting wins. But the "performance" of pieces within that are still worth looking at because a misfiring engine or failing radiator could sabotage the season at some point. The Mavs could win 67 games this year and then have a weakness exposed in round one for example.....I know, I know....too soon.

[Jommybone]

Seems to me that +/- flunks the scientific method by a long shot. There’s no control group, and at least the raw numbers make no attempt to account for the large number of variables. 

The most apt analogy I can think of is measuring the air temp of an individual’s location. You might think we could assume the variables will even out over time, and the Texans will have warmer air space than the Canadians. But that’s a poor guess when you don’t consider how much time they spend outdoors, where the thermostats get set, how tolerant they are of extremes, and a host of other things. 

If we just assume the things we don’t know are a wash, we come to conclusions like Alex Caruso is better than Russel Westbrook.

[BackToSquareOne]

Seems to me that +/- flunks the scientific method by a long shot. There’s no control group, and at least the raw numbers make no attempt to account for the large number of variables. 

The most apt analogy I can think of is measuring the air temp of an individual’s location. You might think we could assume the variables will even out over time, and the Texans will have warmer air space than the Canadians. But that’s a poor guess when you don’t consider how much time they spend outdoors, where the thermostats get set, how tolerant they are of extremes, and a host of other things. 

If we just assume the things we don’t know are a wash, we come to conclusions like Alex Caruso is better than Russel Westbrook.

I like(d) this post but Alex Caruso is better than Brodie ToxicTripleDouble. So, I'm kinda torn.

[mvossman]

Side note: Talking about more advanced +/- metrics. As far as I know Jeremias Engelmann (creator of ESPNs RPM, before they changed the prior/formula multiple times) is still one of the senior analysts. My best guess is that he is doing similar things for the Mavs.

I agree with everything you said here.  The kicker is that raw +/- takes a long, long time to normalize.  Like many seasons.  It is incredibly noisy in 1 game (and even 10 game) samples.  The advanced stats that leverage +/- do a much better job of handling smaller sample data.  The really good ones start with box score stats and then apply +/- logic looking at both teammates and opponents.  I found a really good article which goes over which are considered the best in the industry (you will notice that PER comes in last, which validates both your and my thoughts on that particular statistic).

What is the best advanced statistic for basketball? NBA executives weigh in | HoopsHype

[BackToSquareOne]

I agree with everything you said here.  The kicker is that raw +/- takes a long, long time to normalize.  Like many seasons.  It is incredibly noisy in 1 game (and even 10 game) samples.  The advanced stats that leverage +/- do a much better job of handling smaller sample data.  The really good ones start with box score stats and then apply +/- logic looking at both teammates and opponents.  I found a really good article which goes over which are considered the best in the industry (you will notice that PER comes in last, which validates both your and my thoughts on that particular statistic).

What is the best advanced statistic for basketball? NBA executives weigh in | HoopsHype

Thanks for this. A really interesting read.

[TXBamanut]

I like(d) this post but Alex Caruso is better than Brodie ToxicTripleDouble. So, I'm kinda torn.

It's pretty simple.  What is your definition of "better"?

If it's they play one on one to the death, it's Westbrook all day.

If it's score 25 a night, dominate the ball/do everything on a team with inferior talent, sell tickets and be a possible play in flame out, it's Westbrook all day.

If it's fill in around superior talent and perform a supporting role at a high level on a championship team, it's easily Caruso.

Better is all about what you are asking them to do.  That's where people fail in the better argument.  They assume option 1 and 2 automatically means you are better for the team.  That's not true.

[burekemde]

What +/- tells us is that we should send Luka to the G-league and start Frank. Would anyone dare to follow that based on this stat?

I just learned about the Raptor algorithm - It tells us that Luka is the 165th best player in the NBA.

My suggestion is that they really need to tune that algorithm. They are missing a lot of variables in their equation.

Let's look at it this way. If we don't have Luka, I don't believe we are are now 7-3. We are more like 4-6 at best IMO. Also, we would not have made the playoffs on regular basis since he entered the league and changed that culture at this team.

Basically, it is similar to what Skip is doing. He is finding some not fully representative single stat that highlights Luka is bad, and using that against him. It is all correct, that he needs to improve in some areas. But please, look at the 100 other stats that tell different story, and watch the games please.

There is no doubt in my mind that he leads by example, and makes the toughest situations during the game for his team.

[Kammrath]

What +/- tells us is that we should send Luka to the G-league and start Frank. Would anyone dare to follow that based on this stat?

That. is. not. even. remotely. what. the. stat. is. telling. us.

[BackToSquareOne]

It's pretty simple.  What is your definition of "better"?

If it's they play one on one to the death, it's Westbrook all day.

If it's score 25 a night, dominate the ball/do everything on a team with inferior talent, sell tickets and be a possible play in flame out, it's Westbrook all day.

If it's fill in around superior talent and perform a supporting role at a high level on a championship team, it's easily Caruso.

Better is all about what you are asking them to do.  That's where people fail in the better argument.  They assume option 1 and 2 automatically means you are better for the team.  That's not true.

It seems that we 100% agree.  :thumbsup:  Caruso is a better option for any team. (No team needs 1 or 2 above, tickets just aren't at that premium - cf the Wizards last year.)

[TXBamanut]

It seems that we 100% agree.  :thumbsup:  Caruso is a better option for any team. (No team needs 1 or 2 above, tickets just aren't at that premium - cf the Wizards last year.)

My post wasn't an exhaustive list of possibilities/roles, but just an example of what I was talking about.  So no, that doesn't work for every team.

[BackToSquareOne]

My post wasn't an exhaustive list of possibilities/roles, but just an example of what I was talking about.  So no, that doesn't work for every team.

Ok. That's fine. For whom would Brodie be 'better' (?) - and agreed that varies a lot by team.

Tbh, I thought we were close to agreeing. If we aren't then that's obviously fine - why can't reasonable people have different views? Nevertheless, I would be surprised.

[Jommybone]

I like(d) this post but Alex Caruso is better than Brodie ToxicTripleDouble. So, I'm kinda torn.

Funny. I like your post too. But I don’t discount RW’s bb ability because of his toxicity. In my mind at least, that’s a coaching failure. Not a lack of ability. 

For example, Ben Simmons isn’t a worse player, again in my mind, as a result of his being basically on strike. He’s a very good player feigning injury and illness because he’s on strike, not a worthless player who is contributing less than even Alex f’n Caruso. But I can understand why some would disagree.

[Dahlsim]

+/- is a real number of course, but interpreting it, is where you can go wrong. You can also interpret it correctly and then it's valuable. How to interpret it correctly? One way is to watch the plays and let them describe the reasons behind the +/-. Another one is to use multitude of other stats and combine those together with the +/-, in order to interpret a players performance and contribution. But also this in the end involves watching the plays and judging if the combined statistical interpretation is correct.

Bottom line is this, each stat or combination of stats needs to be verified by watching.

There is no way going around by watching and evaluation. But stat can help.

Maybe on +/- the statement could be applied Trust but verify.    

Generally speaking I treat as a great summary statistic to quickly get a bottom line result from what happened when certain player and certain combination of players were on the floor.  Trust that as being meaningful to some degree because the score is the bottom line so to speak. 

In terms of drawing a lot detailed conclusions, verify by watching or as Kidd says put the eye test to the game and what happening. 

@"DanSchwartzgan" cited the -6 that came from the minutes with double centers.  I didn't know the +/- but just from the eye test it was clear the minutes didn't go well but there was also some clear indication at least one reason to why it wouldn't go well.   

NO really made a concerted effort to push the ball against the super sized lineups, particularly off of misses and beat the slower centers down the floor for great looks without a set defense. 

Dallas on the other hand didn't consistently leverage the big guys to get high % shots and be in rebounding position against the smaller lineup.  When they did use the size advantage it generally went well by my eyes but mostly they seemed to play off the same offensive schemes.  

Without knowing the +/- watching the game it was clear it wasn't working and to some degree it was pretty clear why.  So if you've got a guy on floor, any guy, in what capacity is he being used?  

In the case of the big guys for example if you're going to supersize then you better take advantage of the size shove the high % looks down their throat because your opponent will certainly try and run it down your throat.  Its kill or be killed. 

[Dahlsim]

Funny. I like your post too. But I don’t discount RW’s bb ability because of his toxicity. In my mind at least, that’s a coaching failure. Not a lack of ability. 

For example, Ben Simmons isn’t a worse player, again in my mind, as a result of his being basically on strike. He’s a very good player feigning injury and illness because he’s on strike, not a worthless player who is contributing less than even Alex f’n Caruso. But I can understand why some would disagree.

Simmons is as you said, clearly faking.  He said what he would do, not play in Philly, and that's what he did, except that he and his guys didn't quite calculate correctly how the $$$ suspension thing would play out.   

They made a slight chiropractic adjustment, moving his injury from back to his head and presto, his plan is back on track.    Yeah, I wouldn't be surprised if he surfaces with some monster numbers with whoever he ends up playing for next.  He'll have a chip on his shoulder.

[burekemde]

Maybe on +/- the statement could be applied Trust but verify.    

Generally speaking I treat as a great summary statistic to quickly get a bottom line result from what happened when certain player and certain combination of players were on the floor.  Trust that as being meaningful to some degree because the score is the bottom line so to speak. 

In terms of drawing a lot detailed conclusions, verify by watching or as Kidd says put the eye test to the game and what happening. 

@"DanSchwartzgan" cited the -6 that came from the minutes with double centers.  I didn't know the +/- but just from the eye test it was clear the minutes didn't go well but there was also some clear indication at least one reason to why it wouldn't go well.   

NO really made a concerted effort to push the ball against the super sized lineups, particularly off of misses and beat the slower centers down the floor for great looks without a set defense. 

Dallas on the other hand didn't consistently leverage the big guys to get high % shots and be in rebounding position against the smaller lineup.  When they did use the size advantage it generally went well by my eyes but mostly they seemed to play off the same offensive schemes.  

Without knowing the +/- watching the game it was clear it wasn't working and to some degree it was pretty clear why.  So if you've got a guy on floor, any guy, in what capacity is he being used?  

In the case of the big guys for example if you're going to supersize then you better take advantage of the size shove the high % looks down their throat because your opponent will certainly try and run it down your throat.  Its kill or be killed. 

It's fantastic with eye analysis that explains the stats, thanks for these thoughts Dahlsim. I really like this! There is always a gap in the knowledge between what we see in games, and that raw number on the statsheet. Making that link, closing that gap, is where the best and most difficult analysis is required.

[mavsluvr]

Maybe on +/- the statement could be applied Trust but verify.    

Generally speaking I treat as a great summary statistic to quickly get a bottom line result from what happened when certain player and certain combination of players were on the floor.  Trust that as being meaningful to some degree because the score is the bottom line so to speak. 

In terms of drawing a lot detailed conclusions, verify by watching or as Kidd says put the eye test to the game and what happening. 

@"DanSchwartzgan" cited the -6 that came from the minutes with double centers.  I didn't know the +/- but just from the eye test it was clear the minutes didn't go well but there was also some clear indication at least one reason to why it wouldn't go well.   

NO really made a concerted effort to push the ball against the super sized lineups, particularly off of misses and beat the slower centers down the floor for great looks without a set defense. 

Dallas on the other hand didn't consistently leverage the big guys to get high % shots and be in rebounding position against the smaller lineup.  When they did use the size advantage it generally went well by my eyes but mostly they seemed to play off the same offensive schemes.  

Without knowing the +/- watching the game it was clear it wasn't working and to some degree it was pretty clear why.  So if you've got a guy on floor, any guy, in what capacity is he being used?  

In the case of the big guys for example if you're going to supersize then you better take advantage of the size shove the high % looks down their throat because your opponent will certainly try and run it down your throat.  Its kill or be killed. 

This is actually a legitimate use of short-term +/-. The team can use it to identify runs and use the information to look at video to analyze what went wrong (or right). They can use it for particular lineups in the sense of identifying what happened on the floor with certain lineups, and over time, determine whether there are issues when certain combinations of players are on the floor. 

They can examine, for example, a couple of possessions where the team lost points. Maybe they determine that a certain player made mistakes, and they can use that to try to help him do better. Or, maybe they discover that the problem consisted of a bad call, followed by an opponent hitting a well-defended, extremely low-probability shot, and determine that the players did what they were supposed to do, and no action is required.

The important thing here is that they would use short-term +/- as a tool for identifying what areas they want to take a closer look at. Certainly not to use as a stand-alone number labelling without further examination what players are "winners" and "losers" in terms of the season or their careers.

[fifteenth]

The important thing here is that they would use short-term +/- as a tool for identifying what areas they want to take a closer look at. 

That sounds good to me!

Certainly not to use as a stand-alone number labelling without further examination what players are "winners" and "losers" in terms of the season or their careers.

I agree. Just curious. Do you see people saying something like that here?

[mavsluvr]

I agree. Just curious. Do you see people saying something like that here?

I'm just referring to the Mavs Twitter-verse, lol.

[fifteenth]

I'm just referring to the Mavs Twitter-verse, lol.

I'd like to come up with a mathematical model that predicts the end of civilization as we know it based on when how close real life society in general mirrors the Twitter-verse