Professional and College Basketball Forums banner

1 - 12 of 12 Posts

·
Volleyball Coach
Joined
·
24,902 Posts
Discussion Starter #1
As always, they are doing some great work at 82games.com and they have added even more this season.

Link What I found very interesting in this group of stats is that RJ & Vince's PER (which is purely a measurement of offense) are almost equal at 22.1 and 22.5 respectively. However, their opponets PER is quite different. RJ's opponets have a PER of 16.4 while Vince's have a PER of 13.5. This would make it appear that Vince is playing better defense than RJ; however, the piece of information we don't have readily available is the season PER of the players each was guarding. I'm very curious about what the delta is between the opponets season PERs and their average PER against the Nets.

Padgett's opponets have the highest (on the Nets) PER at 25.1 and McInnis' opponets have the second highest at 22.5. Surprisingly, Zoran's opponets PER is only 13.8

I also took a quick look at Vince's and RJ's individual On/Off court stats. I found the effective FG% allowed stat very interesting. Not only does Vince allow a lower eFG% than RJ, when Vince is off the court, the Nets allow a higher eFG% and when RJ is off the court, the Nets allow a lower eFG%.
 

·
Registered
Joined
·
7,119 Posts
cpawfan said:
As always, they are doing some great work at 82games.com and they have added even more this season.

Link What I found very interesting in this group of stats is that RJ & Vince's PER (which is purely a measurement of offense) are almost equal at 22.1 and 22.5 respectively. However, their opponets PER is quite different. RJ's opponets have a PER of 16.4 while Vince's have a PER of 13.5. This would make it appear that Vince is playing better defense than RJ; however, the piece of information we don't have readily available is the season PER of the players each was guarding. I'm very curious about what the delta is between the opponets season PERs and their average PER against the Nets.

Padgett's opponets have the highest (on the Nets) PER at 25.1 and McInnis' opponets have the second highest at 22.5. Surprisingly, Zoran's opponets PER is only 13.8

I also took a quick look at Vince's and RJ's individual On/Off court stats. I found the effective FG% allowed stat very interesting. Not only does Vince allow a lower eFG% than RJ, when Vince is off the court, the Nets allow a higher eFG% and when RJ is off the court, the Nets allow a lower eFG%.
I prefer looking at the 5-man units, to see the relative value of individual players. The problem with the last figure you mention is that it is almost always RJ--not Vince--who plays with four bench players. It's not surprising that the eFG% is lower with him on the court. So, for instance, I'd compare the eFG% for a unit of McInnis, Zoran, RJ, Cliff, and Jackson to a unit of McInnis, Zoran, Vince, Cliff, and Jackson to have a better picture. As the season goes on, more combinations will log statistically significant minutes.

Anyway, I'm not sure I like the PER, for no other reason than they won't explain how it is calculated.

oh, one more thing--the "new" Roland ratings adds together these two components, and I am not sure that is statistically valid. It seems like you end up counting the contribution of your immediate opponenent twice. I need to give that more thought.
 

·
Registered
Joined
·
13,166 Posts
cpawfan said:
As always, they are doing some great work at 82games.com and they have added even more this season.

Link What I found very interesting in this group of stats is that RJ & Vince's PER (which is purely a measurement of offense) are almost equal at 22.1 and 22.5 respectively. However, their opponets PER is quite different. RJ's opponets have a PER of 16.4 while Vince's have a PER of 13.5. This would make it appear that Vince is playing better defense than RJ; however, the piece of information we don't have readily available is the season PER of the players each was guarding. I'm very curious about what the delta is between the opponets season PERs and their average PER against the Nets.

Padgett's opponets have the highest (on the Nets) PER at 25.1 and McInnis' opponets have the second highest at 22.5. Surprisingly, Zoran's opponets PER is only 13.8

I also took a quick look at Vince's and RJ's individual On/Off court stats. I found the effective FG% allowed stat very interesting. Not only does Vince allow a lower eFG% than RJ, when Vince is off the court, the Nets allow a higher eFG% and when RJ is off the court, the Nets allow a lower eFG%.
The bench defense issue remains a problem as the New York Sun piece noted yesterday: Planinic is the only one whose defense Frank has praised recently. Padgett, McInnis and Jackson are not getting it done. With Vaughn and Murray exiled to the end of the bench, I would like to see if either Johnson or Wright can do anything.
 

·
Registered
Joined
·
1,669 Posts
last night was the first nets game i watched on tv and u could clearly see VC was playing the team game not trying to take over!!!
 

·
Volleyball Coach
Joined
·
24,902 Posts
Discussion Starter #5
Dumpy said:
I prefer looking at the 5-man units, to see the relative value of individual players. The problem with the last figure you mention is that it is almost always RJ--not Vince--who plays with four bench players. It's not surprising that the eFG% is lower with him on the court. So, for instance, I'd compare the eFG% for a unit of McInnis, Zoran, RJ, Cliff, and Jackson to a unit of McInnis, Zoran, Vince, Cliff, and Jackson to have a better picture. As the season goes on, more combinations will log statistically significant minutes.

Anyway, I'm not sure I like the PER, for no other reason than they won't explain how it is calculated.

oh, one more thing--the "new" Roland ratings adds together these two components, and I am not sure that is statistically valid. It seems like you end up counting the contribution of your immediate opponenent twice. I need to give that more thought.
PER is a very good number for looking at offensive production. It isn't perfect, but it is a very good tool. If you would like to have a good discussion on the merits of PER, you might want to venture to the Statistical Analysis subforum of NBA General.
 

·
Premium Member
Joined
·
69,618 Posts
Interesting, interesting interesting. But then again it shouldnt be that suprising just take a look at the first two or 3 games
 

·
Burned
Joined
·
18,772 Posts
Dumpy said:
I prefer looking at the 5-man units, to see the relative value of individual players. The problem with the last figure you mention is that it is almost always RJ--not Vince--who plays with four bench players.
That's the explanation. You saved me the trouble of posting it.
 

·
Registered
Joined
·
620 Posts
cpawfan said:
As always, they are doing some great work at 82games.com and they have added even more this season.

Link What I found very interesting in this group of stats is that RJ & Vince's PER (which is purely a measurement of offense) are almost equal at 22.1 and 22.5 respectively. However, their opponets PER is quite different. RJ's opponets have a PER of 16.4 while Vince's have a PER of 13.5. This would make it appear that Vince is playing better defense than RJ; however, the piece of information we don't have readily available is the season PER of the players each was guarding. I'm very curious about what the delta is between the opponets season PERs and their average PER against the Nets.

Padgett's opponets have the highest (on the Nets) PER at 25.1 and McInnis' opponets have the second highest at 22.5. Surprisingly, Zoran's opponets PER is only 13.8

I also took a quick look at Vince's and RJ's individual On/Off court stats. I found the effective FG% allowed stat very interesting. Not only does Vince allow a lower eFG% than RJ, when Vince is off the court, the Nets allow a higher eFG% and when RJ is off the court, the Nets allow a lower eFG%.
Maybe vince's defense is party because RJ has been guarding better offensive players.
 

·
Registered
Joined
·
16,817 Posts
cpawfan said:
Link What I found very interesting in this group of stats is that RJ & Vince's PER (which is purely a measurement of offense) are almost equal at 22.1 and 22.5 respectively. However, their opponets PER is quite different. RJ's opponets have a PER of 16.4 while Vince's have a PER of 13.5. This would make it appear that Vince is playing better defense than RJ; however, the piece of information we don't have readily available is the season PER of the players each was guarding. I'm very curious about what the delta is between the opponets season PERs and their average PER against the Nets.
Not exactly. It accounts for defensive statistics (blocks and steals), but that's about it.[/QUOTE]
 

·
Registered
Joined
·
16,817 Posts
Dumpy said:
Anyway, I'm not sure I like the PER, for no other reason than they won't explain how it is calculated.
You can calculate PER, but it's rather complicated.

oh, one more thing--the "new" Roland ratings adds together these two components, and I am not sure that is statistically valid. It seems like you end up counting the contribution of your immediate opponenent twice. I need to give that more thought.
Well, it depends. If you are all right with a statistic accounting for an indivdual's impact against another individual and for an individual's impact on the game, it is all right. They do offer each statistic seperately, though, if you do not wish to rely on Roland Ratings.

Discuss it here!

 

·
Registered
Joined
·
7,119 Posts
Premier said:
You can calculate PER, but it's rather complicated.


Well, it depends. If you are all right with a statistic accounting for an indivdual's impact against another individual and for an individual's impact on the game, it is all right. They do offer each statistic seperately, though, if you do not wish to rely on Roland Ratings.

Discuss it here!

Thanks. I was misled when the site just said that the formula was developed by some guy, but didn't describe it or include a link to it. I didn't realise that it was a well-known formula
 

·
Volleyball Coach
Joined
·
24,902 Posts
Discussion Starter #12
Premier said:
You can calculate PER, but it's rather complicated.


Well, it depends. If you are all right with a statistic accounting for an indivdual's impact against another individual and for an individual's impact on the game, it is all right. They do offer each statistic seperately, though, if you do not wish to rely on Roland Ratings.

Discuss it here!

Thanks for the correction and the information.

I knew there was a reason I asked you to mod that forum ;)
 
1 - 12 of 12 Posts
Top