The CardBoard
WBB: Stanford 67 Oregon State 60 - Printable Version

+- The CardBoard (https://thecardboard.org/board)
+-- Forum: C-House! (https://thecardboard.org/board/forum-4.html)
+--- Forum: The CardBoard (https://thecardboard.org/board/forum-5.html)
+--- Thread: WBB: Stanford 67 Oregon State 60 (/thread-2753.html)

Pages: 1 2 3 4 5 6

Re: WBB: Stanford 67 Oregon State 60 - M T - 01-11-2012

(01-08-2012, 11:55 PM)Viking_Guy link Wrote:MT--
I was actually wondering about percentage of available boards, having run into that statistic before.  Do you any comparisons from earlier this year?  I think it does point out that the first and second half were qualitatively different, even if Stanford shot equally poorly (actually, IIRC, identically poorly) in each half.

In the past couple of years, I've kept a spread sheet with the games' statistics in it, including this.  I've been busy elsewhere this year so I hadn't done it until now.

Stanford's OR% and DR% for the previous games this year are below.
OR% = OR/(OR+ Opponent DR)  DR% = DR/(DR + Opponent OR)
(The opponent's OR% = 1 - Stanford DR%) 
I consider that we lost the rebounding battle if OR% + DR% < 100.  We lost against UConn.

We are clearly doing quite well this year, as you'd expect with our talent.  I didn't check, but I'd expect the numbers to come back down when we get to tournament time.

                         
Opp OR% DR%
Texas 49% 61%
Gonzaga 35% 80%
ODU 58% 63%
UConn 30% 65%
Xavier 53% 76%
UC-Davis 52% 100%
Fresno 64% 76%
Princeton 44% 60%
Tenn 44% 66%
CSUB 50% 70%
USC 36% 67%
UCLA 41% 82%
Oregon 38% 69%
OSU 51% 79%
2011-2012 Season 46% 72%
2010-2011 42%70%
2009-2010 41%71%
2008-2009 44%72%
2007-2008 37%71%
2006-2007 37%71%
2005-2006 35%68%
2004-2005 38%68%
2003-2004 36%67%
2002-2003 38%71%
2001-2002 40%68%
2000-2001 39%64%
1999-2000 37%64%
1998-1999 39%64%





Re: WBB: Stanford 67 Oregon State 60 - M T - 01-11-2012

(01-08-2012, 11:55 PM)Viking_Guy link Wrote:MT--
I was actually wondering about percentage of available boards, having run into that statistic before.  Do you any comparisons from earlier this year?  I think it does point out that the first and second half were qualitatively different, even if Stanford shot equally poorly (actually, IIRC, identically poorly) in each half.

In the past couple of years, I've kept a spread sheet with the games' statistics in it, including this.  I've been busy elsewhere this year so I hadn't done it until now.

Stanford's OR% and DR% for the previous games this year are below.
OR% = OR/(OR+ Opponent DR)  DR% = DR/(DR + Opponent OR)
(The opponent's OR% = 1 - Stanford DR%) 
I consider that we lost the rebounding battle if OR% + DR% < 100.  We lost against UConn.

We are clearly doing quite well this year, as you'd expect with our talent.  I didn't check, but I'd expect the numbers to come back down when we get to tournament time.

                         
Opp OR% DR%
Texas 49% 61%
Gonzaga 35% 80%
ODU 58% 63%
UConn 30% 65%
Xavier 53% 76%
UC-Davis 52% 100%
Fresno 64% 76%
Princeton 44% 60%
Tenn 44% 66%
CSUB 50% 70%
USC 36% 67%
UCLA 41% 82%
Oregon 38% 69%
OSU 51% 79%
2011-2012 Season 46% 72%
2010-2011 42%70%
2009-2010 41%71%
2008-2009 44%72%
2007-2008 37%71%
2006-2007 37%71%
2005-2006 35%68%
2004-2005 38%68%
2003-2004 36%67%
2002-2003 38%71%
2001-2002 40%68%
2000-2001 39%64%
1999-2000 37%64%
1998-1999 39%64%





Re: WBB: Stanford 67 Oregon State 60 - M T - 01-11-2012

(01-08-2012, 11:55 PM)Viking_Guy link Wrote:MT--
I was actually wondering about percentage of available boards, having run into that statistic before.  Do you any comparisons from earlier this year?  I think it does point out that the first and second half were qualitatively different, even if Stanford shot equally poorly (actually, IIRC, identically poorly) in each half.

In the past couple of years, I've kept a spread sheet with the games' statistics in it, including this.  I've been busy elsewhere this year so I hadn't done it until now.

Stanford's OR% and DR% for the previous games this year are below.
OR% = OR/(OR+ Opponent DR)  DR% = DR/(DR + Opponent OR)
(The opponent's OR% = 1 - Stanford DR%) 
I consider that we lost the rebounding battle if OR% + DR% < 100.  We lost against UConn.

We are clearly doing quite well this year, as you'd expect with our talent.  I didn't check, but I'd expect the numbers to come back down when we get to tournament time.

                         
Opp OR% DR%
Texas 49% 61%
Gonzaga 35% 80%
ODU 58% 63%
UConn 30% 65%
Xavier 53% 76%
UC-Davis 52% 100%
Fresno 64% 76%
Princeton 44% 60%
Tenn 44% 66%
CSUB 50% 70%
USC 36% 67%
UCLA 41% 82%
Oregon 38% 69%
OSU 51% 79%
2011-2012 Season 46% 72%
2010-2011 42%70%
2009-2010 41%71%
2008-2009 44%72%
2007-2008 37%71%
2006-2007 37%71%
2005-2006 35%68%
2004-2005 38%68%
2003-2004 36%67%
2002-2003 38%71%
2001-2002 40%68%
2000-2001 39%64%
1999-2000 37%64%
1998-1999 39%64%





Re: WBB: Stanford 67 Oregon State 60 - M T - 01-11-2012

(01-08-2012, 11:55 PM)Viking_Guy link Wrote:MT--
I was actually wondering about percentage of available boards, having run into that statistic before.  Do you any comparisons from earlier this year?  I think it does point out that the first and second half were qualitatively different, even if Stanford shot equally poorly (actually, IIRC, identically poorly) in each half.

In the past couple of years, I've kept a spread sheet with the games' statistics in it, including this.  I've been busy elsewhere this year so I hadn't done it until now.

Stanford's OR% and DR% for the previous games this year are below.
OR% = OR/(OR+ Opponent DR)  DR% = DR/(DR + Opponent OR)
(The opponent's OR% = 1 - Stanford DR%) 
I consider that we lost the rebounding battle if OR% + DR% < 100.  We lost against UConn.

We are clearly doing quite well this year, as you'd expect with our talent.  I didn't check, but I'd expect the numbers to come back down when we get to tournament time.

                         
Opp OR% DR%
Texas 49% 61%
Gonzaga 35% 80%
ODU 58% 63%
UConn 30% 65%
Xavier 53% 76%
UC-Davis 52% 100%
Fresno 64% 76%
Princeton 44% 60%
Tenn 44% 66%
CSUB 50% 70%
USC 36% 67%
UCLA 41% 82%
Oregon 38% 69%
OSU 51% 79%
2011-2012 Season 46% 72%
2010-2011 42%70%
2009-2010 41%71%
2008-2009 44%72%
2007-2008 37%71%
2006-2007 37%71%
2005-2006 35%68%
2004-2005 38%68%
2003-2004 36%67%
2002-2003 38%71%
2001-2002 40%68%
2000-2001 39%64%
1999-2000 37%64%
1998-1999 39%64%





Another stat I find interesting - M T - 01-11-2012

Another statistic that I find interesting is what I call effectiveness.  For each player, it is simply a measure of how much more your team scores than the other team while the player is in the game, normalized to a full 40-minutes.  If you play exactly one half and your team gained 3 points on the other team during that time, your effectiveness would be 6.

I think the WNBA once had something similar to this, but it either was dropped or changed.  I started looking at this when Nicole was first in the WNBA.

Most BB statistics highlight one particular aspect of play (naturally).  But how do you compare the effectiveness of a great shooter versus a great defender.  We've all seen players whose contributions don't get well reflected in the statistics.  I believe this stat combines everything you do.  If you make a pass to a covered player, which leads to a turnover, this picks it up.  If your spacing is bad, which makes someone else not able to get the ball to you, the effectiveness picks it up.  If you hound their star player so they never get the ball, it shows up.

One thing I like about this is that it is normalized to a full 40 minutes of play.  For instance, Nneka would have more points if she played a full 40 minutes in each game.  But, if she sits down with 8 points after only 13 minutes in a blowout game, her average points per game is hurt, as compared to someone who plays almost every minute of every game.

The problem with this stat, as with most (except maybe FT%), is that it depends on both the quality of your opponents and the quality of your teammates.  When our second stringers are in against their second stringers, our best player in our second stringers would have an effectiveness stat like our best player in our first team against their first team.  (That's not really true as the two second-string teams might only score 40 points against each other in a full game, but you get the idea.)  The stat would work best if all teams were more closely matched, which is not the case in WBB.

If I select for just the games that were at least somewhat challenging (Texas, Gonzaga, UConn, Xavier, Tenn, the University of South Central, OSU), here are the effectiveness ( = NetPts/40min) numbers for the Stanford players:


                                                                                                           
Player   MinNetPtsEffectiveness
Boothe, Sarah       
70:32
-15
-8.51
Camp, Jasmine     
38:47
-12
-12.38
Green, Alex           
2:48
-2
-28.57
Greenfield, Taylor 
162:37
50
12.30
James, Sara         
22:39
-5
-8.83
Kokenis, Toni     
243:08
100
16.45
La Rocque, Lindy   
200:54
39
  7.77
Mashore, Grace     
  1:09
-2
-69.57
Ogwumike, Chiney 
194:35
78
16.03
Ogwumike, Nnemkadi
199:46
86
17.22
Orrange, Amber   
71:39
  5
  2.79
Payne, Erica       
14:31
-12
-33.07
Ruef, Mikaela       
11:23
  7
24.60
Samuelson, Bonnie 
45:55
15
13.07
Tinkle, Joslyn     
95:52
-11
-4.59


Another stat I find interesting - M T - 01-11-2012

Another statistic that I find interesting is what I call effectiveness.  For each player, it is simply a measure of how much more your team scores than the other team while the player is in the game, normalized to a full 40-minutes.  If you play exactly one half and your team gained 3 points on the other team during that time, your effectiveness would be 6.

I think the WNBA once had something similar to this, but it either was dropped or changed.  I started looking at this when Nicole was first in the WNBA.

Most BB statistics highlight one particular aspect of play (naturally).  But how do you compare the effectiveness of a great shooter versus a great defender.  We've all seen players whose contributions don't get well reflected in the statistics.  I believe this stat combines everything you do.  If you make a pass to a covered player, which leads to a turnover, this picks it up.  If your spacing is bad, which makes someone else not able to get the ball to you, the effectiveness picks it up.  If you hound their star player so they never get the ball, it shows up.

One thing I like about this is that it is normalized to a full 40 minutes of play.  For instance, Nneka would have more points if she played a full 40 minutes in each game.  But, if she sits down with 8 points after only 13 minutes in a blowout game, her average points per game is hurt, as compared to someone who plays almost every minute of every game.

The problem with this stat, as with most (except maybe FT%), is that it depends on both the quality of your opponents and the quality of your teammates.  When our second stringers are in against their second stringers, our best player in our second stringers would have an effectiveness stat like our best player in our first team against their first team.  (That's not really true as the two second-string teams might only score 40 points against each other in a full game, but you get the idea.)  The stat would work best if all teams were more closely matched, which is not the case in WBB.

If I select for just the games that were at least somewhat challenging (Texas, Gonzaga, UConn, Xavier, Tenn, the University of South Central, OSU), here are the effectiveness ( = NetPts/40min) numbers for the Stanford players:


                                                                                                           
Player   MinNetPtsEffectiveness
Boothe, Sarah       
70:32
-15
-8.51
Camp, Jasmine     
38:47
-12
-12.38
Green, Alex           
2:48
-2
-28.57
Greenfield, Taylor 
162:37
50
12.30
James, Sara         
22:39
-5
-8.83
Kokenis, Toni     
243:08
100
16.45
La Rocque, Lindy   
200:54
39
  7.77
Mashore, Grace     
  1:09
-2
-69.57
Ogwumike, Chiney 
194:35
78
16.03
Ogwumike, Nnemkadi
199:46
86
17.22
Orrange, Amber   
71:39
  5
  2.79
Payne, Erica       
14:31
-12
-33.07
Ruef, Mikaela       
11:23
  7
24.60
Samuelson, Bonnie 
45:55
15
13.07
Tinkle, Joslyn     
95:52
-11
-4.59


Another stat I find interesting - M T - 01-11-2012

Another statistic that I find interesting is what I call effectiveness.  For each player, it is simply a measure of how much more your team scores than the other team while the player is in the game, normalized to a full 40-minutes.  If you play exactly one half and your team gained 3 points on the other team during that time, your effectiveness would be 6.

I think the WNBA once had something similar to this, but it either was dropped or changed.  I started looking at this when Nicole was first in the WNBA.

Most BB statistics highlight one particular aspect of play (naturally).  But how do you compare the effectiveness of a great shooter versus a great defender.  We've all seen players whose contributions don't get well reflected in the statistics.  I believe this stat combines everything you do.  If you make a pass to a covered player, which leads to a turnover, this picks it up.  If your spacing is bad, which makes someone else not able to get the ball to you, the effectiveness picks it up.  If you hound their star player so they never get the ball, it shows up.

One thing I like about this is that it is normalized to a full 40 minutes of play.  For instance, Nneka would have more points if she played a full 40 minutes in each game.  But, if she sits down with 8 points after only 13 minutes in a blowout game, her average points per game is hurt, as compared to someone who plays almost every minute of every game.

The problem with this stat, as with most (except maybe FT%), is that it depends on both the quality of your opponents and the quality of your teammates.  When our second stringers are in against their second stringers, our best player in our second stringers would have an effectiveness stat like our best player in our first team against their first team.  (That's not really true as the two second-string teams might only score 40 points against each other in a full game, but you get the idea.)  The stat would work best if all teams were more closely matched, which is not the case in WBB.

If I select for just the games that were at least somewhat challenging (Texas, Gonzaga, UConn, Xavier, Tenn, the University of South Central, OSU), here are the effectiveness ( = NetPts/40min) numbers for the Stanford players:


                                                                                                           
Player   MinNetPtsEffectiveness
Boothe, Sarah       
70:32
-15
-8.51
Camp, Jasmine     
38:47
-12
-12.38
Green, Alex           
2:48
-2
-28.57
Greenfield, Taylor 
162:37
50
12.30
James, Sara         
22:39
-5
-8.83
Kokenis, Toni     
243:08
100
16.45
La Rocque, Lindy   
200:54
39
  7.77
Mashore, Grace     
  1:09
-2
-69.57
Ogwumike, Chiney 
194:35
78
16.03
Ogwumike, Nnemkadi
199:46
86
17.22
Orrange, Amber   
71:39
  5
  2.79
Payne, Erica       
14:31
-12
-33.07
Ruef, Mikaela       
11:23
  7
24.60
Samuelson, Bonnie 
45:55
15
13.07
Tinkle, Joslyn     
95:52
-11
-4.59


Another stat I find interesting - M T - 01-11-2012

Another statistic that I find interesting is what I call effectiveness.  For each player, it is simply a measure of how much more your team scores than the other team while the player is in the game, normalized to a full 40-minutes.  If you play exactly one half and your team gained 3 points on the other team during that time, your effectiveness would be 6.

I think the WNBA once had something similar to this, but it either was dropped or changed.  I started looking at this when Nicole was first in the WNBA.

Most BB statistics highlight one particular aspect of play (naturally).  But how do you compare the effectiveness of a great shooter versus a great defender.  We've all seen players whose contributions don't get well reflected in the statistics.  I believe this stat combines everything you do.  If you make a pass to a covered player, which leads to a turnover, this picks it up.  If your spacing is bad, which makes someone else not able to get the ball to you, the effectiveness picks it up.  If you hound their star player so they never get the ball, it shows up.

One thing I like about this is that it is normalized to a full 40 minutes of play.  For instance, Nneka would have more points if she played a full 40 minutes in each game.  But, if she sits down with 8 points after only 13 minutes in a blowout game, her average points per game is hurt, as compared to someone who plays almost every minute of every game.

The problem with this stat, as with most (except maybe FT%), is that it depends on both the quality of your opponents and the quality of your teammates.  When our second stringers are in against their second stringers, our best player in our second stringers would have an effectiveness stat like our best player in our first team against their first team.  (That's not really true as the two second-string teams might only score 40 points against each other in a full game, but you get the idea.)  The stat would work best if all teams were more closely matched, which is not the case in WBB.

If I select for just the games that were at least somewhat challenging (Texas, Gonzaga, UConn, Xavier, Tenn, the University of South Central, OSU), here are the effectiveness ( = NetPts/40min) numbers for the Stanford players:


                                                                                                           
Player   MinNetPtsEffectiveness
Boothe, Sarah       
70:32
-15
-8.51
Camp, Jasmine     
38:47
-12
-12.38
Green, Alex           
2:48
-2
-28.57
Greenfield, Taylor 
162:37
50
12.30
James, Sara         
22:39
-5
-8.83
Kokenis, Toni     
243:08
100
16.45
La Rocque, Lindy   
200:54
39
  7.77
Mashore, Grace     
  1:09
-2
-69.57
Ogwumike, Chiney 
194:35
78
16.03
Ogwumike, Nnemkadi
199:46
86
17.22
Orrange, Amber   
71:39
  5
  2.79
Payne, Erica       
14:31
-12
-33.07
Ruef, Mikaela       
11:23
  7
24.60
Samuelson, Bonnie 
45:55
15
13.07
Tinkle, Joslyn     
95:52
-11
-4.59


Re: WBB: Stanford 67 Oregon State 60 - CompSci87 - 01-11-2012

Interesting stats, thanks. I have a feeling the effectiveness stat is tough on our second stringers in general. I've noticed that often we have our second stringers in while our opponent still has a lot of first stringers in -- perhaps because the opponent has less depth, or perhaps because we're way ahead and feel comfortable putting in second stringers but our opponent is still fighting to make it respectable, trying to get their starters more experience, vainly hoping to pull out a win, or whatever.


Re: WBB: Stanford 67 Oregon State 60 - CompSci87 - 01-11-2012

Interesting stats, thanks. I have a feeling the effectiveness stat is tough on our second stringers in general. I've noticed that often we have our second stringers in while our opponent still has a lot of first stringers in -- perhaps because the opponent has less depth, or perhaps because we're way ahead and feel comfortable putting in second stringers but our opponent is still fighting to make it respectable, trying to get their starters more experience, vainly hoping to pull out a win, or whatever.


Re: WBB: Stanford 67 Oregon State 60 - CompSci87 - 01-11-2012

Interesting stats, thanks. I have a feeling the effectiveness stat is tough on our second stringers in general. I've noticed that often we have our second stringers in while our opponent still has a lot of first stringers in -- perhaps because the opponent has less depth, or perhaps because we're way ahead and feel comfortable putting in second stringers but our opponent is still fighting to make it respectable, trying to get their starters more experience, vainly hoping to pull out a win, or whatever.


Re: WBB: Stanford 67 Oregon State 60 - CompSci87 - 01-11-2012

Interesting stats, thanks. I have a feeling the effectiveness stat is tough on our second stringers in general. I've noticed that often we have our second stringers in while our opponent still has a lot of first stringers in -- perhaps because the opponent has less depth, or perhaps because we're way ahead and feel comfortable putting in second stringers but our opponent is still fighting to make it respectable, trying to get their starters more experience, vainly hoping to pull out a win, or whatever.


Re: WBB: Stanford 67 Oregon State 60 - washingtonismoney - 01-11-2012

(01-11-2012, 04:50 PM)garvin link Wrote:I'm sort of dubious about the effectiveness stat. Suppose you happen to put in your five minutes against the University of South Bend just as Skylar Diggins goes to the bench with four fouls.

Yeah, plus/minus--which is what it's sort of widely called, I guess--is a difficult stat to trust. There are effects from:
a) who you play against
b) who you play with
c) where you play (some players play much more at home than the road, or vice versa)

And so on. Various basketball statisticians throw various adjustments for stuff into it, but the consensus is that it takes several years for the noise to get filtered out and the signal to be heard. Of course, by that point, the data is not particularly helpful since your players have graduated, in this case.

Re: WBB: Stanford 67 Oregon State 60 - washingtonismoney - 01-11-2012

(01-11-2012, 04:50 PM)garvin link Wrote:I'm sort of dubious about the effectiveness stat. Suppose you happen to put in your five minutes against the University of South Bend just as Skylar Diggins goes to the bench with four fouls.

Yeah, plus/minus--which is what it's sort of widely called, I guess--is a difficult stat to trust. There are effects from:
a) who you play against
b) who you play with
c) where you play (some players play much more at home than the road, or vice versa)

And so on. Various basketball statisticians throw various adjustments for stuff into it, but the consensus is that it takes several years for the noise to get filtered out and the signal to be heard. Of course, by that point, the data is not particularly helpful since your players have graduated, in this case.

Re: WBB: Stanford 67 Oregon State 60 - washingtonismoney - 01-11-2012

(01-11-2012, 04:50 PM)garvin link Wrote:I'm sort of dubious about the effectiveness stat. Suppose you happen to put in your five minutes against the University of South Bend just as Skylar Diggins goes to the bench with four fouls.

Yeah, plus/minus--which is what it's sort of widely called, I guess--is a difficult stat to trust. There are effects from:
a) who you play against
b) who you play with
c) where you play (some players play much more at home than the road, or vice versa)

And so on. Various basketball statisticians throw various adjustments for stuff into it, but the consensus is that it takes several years for the noise to get filtered out and the signal to be heard. Of course, by that point, the data is not particularly helpful since your players have graduated, in this case.

Re: WBB: Stanford 67 Oregon State 60 - washingtonismoney - 01-11-2012

(01-11-2012, 04:50 PM)garvin link Wrote:I'm sort of dubious about the effectiveness stat. Suppose you happen to put in your five minutes against the University of South Bend just as Skylar Diggins goes to the bench with four fouls.

Yeah, plus/minus--which is what it's sort of widely called, I guess--is a difficult stat to trust. There are effects from:
a) who you play against
b) who you play with
c) where you play (some players play much more at home than the road, or vice versa)

And so on. Various basketball statisticians throw various adjustments for stuff into it, but the consensus is that it takes several years for the noise to get filtered out and the signal to be heard. Of course, by that point, the data is not particularly helpful since your players have graduated, in this case.

Re: WBB: Stanford 67 Oregon State 60 - M T - 01-11-2012

(I thought Plus-Minus is used as an absolute measure, where the players that play the most have the biggest numbers.  To me it is critical that this be normalized over playing time.)

Exactly the same thing ("difficult stat to trust") can be said about rebounding, shooting percentage, number of steals, turnovers, etc.  If their best defenders are out, then your offensive stats go up. If your point guard teammate goes out, your offensive stats go down.  If their best playmakers are out, then your defensive stats go up.  Or, as the previous poster said:
Quote:There are effects from:
a) who you play against
b) who you play with
c) where you play (some players play much more at home than the road, or vice versa)
So, do we simply say that anything except FT% is a difficult stat to trust?  What would the broadcasters do without stats to spout (along with things like how often a team wins on the road in a state that begins with T on nights that begin with T)?  What would fans argue about?  ;) (Sorry for channeling Brent Musberger for a moment.)

I absolutely agree that it is difficult to compare two players on different teams that played different schedules for any of the stats beyond FT%.  But within the PAC-10's round-robin schedule, you could start to compare because of the common opponents.

I intentionally didn't analyze the numbers above in order to allow others a chance to comment first.  Notice that Toni has almost the same rating as Nneka.  I don't think any one thinks Toni is as talented a BB player as Nneka.  I think this reflects the fact that Toni is a necessary part of the effectiveness of Stanford.  When she's out, there's no one else on the team that can do what she does.  I knew Toni had been quite effective, but I didn't realize she was this effective.

If you look at individual stats, in the Pac-12, Toni ranks #27 in scoring, #8 in assists (but #1 in A/TO ratio), #15 in 3PT.  That's it.  That doesn't sound like a player that is almost as effective for Stanford as consensus All-American Nneka.  So, what is it?  Is it that she shuts down the other team's offense better, does she disrupt their defense more by her cuts, etc.?  I feel this number integrates over all those other things a BB player does that helps the outcome.

Here is a caveat though...  While Toni's rating is only affected by what happens when she's in the game, comparing her to others is affected by the capabilities of her replacement.  So, Toni gets a 16.45 just from when she's in the game.  Nneka gets a 17.22 over her time.  But maybe Nneka would get a 20 when Toni was in and only a 10 when Amber is the point guard.  In effect, Toni's seemingly higher-than-expected team ranking of her rating may be due to her filling some necessary role that others can't.

I guess the next thing I'd want to put together are effectiveness of lineups.  What is the effectiveness of Nneka, Chiney, Toni, Lindy, Taylor versus the same group with Amber instead of Toni?  What is the effectiveness of that first group versus Nneka, Chiney, Toni, Joslyn, Bonnie?  etc.

Re: WBB: Stanford 67 Oregon State 60 - M T - 01-11-2012

(I thought Plus-Minus is used as an absolute measure, where the players that play the most have the biggest numbers.  To me it is critical that this be normalized over playing time.)

Exactly the same thing ("difficult stat to trust") can be said about rebounding, shooting percentage, number of steals, turnovers, etc.  If their best defenders are out, then your offensive stats go up. If your point guard teammate goes out, your offensive stats go down.  If their best playmakers are out, then your defensive stats go up.  Or, as the previous poster said:
Quote:There are effects from:
a) who you play against
b) who you play with
c) where you play (some players play much more at home than the road, or vice versa)
So, do we simply say that anything except FT% is a difficult stat to trust?  What would the broadcasters do without stats to spout (along with things like how often a team wins on the road in a state that begins with T on nights that begin with T)?  What would fans argue about?  ;) (Sorry for channeling Brent Musberger for a moment.)

I absolutely agree that it is difficult to compare two players on different teams that played different schedules for any of the stats beyond FT%.  But within the PAC-10's round-robin schedule, you could start to compare because of the common opponents.

I intentionally didn't analyze the numbers above in order to allow others a chance to comment first.  Notice that Toni has almost the same rating as Nneka.  I don't think any one thinks Toni is as talented a BB player as Nneka.  I think this reflects the fact that Toni is a necessary part of the effectiveness of Stanford.  When she's out, there's no one else on the team that can do what she does.  I knew Toni had been quite effective, but I didn't realize she was this effective.

If you look at individual stats, in the Pac-12, Toni ranks #27 in scoring, #8 in assists (but #1 in A/TO ratio), #15 in 3PT.  That's it.  That doesn't sound like a player that is almost as effective for Stanford as consensus All-American Nneka.  So, what is it?  Is it that she shuts down the other team's offense better, does she disrupt their defense more by her cuts, etc.?  I feel this number integrates over all those other things a BB player does that helps the outcome.

Here is a caveat though...  While Toni's rating is only affected by what happens when she's in the game, comparing her to others is affected by the capabilities of her replacement.  So, Toni gets a 16.45 just from when she's in the game.  Nneka gets a 17.22 over her time.  But maybe Nneka would get a 20 when Toni was in and only a 10 when Amber is the point guard.  In effect, Toni's seemingly higher-than-expected team ranking of her rating may be due to her filling some necessary role that others can't.

I guess the next thing I'd want to put together are effectiveness of lineups.  What is the effectiveness of Nneka, Chiney, Toni, Lindy, Taylor versus the same group with Amber instead of Toni?  What is the effectiveness of that first group versus Nneka, Chiney, Toni, Joslyn, Bonnie?  etc.

Re: WBB: Stanford 67 Oregon State 60 - M T - 01-11-2012

(I thought Plus-Minus is used as an absolute measure, where the players that play the most have the biggest numbers.  To me it is critical that this be normalized over playing time.)

Exactly the same thing ("difficult stat to trust") can be said about rebounding, shooting percentage, number of steals, turnovers, etc.  If their best defenders are out, then your offensive stats go up. If your point guard teammate goes out, your offensive stats go down.  If their best playmakers are out, then your defensive stats go up.  Or, as the previous poster said:
Quote:There are effects from:
a) who you play against
b) who you play with
c) where you play (some players play much more at home than the road, or vice versa)
So, do we simply say that anything except FT% is a difficult stat to trust?  What would the broadcasters do without stats to spout (along with things like how often a team wins on the road in a state that begins with T on nights that begin with T)?  What would fans argue about?  ;) (Sorry for channeling Brent Musberger for a moment.)

I absolutely agree that it is difficult to compare two players on different teams that played different schedules for any of the stats beyond FT%.  But within the PAC-10's round-robin schedule, you could start to compare because of the common opponents.

I intentionally didn't analyze the numbers above in order to allow others a chance to comment first.  Notice that Toni has almost the same rating as Nneka.  I don't think any one thinks Toni is as talented a BB player as Nneka.  I think this reflects the fact that Toni is a necessary part of the effectiveness of Stanford.  When she's out, there's no one else on the team that can do what she does.  I knew Toni had been quite effective, but I didn't realize she was this effective.

If you look at individual stats, in the Pac-12, Toni ranks #27 in scoring, #8 in assists (but #1 in A/TO ratio), #15 in 3PT.  That's it.  That doesn't sound like a player that is almost as effective for Stanford as consensus All-American Nneka.  So, what is it?  Is it that she shuts down the other team's offense better, does she disrupt their defense more by her cuts, etc.?  I feel this number integrates over all those other things a BB player does that helps the outcome.

Here is a caveat though...  While Toni's rating is only affected by what happens when she's in the game, comparing her to others is affected by the capabilities of her replacement.  So, Toni gets a 16.45 just from when she's in the game.  Nneka gets a 17.22 over her time.  But maybe Nneka would get a 20 when Toni was in and only a 10 when Amber is the point guard.  In effect, Toni's seemingly higher-than-expected team ranking of her rating may be due to her filling some necessary role that others can't.

I guess the next thing I'd want to put together are effectiveness of lineups.  What is the effectiveness of Nneka, Chiney, Toni, Lindy, Taylor versus the same group with Amber instead of Toni?  What is the effectiveness of that first group versus Nneka, Chiney, Toni, Joslyn, Bonnie?  etc.

Re: WBB: Stanford 67 Oregon State 60 - M T - 01-11-2012

(I thought Plus-Minus is used as an absolute measure, where the players that play the most have the biggest numbers.  To me it is critical that this be normalized over playing time.)

Exactly the same thing ("difficult stat to trust") can be said about rebounding, shooting percentage, number of steals, turnovers, etc.  If their best defenders are out, then your offensive stats go up. If your point guard teammate goes out, your offensive stats go down.  If their best playmakers are out, then your defensive stats go up.  Or, as the previous poster said:
Quote:There are effects from:
a) who you play against
b) who you play with
c) where you play (some players play much more at home than the road, or vice versa)
So, do we simply say that anything except FT% is a difficult stat to trust?  What would the broadcasters do without stats to spout (along with things like how often a team wins on the road in a state that begins with T on nights that begin with T)?  What would fans argue about?  ;) (Sorry for channeling Brent Musberger for a moment.)

I absolutely agree that it is difficult to compare two players on different teams that played different schedules for any of the stats beyond FT%.  But within the PAC-10's round-robin schedule, you could start to compare because of the common opponents.

I intentionally didn't analyze the numbers above in order to allow others a chance to comment first.  Notice that Toni has almost the same rating as Nneka.  I don't think any one thinks Toni is as talented a BB player as Nneka.  I think this reflects the fact that Toni is a necessary part of the effectiveness of Stanford.  When she's out, there's no one else on the team that can do what she does.  I knew Toni had been quite effective, but I didn't realize she was this effective.

If you look at individual stats, in the Pac-12, Toni ranks #27 in scoring, #8 in assists (but #1 in A/TO ratio), #15 in 3PT.  That's it.  That doesn't sound like a player that is almost as effective for Stanford as consensus All-American Nneka.  So, what is it?  Is it that she shuts down the other team's offense better, does she disrupt their defense more by her cuts, etc.?  I feel this number integrates over all those other things a BB player does that helps the outcome.

Here is a caveat though...  While Toni's rating is only affected by what happens when she's in the game, comparing her to others is affected by the capabilities of her replacement.  So, Toni gets a 16.45 just from when she's in the game.  Nneka gets a 17.22 over her time.  But maybe Nneka would get a 20 when Toni was in and only a 10 when Amber is the point guard.  In effect, Toni's seemingly higher-than-expected team ranking of her rating may be due to her filling some necessary role that others can't.

I guess the next thing I'd want to put together are effectiveness of lineups.  What is the effectiveness of Nneka, Chiney, Toni, Lindy, Taylor versus the same group with Amber instead of Toni?  What is the effectiveness of that first group versus Nneka, Chiney, Toni, Joslyn, Bonnie?  etc.