This is Google's cache of viewtopic.php?t=183. It is a snapshot of the page as it appeared on Feb 12, 2011 19:38:21 GMT. The current page could have changed in the meantime. Learn more

Text-only version
These search terms are highlighted: ed küpfer  
APBRmetrics :: View topic - Unstoppable force vs. immovable object
APBRmetrics Forum Index APBRmetrics
The statistical revolution will not be televised.
 
 FAQFAQ   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups   RegisterRegister 
 ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in 

Unstoppable force vs. immovable object

 
Post new topic   Reply to topic    APBRmetrics Forum Index -> General discussion
View previous topic :: View next topic  
Author Message
mavs128



Joined: 25 Feb 2005
Posts: 32
Location: Dallas, Texas

PostPosted: Fri Apr 15, 2005 12:01 pm    Post subject: Unstoppable force vs. immovable object Reply with quote

Question: does anyone know of any studies done that show how a team who is dominant in one area is affected when they play a team that is dominant in the opposite area?

For instance, if the playoffs started today Seattle would be matched up against Houston. Seattle has the top OReb% in the league at 32.5%.....but on the flip side, Houston has the top DReb% in the league at 74.4%. I'm wondering if either sides' dominance is expected to come out more, or if it's expected to even off at 29% OReb for Seattle and 71% DReb for Houston, or what?
Back to top
View user's profile Send private message
jambalaya



Joined: 30 Jan 2005
Posts: 282

PostPosted: Fri Apr 15, 2005 12:51 pm    Post subject: rebounding matchup Reply with quote

i think it will be close on your DR/OR question.

at first i thought reggie evans will have a field day against the remaining houston power forwards. but he only got 7 and 8 total rebounds the last two times out against them. i would assume van grundy's emphasis on boxing out has some effect.

yao ming should get a good rate of defensive rebounds. whether fortson is a major factor in such a playoff matchup is a main unknown.

i would guess sura and mcgrady will do pretty well rebounding under our bucket.

how big collison is off the bench is the other key.

houston's starting five might win or keep it close in their matchup against the sonic starting five but the sonic bench should win theirs. in the playoffs the bench's role usually shrinks but maybe not for the sonics.
Back to top
View user's profile Send private message
Ed Küpfer



Joined: 30 Dec 2004
Posts: 783
Location: Toronto

PostPosted: Fri Apr 15, 2005 1:01 pm    Post subject: Re: Unstoppable force vs. immovable object Reply with quote

mavs128 wrote:
Question: does anyone know of any studies done that show how a team who is dominant in one area is affected when they play a team that is dominant in the opposite area?


I tried this once with ORTG vs DRTG, and didn't get very far. The amount of noise is enormous. I know DeanO did something similar with game pace. Other than that, I don't know of anything. It's a good idea though, especially on a smaller scale -- say, for REB% or shooting or turnovers. Maybe I'll try something over the weekend.
_________________
ed
Back to top
View user's profile Send private message Send e-mail
Ed Küpfer



Joined: 30 Dec 2004
Posts: 783
Location: Toronto

PostPosted: Mon Apr 18, 2005 6:31 pm    Post subject: Re: Unstoppable force vs. immovable object Reply with quote

Just fooling around, and I came up with this:

Team 1 excected EFG% = 1 / (1 + EXP(-(-3.1 + 3.5 * EFG_team1 + 2.9 * defEFG_team2)))

where EFG_team1 is Team 1's EFG% and defEFG_team2 is Team 2's defensive EFG%. The RMSE on that one is about 6%, meaning 2/3 of all games will be within 6% of the expected amount.

Team 1 expected TO% = 1 / (1 + EXP(-(-3.6 + 5.8 * TO%_team1 + 6.1 * defTO%_team2)))

where TO%_team1 is Team 1's TOs per possession, and defTO%_team2 is Team 2's defensive TOs per possession. The RMSE is about 4%.

Team 1 expected OR% = 1 / (1 + EXP(-(4.1*OR%_team1 - 3.1 * DR%_team2)))

The RMSE on this one isn't very good, a little less than 8%.
_________________
ed
Back to top
View user's profile Send private message Send e-mail
HoopStudies



Joined: 30 Dec 2004
Posts: 705
Location: Near Philadelphia, PA

PostPosted: Mon Apr 18, 2005 6:54 pm    Post subject: Re: Unstoppable force vs. immovable object Reply with quote

Ed Küpfer wrote:
Just fooling around, and I came up with this:

Team 1 excected EFG% = 1 / (1 + EXP(-(-3.1 + 3.5 * EFG_team1 + 2.9 * defEFG_team2)))

where EFG_team1 is Team 1's EFG% and defEFG_team2 is Team 2's defensive EFG%. The RMSE on that one is about 6%, meaning 2/3 of all games will be within 6% of the expected amount.

Team 1 expected TO% = 1 / (1 + EXP(-(-3.6 + 5.8 * TO%_team1 + 6.1 * defTO%_team2)))

where TO%_team1 is Team 1's TOs per possession, and defTO%_team2 is Team 2's defensive TOs per possession. The RMSE is about 4%.

Team 1 expected OR% = 1 / (1 + EXP(-(4.1*OR%_team1 - 3.1 * DR%_team2)))

The RMSE on this one isn't very good, a little less than 8%.


What is the significance of the coefficients? This generally seems to suggest that offense wins but are those differences significant?
_________________
Dean Oliver
Author, Basketball on Paper
The postings are my own & don't necess represent positions, strategies or opinions of employers.
Back to top
View user's profile Send private message Visit poster's website
Ed Küpfer



Joined: 30 Dec 2004
Posts: 783
Location: Toronto

PostPosted: Mon Apr 18, 2005 7:01 pm    Post subject: Re: Unstoppable force vs. immovable object Reply with quote

HoopStudies wrote:
What is the significance of the coefficients? This generally seems to suggest that offense wins but are those differences significant?


Everything was significant at 0.001. I should note that the equations above were all more accurate than the log5 type-estimates.
_________________
ed
Back to top
View user's profile Send private message Send e-mail
HoopStudies



Joined: 30 Dec 2004
Posts: 705
Location: Near Philadelphia, PA

PostPosted: Mon Apr 18, 2005 7:28 pm    Post subject: Re: Unstoppable force vs. immovable object Reply with quote

Ed Küpfer wrote:
HoopStudies wrote:
What is the significance of the coefficients? This generally seems to suggest that offense wins but are those differences significant?


Everything was significant at 0.001. I should note that the equations above were all more accurate than the log5 type-estimates.


What were the confidence intervals around the coefficients?
_________________
Dean Oliver
Author, Basketball on Paper
The postings are my own & don't necess represent positions, strategies or opinions of employers.
Back to top
View user's profile Send private message Visit poster's website
Ed Küpfer



Joined: 30 Dec 2004
Posts: 783
Location: Toronto

PostPosted: Mon Apr 18, 2005 7:43 pm    Post subject: Re: Unstoppable force vs. immovable object Reply with quote

HoopStudies wrote:
Ed Küpfer wrote:
HoopStudies wrote:
What is the significance of the coefficients? This generally seems to suggest that offense wins but are those differences significant?


Everything was significant at 0.001. I should note that the equations above were all more accurate than the log5 type-estimates.


What were the confidence intervals around the coefficients?


I should have mentioned that I only analysed home team results.

EFG%:

Code:
                                                 Odds     95% CI
Predictor      Coef    SE Coef        Z      P  Ratio  Lower  Upper
Constant   -3.14682  0.0161214  -195.20  0.000
1oEFG%      3.50320  0.0265769   131.81  0.000  33.22  31.54  35.00
2dEFG%      2.93039  0.0290748   100.79  0.000  18.73  17.70  19.83


TO%
Code:
                                                                95% CI
Predictor      Coef    SE Coef        Z      P  Odds Ratio   Lower   Upper
Constant   -3.59823  0.0119828  -300.28  0.000
1oTO%       5.77781  0.0578701    99.84  0.000      323.05  288.41  361.85
2dTO%       6.07966  0.0545186   111.52  0.000      436.88  392.61  486.15


OR%
Code:

                                                  Odds     95% CI
Predictor       Coef    SE Coef        Z      P  Ratio  Lower  Upper
Constant   0.0854750  0.0203862     4.19  0.000
1oOR%        4.17706  0.0212634   196.44  0.000  65.17  62.51  67.95
2dDR%       -3.10289  0.0247316  -125.46  0.000   0.04   0.04   0.05

_________________
ed
Back to top
View user's profile Send private message Send e-mail
Ed Küpfer



Joined: 30 Dec 2004
Posts: 783
Location: Toronto

PostPosted: Mon Apr 18, 2005 7:59 pm    Post subject: Re: Unstoppable force vs. immovable object Reply with quote

It evens out once you include away teams.

EFG%:
Code:

Predictor       Coef    SE Coef        Z      P  Ratio  Lower  Upper
Constant    -3.09752  0.0114094  -271.49  0.000
HOME       0.0460184  0.0007412    62.09  0.000   1.05   1.05   1.05
1oEFG%       3.22721  0.0187841   171.81  0.000  25.21  24.30  26.15
2dEFG%       3.00915  0.0205559   146.39  0.000  20.27  19.47  21.10


TO%
Code:
                                                                  95% CI
Predictor        Coef    SE Coef        Z      P  Odds Ratio   Lower   Upper
Constant     -3.65004  0.0084608  -431.40  0.000
HOME       -0.0284064  0.0010110   -28.10  0.000        0.97    0.97    0.97
1oTO%         6.05658  0.0407233   148.73  0.000      426.91  394.16  462.39
2dTO%         6.29438  0.0384019   163.91  0.000      541.52  502.26  583.85


OR%

Code:
                                      Odds     95% CI
Predictor       Coef    SE Coef        Z      P  Ratio  Lower  Upper
Constant    -2.99216  0.0057565  -519.79  0.000
HOME       0.0769602  0.0008089    95.14  0.000   1.08   1.08   1.08
1oOR%        3.78669  0.0151349   250.20  0.000  44.11  42.82  45.44
2dOR%        3.15742  0.0176015   179.38  0.000  23.51  22.71  24.34

_________________
ed
Back to top
View user's profile Send private message Send e-mail
Ed Küpfer



Joined: 30 Dec 2004
Posts: 783
Location: Toronto

PostPosted: Mon Apr 18, 2005 11:46 pm    Post subject: Re: Unstoppable force vs. immovable object Reply with quote

And to fill out the fourth factor,

FTA/poss = - 0.216 + 0.0108 * Home + 0.886 * 1FTA + 0.917 * 2FTA

where Home is a dummy, 1FTA is Team 1's FTA/poss, and 2FTA is team 2's FTA allowed/poss.

Code:
Predictor       Coef    SE Coef       T      P
Constant   -0.216195   0.006288  -34.38  0.000
Home       0.0108320  0.0008882   12.20  0.000
1FTA         0.88636    0.01859   47.69  0.000
2FTA         0.91714    0.01631   56.25  0.000


S = 0.0759282   R-Sq = 17.9%   R-Sq(adj) = 17.9%

_________________
ed
Back to top
View user's profile Send private message Send e-mail
Display posts from previous:   
Post new topic   Reply to topic    APBRmetrics Forum Index -> General discussion All times are GMT - 5 Hours
Page 1 of 1

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum


Powered by phpBB © 2001, 2005 phpBB Group