Author Message
thickbutcurious
Joined: 05 Jan 2010
Posts: 6
PostPosted: Sun Jan 24, 2010 11:04 pm Post subject: Who controls/more affects the pace? Reply with quote
I've done a fairly ham fisted-attempt at finding out whether the home or road team,the quicker or slower team has a greater impact on the pace of play.
I would be interested in whether anybody here has done the same properly,mainly to see if my results come remotely close to matching a properly done analysis.
My prior guess was that the teams would play closer to the home team's pace and slightly nearer the faster pace so I wanted to see if that was correct.
I also intended but it proved beynd me to see if a big difference in pace meant a team was more likely to outperform the efficiency margin prediction adjusted for likely pace I'd done.
i.e. does playing at a quick pace get a more deliberate opponent rushed into keeping up and thus lose efficiency or does an opponent playing very slowly frustrate a quicker team into playing worse,sadly I've no idea.
Iain
Back to top
View user's profile Send private message
thickbutcurious
Joined: 05 Jan 2010
Posts: 6
PostPosted: Mon Jan 25, 2010 12:47 pm Post subject: Reply with quote
As nobody has answered I might as well say what my answers were.
(Home team pace x 1.25 plus road team pace)/2.25 was the best predictor.
Faster teams slowed down to their opponent's pace more.
I'm assuming without explaining how I came up with that it's worthless,but I'm not saying as I know anyone here could do better and I dislike being mocked and/or patronised.
I reckon I sound annoyed here,I'm not.
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 807
PostPosted: Mon Jan 25, 2010 1:04 pm Post subject: Reply with quote
Do you mean
(Home team pace + 1.25 plus road team pace) /2.25?
Sometimes threads don't get answers for various reasons. Not a lot of people understand or are that interested in pace, thinking it is not that important or too hard to predict. I think it could be possible to get a better handle on it and would be worthwhile. There is some good stuff about pace in earlier threads; if you haven't already seen, it would be worth looking up via a search.
It does start to help more to see what your formula is now.
And by Home and Road team pace do you mean their overall pace or just when they are at home or on the road?
If faster teams slowed down to their opponent's pace more that would be good to incorporate in addition to home/road. You could do the same if the more winning or losing team has more influence or same with the more or less experienced team.
Instead of a weighted average of home and road it could become a weighted average of 4, 6 or 8 pace variables. Or more.
Last edited by Crow on Mon Jan 25, 2010 2:20 pm; edited 5 times in total
Back to top
View user's profile Send private message
thickbutcurious
Joined: 05 Jan 2010
Posts: 6
PostPosted: Mon Jan 25, 2010 1:46 pm Post subject: Reply with quote
Crow wrote:
Do you mean
(Home team pace + 1.25 plus road team pace) /2.25?
Sometimes threads don't get answers for various reasons. Not a lot of people understand or are that interested in pace, thinking it is not that important or too hard to predict. I think it could be possible to get a better handle on it and would be worthwhile. There is some good stuff about pace in earlier threads; if you haven't already seen, it would be worth looking up via a search. I haven't had the chance to do that or comment on substance.
It does start to help more to see what your formula is now.
And by Home and Road team pace do you mean their overall pace or just when they are at home or on the road?
Sorry I realised I sounded petulant but was too lazy to do more than just add the last sentence to counteract that.
I meant multiplied by 1.25 and their overall pace.
I did do a search but don't trust message board searches in general,and thought there might be more that someone would point me to.
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 807
PostPosted: Mon Jan 25, 2010 2:01 pm Post subject: Reply with quote
(Home team pace x 1.25 plus road team pace)/2.25
will be a huge number, on the order of 4-5,000 if you are using pace as possession per game. But I guess the formula would be ok if you are using pace standardized relative to average league pace.
Pace is often presented as pace per game for all possessions by both but it might be easier to predict pace for each team when they have the ball and combine the predictions.
I haven't had the chance to re-read and digest them all again but these appear to be the best links
http://sonicscentral.com/apbrmetrics/vi ... light=pace
http://sonicscentral.com/apbrmetrics/vi ... light=pace
http://sonicscentral.com/apbrmetrics/vi ... light=pace
http://sonicscentral.com/apbrmetrics/vi ... light=pace
http://sonicscentral.com/apbrmetrics/vi ... light=pace
http://sonicscentral.com/apbrmetrics/vi ... light=pace
http://sonicscentral.com/apbrmetrics/vi ... light=pace
http://sonicscentral.com/apbrmetrics/vi ... light=pace
http://basketballprospectus.com/article ... icleid=633
If you want more discussion about pace I'd read these, add them to your thinking and formula or formulate the specific questions you want feedback on.
Last edited by Crow on Mon Jan 25, 2010 2:23 pm; edited 1 time in total
Back to top
View user's profile Send private message
thickbutcurious
Joined: 05 Jan 2010
Posts: 6
PostPosted: Mon Jan 25, 2010 2:22 pm Post subject: Reply with quote
Crow wrote:
(Home team pace x 1.25 plus road team pace)/2.25
will be a huge number, on the order of 4-5,000 if you are using pace as possession per game. But I guess the formula would be ok if you are using pace standardized relative to average league pace.
Pace is often presented as pace per game for all possessions by both but it might be easier to predict pace for each team when they have the ball and combine the predictions.
Using Boston vs LAC tonight with hollinger pace for each team my rough guess is found by
(1.25*94+94.4)/2.25 a guess of 94.18 this still wasn't a good guess but was the best weighting just including the teams' overall pace.Obviously a bigger difference would make the weighting more relevant,although even then not much.What I found was far less interesting and meaningful than what I tried to find.
I had missed some of those links,given my lack of expertise they should occupy me for a while.
Thanks
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 807
PostPosted: Mon Jan 25, 2010 2:26 pm Post subject: Reply with quote
Sorry, when I read "Home team pace x 1.25 plus road team pace" I focused on the sign "X" and missed the word "plus" that appear later and the hierarchy of the operators was unclear to me.
Look forward to hearing your further thoughts on the subject. Definitely room for a better summary of the issues and a better projection metric.
Last edited by Crow on Mon Feb 01, 2010 11:04 pm; edited 2 times in total
Back to top
View user's profile Send private message
TwoTrey
Joined: 10 Nov 2009
Posts: 6
PostPosted: Mon Feb 01, 2010 8:08 pm Post subject: Reply with quote
Semi-ot
Is there a viewable source with a Pace Adjusted Points Per Game statistic?
Back to top
View user's profile Send private message
Ed Küpfer
Joined: 30 Dec 2004
Posts: 785
Location: Toronto
PostPosted: Mon Feb 01, 2010 9:08 pm Post subject: Reply with quote
TwoTrey wrote:
Is there a viewable source with a Pace Adjusted Points Per Game statistic?
That is simply points/possession, widely available in the points/100 possessions flavour as Offensive Rating, eg:
_________________
ed
Back to top
View user's profile Send private message Send e-mail
TwoTrey
Joined: 10 Nov 2009
Posts: 6
PostPosted: Mon Feb 01, 2010 9:51 pm Post subject: Reply with quote
Ed Küpfer wrote:
TwoTrey wrote:
Is there a viewable source with a Pace Adjusted Points Per Game statistic?
That is simply points/possession, widely available in the points/100 possessions flavour as Offensive Rating, eg:
My bad, should have re-read my own post.
Is there a viewable source with an Individual Pace Adjusted Points Per Game statistic?
Back to top
View user's profile Send private message
Joe
Joined: 27 Sep 2009
Posts: 94
Location: Long Island, NY
PostPosted: Mon Feb 01, 2010 9:56 pm Post subject: Reply with quote
http://www.draftexpress.com/stats.php?s ... l&qual=all
_________________
http://www.hoopdata.com
Back to top
View user's profile Send private message Visit poster's website
TwoTrey
Joined: 10 Nov 2009
Posts: 6
PostPosted: Mon Feb 01, 2010 10:45 pm Post subject: Reply with quote
Thank you
Back to top
View user's profile Send private message
findingneema
Joined: 25 Feb 2008
Posts: 34
Location: Atlanta, GA
PostPosted: Sat Feb 13, 2010 3:29 pm Post subject: Reply with quote
You can get a far better fit from just a simple y = a*home + b*away + c (which of course has more parameters). Also, I get a much smaller effect than you.
From the 08-09 season:
Code:
Call:
lm(formula = pace ~ home + away)
Residuals:
Min 1Q Median 3Q Max
-11.2019 -2.3346 -0.0829 2.3502 12.6572
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -95.89404 4.56902 -20.99 <2e-16 ***
away 1.02769 0.03541 29.02 <2e-16 ***
home 1.04047 0.03541 29.39 <2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 3.604 on 1227 degrees of freedom
Multiple R-squared: 0.5739, Adjusted R-squared: 0.5732
F-statistic: 826.2 on 2 and 1227 DF, p-value: < 2.2e-16
Code:
Formula: pace ~ (away + a * home)/(1+a)
Parameters:
Estimate Std. Error t value Pr(>|t|)
a 1.0259 0.1177 8.712 <2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 4.199 on 1229 degrees of freedom
Number of iterations to convergence: 2
Achieved convergence tolerance: 2.019e-08