Adj Efficiencies,Motion Chart + new site (DSMok1, 2011)
Posted: Fri Apr 15, 2011 1:47 am
page 1 of 2
Author Message
DSMok1
Joined: 05 Aug 2009
Posts: 611
Location: Where the wind comes sweeping down the plains
PostPosted: Wed Jan 05, 2011 5:24 pm Post subject: NBA Adjusted Efficiencies: Google Motion Chart (+ new site) Reply with quote
I've been working on adjusting efficiencies, recently, and have tabulated every game this year and each team's adjusted efficiency (adjusted for location, opponent, and rest advantage).
I thought this would look good in a Google Motion Chart--to do this, I created a 5-game moving average for each team. This somewhat smooths out the data and helps one to see the trends for each team.
Hint: Orlando is looking good with Hedo!
The visualization is at my new sports stats blog, at http://godismyjudgeok.com/DStats/2011/n ... -1-4-2011/
I dusted off my site design skills and borrowed a lot from my non-sports site's design. This is my first foray into a full-fledged CMS... so there could be some glitches.
_________________
GodismyJudgeOK.com/DStats
Twitter.com/DSMok1
Back to top
View user's profile Send private message Send e-mail Visit poster's website
EvanZ
Joined: 22 Nov 2010
Posts: 305
PostPosted: Wed Jan 05, 2011 5:43 pm Post subject: Reply with quote
Very cool. How do you link the data to the chart anyway? Is it a Google spreadsheet?
_________________
http://www.thecity2.com
http://www.ibb.gatech.edu/evan-zamir
Back to top
View user's profile Send private message
DSMok1
Joined: 05 Aug 2009
Posts: 611
Location: Where the wind comes sweeping down the plains
PostPosted: Wed Jan 05, 2011 7:14 pm Post subject: Reply with quote
EvanZ wrote:
Very cool. How do you link the data to the chart anyway? Is it a Google spreadsheet?
I just coded it directly; it's not too hard to create the strings in Excel. If you view the source of the page, you'll quickly see all of the data points in the <head> portion of the page.
The Google page showing how to do this is here: http://code.google.com/apis/visualizati ... chart.html
_________________
GodismyJudgeOK.com/DStats
Twitter.com/DSMok1
Back to top
View user's profile Send private message Send e-mail Visit poster's website
DSMok1
Joined: 05 Aug 2009
Posts: 611
Location: Where the wind comes sweeping down the plains
PostPosted: Tue Jan 11, 2011 3:58 pm Post subject: Reply with quote
Updated, using 7-game weighted trailing moving averages, rather than 5 game simple moving averages.
http://godismyjudgeok.com/DStats/2011/n ... 1-11-2011/
Also a look at Orlando's team numbers.
_________________
GodismyJudgeOK.com/DStats
Twitter.com/DSMok1
Back to top
View user's profile Send private message Send e-mail Visit poster's website
Crow
Joined: 20 Jan 2009
Posts: 825
PostPosted: Tue Jan 11, 2011 6:58 pm Post subject: Reply with quote
The 7-game weighted moving averages as currently constructed sure move around a lot day to day.
The visual impression from your chart as of 1/8/2011 is very different than the full season chart here http://www.nbastuffer.com/2010-2011_NBA ... Stats.html.
For the season I am not sure I recall ever seeing the teams so spaced apart and evenly so in all 4 quadrants. On the 7-game weighted moving averages there is, at least temporarily, a very strong pattern of teams being good or bad on both efficiencies.
Back to top
View user's profile Send private message
erivera7
Joined: 19 Jan 2009
Posts: 185
Location: Chicago, IL
PostPosted: Tue Jan 11, 2011 7:50 pm Post subject: Reply with quote
Easy to say that Turkoglu's contributions could be small sample size, but I see them as legitimate. It's a perfect convergence of situations for Turkoglu, and he's been taking advantage of them for some stellar basketball since the trade.
_________________
@erivera7
I cover the Orlando Magic - Magic Basketball
Back to top
View user's profile Send private message
DSMok1
Joined: 05 Aug 2009
Posts: 611
Location: Where the wind comes sweeping down the plains
PostPosted: Tue Jan 11, 2011 8:07 pm Post subject: Reply with quote
erivera7 wrote:
Easy to say that Turkoglu's contributions could be small sample size, but I see them as legitimate. It's a perfect convergence of situations for Turkoglu, and he's been taking advantage of them for some stellar basketball since the trade.
I'm probably going to do a feature post on the trade and the changes in Orlando. Have you done a retrospective look at what the trade meant yet? Turkoglu's case is just what statistician's don't really like to see: proof that context matters... a lot. So Turkoglu on Orlando is not the same as he is elsewhere. How to know that ahead of time would be remarkably difficult to disentangle, if not impossible. How does one quantify fit?
_________________
GodismyJudgeOK.com/DStats
Twitter.com/DSMok1
Back to top
View user's profile Send private message Send e-mail Visit poster's website
erivera7
Joined: 19 Jan 2009
Posts: 185
Location: Chicago, IL
PostPosted: Tue Jan 11, 2011 8:15 pm Post subject: Reply with quote
DSMok1 wrote:
I'm probably going to do a feature post on the trade and the changes in Orlando. Have you done a retrospective look at what the trade meant yet? Turkoglu's case is just what statistician's don't really like to see: proof that context matters... a lot. So Turkoglu on Orlando is not the same as he is elsewhere. How to know that ahead of time would be remarkably difficult to disentangle, if not impossible. How does one quantify fit?
Haven't done a deep look yet. I know Kevin Pelton wrote about the Magic in his recent Trend Watch at ESPN and noted their schedule-adjusted point differential skyrocketed since the trades.
As for Turkoglu, context does matter a lot. Agreed. Like I said previously, he's taking advantage of a convergence of factors that benefits him -- allowed to be a playmaker in the offense, deep-rooted chemistry with Dwight Howard in the pick and roll, trusted by Van Gundy to make plays, etc.
Turkoglu's numbers don't jump off the page, but his impact on Orlando is palpable. He's a big reason for the nine-game winning streak.
Should we have seen this coming? I knew Turkoglu would fit like a glove with the Magic, but these results have been remarkable. Triple-double and 17-assist game in his first 11 games back. I didn't see that coming at all, even though I knew Turkoglu would be better off with the basketball in his hands again.
_________________
@erivera7
I cover the Orlando Magic - Magic Basketball
Back to top
View user's profile Send private message
DSMok1
Joined: 05 Aug 2009
Posts: 611
Location: Where the wind comes sweeping down the plains
PostPosted: Wed Jan 12, 2011 4:28 pm Post subject: Reply with quote
A new post with a new motion chart: comparing the NBA's top 35 point guards, using any and every advanced stat available.
http://godismyjudgeok.com/DStats/2011/g ... son-chart/
_________________
GodismyJudgeOK.com/DStats
Twitter.com/DSMok1
Back to top
View user's profile Send private message Send e-mail Visit poster's website
Mike G
Joined: 14 Jan 2005
Posts: 3621
Location: Hendersonville, NC
PostPosted: Wed Jan 12, 2011 4:53 pm Post subject: Reply with quote
Are you adjusting for home assist inflation?
For % of competition who are starters?
_________________
`
36% of all statistics are wrong
Back to top
View user's profile Send private message Send e-mail
DSMok1
Joined: 05 Aug 2009
Posts: 611
Location: Where the wind comes sweeping down the plains
PostPosted: Wed Jan 12, 2011 5:33 pm Post subject: Reply with quote
Mike G wrote:
Are you adjusting for home assist inflation?
For % of competition who are starters?
Nope, the numbers are straight from Basketball Reference.
_________________
GodismyJudgeOK.com/DStats
Twitter.com/DSMok1
Back to top
View user's profile Send private message Send e-mail Visit poster's website
DSMok1
Joined: 05 Aug 2009
Posts: 611
Location: Where the wind comes sweeping down the plains
PostPosted: Fri Jan 14, 2011 5:23 pm Post subject: Reply with quote
I just posted updated adjusted efficiency ratings, and looked at a couple of questions people have had:
How much does adjusting for rest days impact the ratings?
What is the impact of minimizing residuals^1.5 vs. residuals^2 or something else?
http://godismyjudgeok.com/DStats/2011/n ... 1-13-2011/
_________________
GodismyJudgeOK.com/DStats
Twitter.com/DSMok1
Back to top
View user's profile Send private message Send e-mail Visit poster's website
Ryan J. Parker
Joined: 23 Mar 2007
Posts: 711
Location: Raleigh, NC
PostPosted: Fri Jan 14, 2011 5:40 pm Post subject: Reply with quote
How exactly are you raising a negative residual to the 1.5 power?
Even R can't do it!
> (-5)^(1.5)
[1] NaN
_________________
I am a basketball geek.
Back to top
View user's profile Send private message Visit poster's website
DSMok1
Joined: 05 Aug 2009
Posts: 611
Location: Where the wind comes sweeping down the plains
PostPosted: Fri Jan 14, 2011 5:42 pm Post subject: Reply with quote
Ryan J. Parker wrote:
How exactly are you raising a negative residual to the 1.5 power?
Even R can't do it!
> (-5)^(1.5)
[1] NaN
Laughing (abs(-5))^1.5
_________________
GodismyJudgeOK.com/DStats
Twitter.com/DSMok1
Back to top
View user's profile Send private message Send e-mail Visit poster's website
Ryan J. Parker
Joined: 23 Mar 2007
Posts: 711
Location: Raleigh, NC
PostPosted: Fri Jan 14, 2011 5:48 pm Post subject: Reply with quote
Laughing
That's mathematically intractable (which is one reason why 2 is a lot), but interesting to see.
Gotta spend some time away from these stubborn statisticians!
page 2 of 2
Author Message
Ryan J. Parker
Joined: 23 Mar 2007
Posts: 708
Location: Raleigh, NC
PostPosted: Fri Jan 14, 2011 5:50 pm Post subject: Reply with quote
Oh, and for future clarity, you might want to say you're using the absolute deviation raised to whatever power. Cool
_________________
I am a basketball geek.
Back to top
View user's profile Send private message Visit poster's website
Jeff Fogle
Joined: 11 Jan 2011
Posts: 62
PostPosted: Fri Jan 14, 2011 5:52 pm Post subject: Reply with quote
DS,
Haven't had a chance to ask this yet. When you were doing your study on the impact of back-to-back's, did you notice if there was a trend toward a larger influence deeper in the season?
I know, well in the past, the legal betting markets used to increase their adjustments for a b2b as the season progressed. Meaning...not much of an adjustment early in the season when everyone was relatively fresh...but then adjustments of as much as 3-4 points in the line from roughly our current point in the season onward (especially if it was a 4th game in 5 nights) because of cumulative fatigue.
The markets seem to adjust less than they used to, as pace has decreased over time, and coaches may have gotten smarter about how they pace their players for quickfire challenges and the full season. Was wondering if your data suggested 1.5 solid through a season...or if it was more like 0.75 in the first quarter of the season, but as much as 2.25 or even more after a certain point.
Thanks in advance if you have a chance to respond...
Back to top
View user's profile Send private message
DSMok1
Joined: 05 Aug 2009
Posts: 602
Location: Where the wind comes sweeping down the plains
PostPosted: Wed Jan 19, 2011 4:36 pm Post subject: Reply with quote
A new post is up, using a motion chart to explore the NBA's centers and C/PF's.
Also, an introductory look into the world of K-Means Clustering as a means of classifying NBA player positions/positionality.
http://godismyjudgeok.com/DStats/2011/g ... lustering/
_________________
GodismyJudgeOK.com/DStats
Author Message
DSMok1
Joined: 05 Aug 2009
Posts: 611
Location: Where the wind comes sweeping down the plains
PostPosted: Wed Jan 05, 2011 5:24 pm Post subject: NBA Adjusted Efficiencies: Google Motion Chart (+ new site) Reply with quote
I've been working on adjusting efficiencies, recently, and have tabulated every game this year and each team's adjusted efficiency (adjusted for location, opponent, and rest advantage).
I thought this would look good in a Google Motion Chart--to do this, I created a 5-game moving average for each team. This somewhat smooths out the data and helps one to see the trends for each team.
Hint: Orlando is looking good with Hedo!
The visualization is at my new sports stats blog, at http://godismyjudgeok.com/DStats/2011/n ... -1-4-2011/
I dusted off my site design skills and borrowed a lot from my non-sports site's design. This is my first foray into a full-fledged CMS... so there could be some glitches.
_________________
GodismyJudgeOK.com/DStats
Twitter.com/DSMok1
Back to top
View user's profile Send private message Send e-mail Visit poster's website
EvanZ
Joined: 22 Nov 2010
Posts: 305
PostPosted: Wed Jan 05, 2011 5:43 pm Post subject: Reply with quote
Very cool. How do you link the data to the chart anyway? Is it a Google spreadsheet?
_________________
http://www.thecity2.com
http://www.ibb.gatech.edu/evan-zamir
Back to top
View user's profile Send private message
DSMok1
Joined: 05 Aug 2009
Posts: 611
Location: Where the wind comes sweeping down the plains
PostPosted: Wed Jan 05, 2011 7:14 pm Post subject: Reply with quote
EvanZ wrote:
Very cool. How do you link the data to the chart anyway? Is it a Google spreadsheet?
I just coded it directly; it's not too hard to create the strings in Excel. If you view the source of the page, you'll quickly see all of the data points in the <head> portion of the page.
The Google page showing how to do this is here: http://code.google.com/apis/visualizati ... chart.html
_________________
GodismyJudgeOK.com/DStats
Twitter.com/DSMok1
Back to top
View user's profile Send private message Send e-mail Visit poster's website
DSMok1
Joined: 05 Aug 2009
Posts: 611
Location: Where the wind comes sweeping down the plains
PostPosted: Tue Jan 11, 2011 3:58 pm Post subject: Reply with quote
Updated, using 7-game weighted trailing moving averages, rather than 5 game simple moving averages.
http://godismyjudgeok.com/DStats/2011/n ... 1-11-2011/
Also a look at Orlando's team numbers.
_________________
GodismyJudgeOK.com/DStats
Twitter.com/DSMok1
Back to top
View user's profile Send private message Send e-mail Visit poster's website
Crow
Joined: 20 Jan 2009
Posts: 825
PostPosted: Tue Jan 11, 2011 6:58 pm Post subject: Reply with quote
The 7-game weighted moving averages as currently constructed sure move around a lot day to day.
The visual impression from your chart as of 1/8/2011 is very different than the full season chart here http://www.nbastuffer.com/2010-2011_NBA ... Stats.html.
For the season I am not sure I recall ever seeing the teams so spaced apart and evenly so in all 4 quadrants. On the 7-game weighted moving averages there is, at least temporarily, a very strong pattern of teams being good or bad on both efficiencies.
Back to top
View user's profile Send private message
erivera7
Joined: 19 Jan 2009
Posts: 185
Location: Chicago, IL
PostPosted: Tue Jan 11, 2011 7:50 pm Post subject: Reply with quote
Easy to say that Turkoglu's contributions could be small sample size, but I see them as legitimate. It's a perfect convergence of situations for Turkoglu, and he's been taking advantage of them for some stellar basketball since the trade.
_________________
@erivera7
I cover the Orlando Magic - Magic Basketball
Back to top
View user's profile Send private message
DSMok1
Joined: 05 Aug 2009
Posts: 611
Location: Where the wind comes sweeping down the plains
PostPosted: Tue Jan 11, 2011 8:07 pm Post subject: Reply with quote
erivera7 wrote:
Easy to say that Turkoglu's contributions could be small sample size, but I see them as legitimate. It's a perfect convergence of situations for Turkoglu, and he's been taking advantage of them for some stellar basketball since the trade.
I'm probably going to do a feature post on the trade and the changes in Orlando. Have you done a retrospective look at what the trade meant yet? Turkoglu's case is just what statistician's don't really like to see: proof that context matters... a lot. So Turkoglu on Orlando is not the same as he is elsewhere. How to know that ahead of time would be remarkably difficult to disentangle, if not impossible. How does one quantify fit?
_________________
GodismyJudgeOK.com/DStats
Twitter.com/DSMok1
Back to top
View user's profile Send private message Send e-mail Visit poster's website
erivera7
Joined: 19 Jan 2009
Posts: 185
Location: Chicago, IL
PostPosted: Tue Jan 11, 2011 8:15 pm Post subject: Reply with quote
DSMok1 wrote:
I'm probably going to do a feature post on the trade and the changes in Orlando. Have you done a retrospective look at what the trade meant yet? Turkoglu's case is just what statistician's don't really like to see: proof that context matters... a lot. So Turkoglu on Orlando is not the same as he is elsewhere. How to know that ahead of time would be remarkably difficult to disentangle, if not impossible. How does one quantify fit?
Haven't done a deep look yet. I know Kevin Pelton wrote about the Magic in his recent Trend Watch at ESPN and noted their schedule-adjusted point differential skyrocketed since the trades.
As for Turkoglu, context does matter a lot. Agreed. Like I said previously, he's taking advantage of a convergence of factors that benefits him -- allowed to be a playmaker in the offense, deep-rooted chemistry with Dwight Howard in the pick and roll, trusted by Van Gundy to make plays, etc.
Turkoglu's numbers don't jump off the page, but his impact on Orlando is palpable. He's a big reason for the nine-game winning streak.
Should we have seen this coming? I knew Turkoglu would fit like a glove with the Magic, but these results have been remarkable. Triple-double and 17-assist game in his first 11 games back. I didn't see that coming at all, even though I knew Turkoglu would be better off with the basketball in his hands again.
_________________
@erivera7
I cover the Orlando Magic - Magic Basketball
Back to top
View user's profile Send private message
DSMok1
Joined: 05 Aug 2009
Posts: 611
Location: Where the wind comes sweeping down the plains
PostPosted: Wed Jan 12, 2011 4:28 pm Post subject: Reply with quote
A new post with a new motion chart: comparing the NBA's top 35 point guards, using any and every advanced stat available.
http://godismyjudgeok.com/DStats/2011/g ... son-chart/
_________________
GodismyJudgeOK.com/DStats
Twitter.com/DSMok1
Back to top
View user's profile Send private message Send e-mail Visit poster's website
Mike G
Joined: 14 Jan 2005
Posts: 3621
Location: Hendersonville, NC
PostPosted: Wed Jan 12, 2011 4:53 pm Post subject: Reply with quote
Are you adjusting for home assist inflation?
For % of competition who are starters?
_________________
`
36% of all statistics are wrong
Back to top
View user's profile Send private message Send e-mail
DSMok1
Joined: 05 Aug 2009
Posts: 611
Location: Where the wind comes sweeping down the plains
PostPosted: Wed Jan 12, 2011 5:33 pm Post subject: Reply with quote
Mike G wrote:
Are you adjusting for home assist inflation?
For % of competition who are starters?
Nope, the numbers are straight from Basketball Reference.
_________________
GodismyJudgeOK.com/DStats
Twitter.com/DSMok1
Back to top
View user's profile Send private message Send e-mail Visit poster's website
DSMok1
Joined: 05 Aug 2009
Posts: 611
Location: Where the wind comes sweeping down the plains
PostPosted: Fri Jan 14, 2011 5:23 pm Post subject: Reply with quote
I just posted updated adjusted efficiency ratings, and looked at a couple of questions people have had:
How much does adjusting for rest days impact the ratings?
What is the impact of minimizing residuals^1.5 vs. residuals^2 or something else?
http://godismyjudgeok.com/DStats/2011/n ... 1-13-2011/
_________________
GodismyJudgeOK.com/DStats
Twitter.com/DSMok1
Back to top
View user's profile Send private message Send e-mail Visit poster's website
Ryan J. Parker
Joined: 23 Mar 2007
Posts: 711
Location: Raleigh, NC
PostPosted: Fri Jan 14, 2011 5:40 pm Post subject: Reply with quote
How exactly are you raising a negative residual to the 1.5 power?
Even R can't do it!
> (-5)^(1.5)
[1] NaN
_________________
I am a basketball geek.
Back to top
View user's profile Send private message Visit poster's website
DSMok1
Joined: 05 Aug 2009
Posts: 611
Location: Where the wind comes sweeping down the plains
PostPosted: Fri Jan 14, 2011 5:42 pm Post subject: Reply with quote
Ryan J. Parker wrote:
How exactly are you raising a negative residual to the 1.5 power?
Even R can't do it!
> (-5)^(1.5)
[1] NaN
Laughing (abs(-5))^1.5
_________________
GodismyJudgeOK.com/DStats
Twitter.com/DSMok1
Back to top
View user's profile Send private message Send e-mail Visit poster's website
Ryan J. Parker
Joined: 23 Mar 2007
Posts: 711
Location: Raleigh, NC
PostPosted: Fri Jan 14, 2011 5:48 pm Post subject: Reply with quote
Laughing
That's mathematically intractable (which is one reason why 2 is a lot), but interesting to see.
Gotta spend some time away from these stubborn statisticians!
page 2 of 2
Author Message
Ryan J. Parker
Joined: 23 Mar 2007
Posts: 708
Location: Raleigh, NC
PostPosted: Fri Jan 14, 2011 5:50 pm Post subject: Reply with quote
Oh, and for future clarity, you might want to say you're using the absolute deviation raised to whatever power. Cool
_________________
I am a basketball geek.
Back to top
View user's profile Send private message Visit poster's website
Jeff Fogle
Joined: 11 Jan 2011
Posts: 62
PostPosted: Fri Jan 14, 2011 5:52 pm Post subject: Reply with quote
DS,
Haven't had a chance to ask this yet. When you were doing your study on the impact of back-to-back's, did you notice if there was a trend toward a larger influence deeper in the season?
I know, well in the past, the legal betting markets used to increase their adjustments for a b2b as the season progressed. Meaning...not much of an adjustment early in the season when everyone was relatively fresh...but then adjustments of as much as 3-4 points in the line from roughly our current point in the season onward (especially if it was a 4th game in 5 nights) because of cumulative fatigue.
The markets seem to adjust less than they used to, as pace has decreased over time, and coaches may have gotten smarter about how they pace their players for quickfire challenges and the full season. Was wondering if your data suggested 1.5 solid through a season...or if it was more like 0.75 in the first quarter of the season, but as much as 2.25 or even more after a certain point.
Thanks in advance if you have a chance to respond...
Back to top
View user's profile Send private message
DSMok1
Joined: 05 Aug 2009
Posts: 602
Location: Where the wind comes sweeping down the plains
PostPosted: Wed Jan 19, 2011 4:36 pm Post subject: Reply with quote
A new post is up, using a motion chart to explore the NBA's centers and C/PF's.
Also, an introductory look into the world of K-Means Clustering as a means of classifying NBA player positions/positionality.
http://godismyjudgeok.com/DStats/2011/g ... lustering/
_________________
GodismyJudgeOK.com/DStats