|
APBRmetrics The statistical revolution will not be televised.
|
View previous topic :: View next topic |
Author |
Message |
GB
Joined: 29 Jun 2005 Posts: 4
|
Posted: Thu Jun 30, 2005 1:23 pm Post subject: |
|
|
My name is GB...I'm a longtime hoops junkie. I'll mostly be a lurker and leamer.
Great board. I've read almost all the way through it and the 'Moneyball' articles about the Sonics... |
|
Back to top |
|
|
bballfan72031
Joined: 13 Feb 2005 Posts: 54
|
Posted: Tue Jul 19, 2005 11:52 pm Post subject: |
|
|
What the hey, I'll make a post here.
I'm a 15 year old kid living in Portland, Oregon, who just lives for basketball. For quite a long time now, I've been a lot more interested in the managing parts of basketball than playing competitively (creating trades is the most fun thing in the world for me).
I've seen a lot of the value in statistics over the years, and so I love learning more about them. And in the fact that I've always kinda liked numbers and patterns, and statistics seem really cool to me. I've been lurking here for....I dunno....has it been a couple of years? I'm not really sure. Started reading the posts at the old yahoo site (and I must say, I'm very happy with the shift to the message board). I was first linked to this site over at Hoopsworld.
Anyway....you won't see me logged on much because I only log on to post, but I'm still here lurking as a guest a lot of the time. I don't expect to add much, if at all, to the board, as I just don't know that much (particularly with the mathematics....I'm often lost when reading some of the posts here). But I really enjoy reading everything everyone has to say here, and I appreciate that you all allow me to listen and learn from ya.
Awesome site! Keep it up.
Last edited by bballfan72031 on Sun Jul 30, 2006 12:33 pm; edited 2 times in total |
|
Back to top |
|
|
Furps
Joined: 25 Jul 2005 Posts: 2 Location: Washington, DC
|
Posted: Mon Jul 25, 2005 1:13 pm Post subject: Great to be here... |
|
|
I am just getting started, and have truly enjoyed reading everyone's posts.
I'm particularly interested in a basketball equivalent of Bill James' pythagorean projections. Does this community post projections using a basketball version of runs scored & runs allowed before the season? Does anyone? Thanks. |
|
Back to top |
|
|
kjb
Joined: 03 Jan 2005 Posts: 865 Location: Washington, DC
|
Posted: Mon Jul 25, 2005 4:14 pm Post subject: Re: Great to be here... |
|
|
Furps wrote: | I am just getting started, and have truly enjoyed reading everyone's posts.
I'm particularly interested in a basketball equivalent of Bill James' pythagorean projections. Does this community post projections using a basketball version of runs scored & runs allowed before the season? Does anyone? Thanks. |
Welcome -- There is an NBA pythagorean formula. Justin includes them at his terrific site www.basketball-reference.com.
You can calculate it yourself using:
pts^14 / (pts^14) + (opp_pts^14)
For example: last season's Wizards scored 8241 points and allowed 8268. So, it's (8241^14) / (8241^14) + (8268 ^14) -- for a pythagorean winning percentage of .488. Multiply by 82 games, and get a project win-loss record of 40-42.
On his site, Justin also includes individual won-loss records, which also uses the Pythagorean method. This is done using offensive and defensive ratings, which (using Dean Oliver's methods from "Basketball On Paper") are expressed in pts per 100 possessions. |
|
Back to top |
|
|
HoopStudies
Joined: 30 Dec 2004 Posts: 705 Location: Near Philadelphia, PA
|
Posted: Mon Jul 25, 2005 4:42 pm Post subject: Re: Great to be here... |
|
|
WizardsKev wrote: | Furps wrote: | I am just getting started, and have truly enjoyed reading everyone's posts.
I'm particularly interested in a basketball equivalent of Bill James' pythagorean projections. Does this community post projections using a basketball version of runs scored & runs allowed before the season? Does anyone? Thanks. |
Welcome -- There is an NBA pythagorean formula. Justin includes them at his terrific site www.basketball-reference.com.
You can calculate it yourself using:
pts^14 / (pts^14) + (opp_pts^14)
For example: last season's Wizards scored 8241 points and allowed 8268. So, it's (8241^14) / (8241^14) + (8268 ^14) -- for a pythagorean winning percentage of .488. Multiply by 82 games, and get a project win-loss record of 40-42.
On his site, Justin also includes individual won-loss records, which also uses the Pythagorean method. This is done using offensive and defensive ratings, which (using Dean Oliver's methods from "Basketball On Paper") are expressed in pts per 100 possessions. |
Not to overload you, but there are at least two others ways of doing this sort of thing.
John Hollinger has a simple translation of net pts/game into projected winning %.
I have a more sophisticated method that uses both average scores (or pts/possession) and variation of scores. That's called a Bell Curve method (or Correlated Gaussian before I decided that the name scared people). It's much more powerful, but not as convenient.
In general, you'll find that basketball methods aren't nearly as consolidated as baseball's. We're younger, the game is more complex, and we don't have the one writer selling 100,000 books on basketball sabermetrics every year to unify terminology. It took us until fairly recently to accept one definition of "possession", which is about as fundamental as an out in baseball. And, to set that straight, a possession is the period of time between when one team has control of the ball and when the other does. So a missed shot, an offensive rebound, and a made shot is just one possession (and two "plays"). _________________ Dean Oliver
Author, Basketball on Paper
The postings are my own & don't necess represent positions, strategies or opinions of employers. |
|
Back to top |
|
|
Kevin Pelton Site Admin
Joined: 30 Dec 2004 Posts: 979 Location: Seattle
|
Posted: Mon Jul 25, 2005 11:31 pm Post subject: Re: Great to be here... |
|
|
Furps wrote: | Does this community post projections using a basketball version of runs scored & runs allowed before the season? Does anyone? Thanks. |
I'm not quite sure what specifically you're asking here -- taking the lessons from the previous season's Pythagorean record, or predicting records based on player ratings and projection systems?
If it's the latter, I did this:
http://www.nba.com/sonics/news/nbapreview04.html
and John Hollinger did it informally at the old version of this board:
http://sports.groups.yahoo.com/group/APBR_analysis/message/5195
There are some of those results I should be hiding. |
|
Back to top |
|
|
Furps
Joined: 25 Jul 2005 Posts: 2 Location: Washington, DC
|
Posted: Tue Jul 26, 2005 10:40 am Post subject: Re: Great to be here... |
|
|
admin wrote: | Furps wrote: | Does this community post projections using a basketball version of runs scored & runs allowed before the season? Does anyone? Thanks. |
I'm not quite sure what specifically you're asking here -- taking the lessons from the previous season's Pythagorean record, or predicting records based on player ratings and projection systems?
If it's the latter, I did this:
http://www.nba.com/sonics/news/nbapreview04.html
and John Hollinger did it informally at the old version of this board:
http://sports.groups.yahoo.com/group/APBR_analysis/message/5195
There are some of those results I should be hiding. |
Yes. This is what I was asking about. Looking for pre-season projections of a team's pythag record...Have done some work with baseball Win Shares in the past, but found I was more interested in projections - figured I'd try and find what was out there in the hoops world.
Great stuff. Thanks much.
Already love the site. Does that mean I'm now a Sonics fan?
|
|
Back to top |
|
|
Ed Küpfer
Joined: 30 Dec 2004 Posts: 787 Location: Toronto
|
Posted: Tue Jul 26, 2005 10:53 am Post subject: Re: Great to be here... |
|
|
Furps wrote: | Yes. This is what I was asking about. Looking for pre-season projections of a team's pythag record...Have done some work with baseball Win Shares in the past, but found I was more interested in projections - figured I'd try and find what was out there in the hoops world. |
Did someone say "Pythagorean"?
Quote: | Already love the site. Does that mean I'm now a Sonics fan? |
It's like a law of nature. By virtue of being stats dorks we are inherently Sonics fans, the same way guitar players are automatically Hendrix fans. _________________ ed |
|
Back to top |
|
|
capnhistory
Joined: 27 Jul 2005 Posts: 62 Location: Durham, North Carolina
|
Posted: Fri Jul 29, 2005 9:46 am Post subject: Salutations |
|
|
I'm a recent convert to basketball in general, so I'm still developing my knowledge of the league's history and the technical aspects of the game. I've become fascinated with using the statistics to find deeper truths about teams, leagues and players. I'm still finishing Dean's book, and so far I love it. I'm working on writing a program for my TI-83 calculator that can caclulate advanced stats from the basic numbers. It won't be able to do whata sophisticated computer program could, but I find it really helpful, for quick studies of boxscores or single season numbers. I still have some kinks to work out, but I may share it with you if anyone's interested. (It strikes me that not many people may be walking around with TI-83's, but I think it's a wonderful tool.) _________________ Throw it down big man!
More expansive basketball babble at a slower pace The Captain of History |
|
Back to top |
|
|
mathayus
Joined: 15 Aug 2005 Posts: 212
|
Posted: Mon Aug 15, 2005 3:16 pm Post subject: |
|
|
Hello all,
I am a long time basketball fan and numbers geek who just dove into the online basketball community this last season and and just found out about this forum recently. I've been pretty active over at RealGM for the past 6 months or so. I am excited to get involved with those who are actually on the frontier of this stuff.
As for my background, well I am a 27-yeard old doctoral student of artificial intelligence in California. Strong at math, and I've always loved "statistics" as I've known it, but I really don't have any background in the science of statistics worth mentioning.
My own independent exploration into quantifying basketball are limited and primitive, but I'll share them here for reference anyway. During the 03-04 season I began using old-school statistics to try to come up with a rating for how valuable a player is. A good experience, but a frustrating one. Two players in particular forced me to conclude that old school stats just weren't enough: Kirilenko and Nash. I'm sure you can guess what my problems were just by knowing what these players bring to the floor. So, it has been exciting for me to discover the new stats that are available online which address areas that weren't well represented before, and it's also great to find a place like this and actually see debates about this stuff.
I also have also been developing and refining a formula to estimate Hall of Fame worthiness. Which is an interesting process, although also rather frustrating for the basic reason that it's hard to gather complete statistics for players who are actually old enough to be in the Hall of Fame. Actually, if you have any suggestions for gathering data, please let me know.
Thanks,
Matt |
|
Back to top |
|
|
MDC
Joined: 11 Jun 2005 Posts: 36
|
Posted: Fri Aug 19, 2005 2:27 pm Post subject: |
|
|
Hello everyone, my name is Nicholas Hunt. I'm transferring to UC San Diego this fall as a major in Computer Science. I have a passion for basketball and videogames. I have devoured the work of the people on this forum, including Dean's book and Hollinger's books. I'm really interested in the adjusted +/- stuff that Dan has been doing and the defensive game charting work that Dean promotes, 82games is attempting, and Kevin Broom has wrriten about. My biggest pet peeve is the prevailence of flawed measures used by professional 'analysts' in the media, like points allowed per game and FG%. I haven't taken my statistical courses yet, nor have I used Excel enough, but I soon hope to be able to contribute research of my own. My home team is the Warriors (which has been brutal).
Nick |
|
Back to top |
|
|
John Hollinger
Joined: 14 Feb 2005 Posts: 175
|
Posted: Fri Sep 16, 2005 11:19 am Post subject: |
|
|
Hey everyone, sorry it took me so long to stopy by the new place but I love what you've done with it.
I noticed Dean's comment on the Pythagorean method and I wanted to note that I've converted to using the "Pythagorean 16.5" method. The "2.75" method I had been using worked well for average teams but for the very worst and very best it broke down, and unfortunately those are the teams that we tend to study the most. This year's Pro Basketball Forecast makes the switch.
John |
|
Back to top |
|
|
yakustk
Joined: 16 Sep 2005 Posts: 15
|
Posted: Fri Sep 16, 2005 11:33 pm Post subject: |
|
|
John, just got your 04/05 book and am looking forward to the next one...any idea when it is going to be released?
I just joined tonight after lurking a bit...great stuff all around. I've enjoyed doing research into baseball and want to learn some basketball as well. Are NBA boxscores available in Excel like baseball stats are at retrosheet.org? |
|
Back to top |
|
|
John Hollinger
Joined: 14 Feb 2005 Posts: 175
|
Posted: Sat Sep 17, 2005 8:51 pm Post subject: |
|
|
The book should be out within the next two weeks, so hopefully everybody has it in their hands soon.
John |
|
Back to top |
|
|
tmorgan
Joined: 30 Sep 2005 Posts: 2 Location: Lexington, KY
|
Posted: Fri Sep 30, 2005 1:51 pm Post subject: |
|
|
Just joined. I'm a teacher by trade, and a baseball fan at heart (hope that's not sacrilege here), but basketball is a clear #2. I'm a native Michigander and a Pistons diehard, but I've enjoyed reading the information here enough to become a closet Sonics fan. Guess that's all... I just felt a little dirty lurking, so I registered. |
|
Back to top |
|
|
|
|
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
|