TheStar.com - Analyze this: NBA '04 (September 19, 2003)
Last year, Oliver wrote to James asking for guidance on his basketball research only to be sent back word that James could offer no help because he was planning a basketball tome of his own. It's unclear whether James is at work on a hoops-inspired manuscript. Nevertheless, Oliver felt an urgency to get his work to market.
--posted by TangoTiger at 09:43 AM EDT
Posted 10:14 a.m.,
September 19, 2003
(#1) -
Scoriano
It could be James or could be STATS, Inc. to which he is referring.
Posted 10:42 a.m.,
September 19, 2003
(#2) -
Andrew Edwards
Some has got to do this for hockey too. I'm not enough of a fan to do it myself, but it's so obvious to me that goaltenders are seriously misevaluated that I go crazy everytime I hear golaie stats discussed.
Posted 11:06 a.m.,
September 19, 2003
(#3) -
tangotiger
Funny thing, Andrew. I've been doing the goalie stuff for a few years now. I've got a system in place that I think is pretty simple and logical. Same thing for plus/minus for players. I've got "sim scores" for players as well.
But, seeing my time management skills aren't up-to-speed yet, I'm not sure when I can deliver on this publicly, if ever.
Posted 3:04 p.m.,
September 19, 2003
(#4) -
Michael Humphreys
Sometime in the past year the New York Times had an article about a basketball evaluation system that sounds like the +/- system alluded to in the posted article. The +/- system seems to evaluate each combination of 5 players that has been on the court and compares how that combination does compared to the others. How that might be translated into individual ratings is beyond me, although it might be something as simple as comparing the mean scores of *four* player combinations with and without the player being rated, and some adapation of a t-test for scale and statistical significance of the player's impact. The limitation of that approach might be that it only captures the interactive impact of a player with that particular combination (or all four-player combinations for that team), not the player's context-independent skill in a "average" environment. The other problems might be that you would probably want to adjust ratings for each combination for opponent quality and home/road impact, which is very signficant in basketball. The sample sizes could get quite small very quickly.
Basketball has at least some counting stats that are meaningful, and people might already have a good idea of the average opportunity cost of a missed 3-point shot and missed 2-point shot, as well as the average value of a rebound. I hadn't realized the subjectiveness of credit for assists. It makes sense that their might be an offsetting cost to high "steal" rates.
Has anyone tried something similar to the +/- system for soccer? The scores in soccer are so low that it might be difficult. I would imagine that Tango's approach for rating hockey goalies might be transferable to soccer goalies.
Since football season is here, does anyone have any suggestions for good "sabermetric" books on football?
Posted 3:41 p.m.,
September 19, 2003
(#5) -
tangotiger
I agree that soccer won't translate as well, since you would get enough sample to do your testing on.
The way to do the plus/minus thing is virtually the same as you would do with strength-of-schedule. I would call this "strength-of-context" (SoC).
Jason Kidd + player1 + player2 + player3 + player4 + opp1 + opp2 + opp3 + opp4 + opp5 + Home/Away = 14 pts for + 11 pts allowed over 30 minutes
You do this for every single combination of teammates and opponents. And for every player. It simply becomes a mathematical problem.
I agree you lose sample size at this level, but there's ways around it. (I wrote to a basketball exec asking him for the pbp files, and I'd do it for free, but no dice. Funny isn't it? I would imagine if I told him I'd do it for 10,000$ that he would take me more seriously.)
Posted 5:09 p.m.,
September 22, 2003
(#6) -
Jim
As for football books, try The Hidden Game of Football by Carroll and Palmer, Pro Football Prospectus by Lahman and Greanier, and for a couple oldie but goodies at your used book store, The New Thinking Man's Guide to Pro Football by Paul Zimmerman and Football by the Numbers by Allen Barra and George Ignatin.
There's also some interesting stuff going on at footballoutsiders.com.
They seem to have patterned themselves after Baseball Primer.
Posted 8:07 p.m.,
September 22, 2003
(#7) -
David Smyth
Maybe you guys aren't familiar with the basketball stat analysis books from the late 80', early 90s. The most prominent was probably Dave Heeren and his TENDEX system. This was a linear system, giving x points for each (of 10) individual official outcome. He had an adjustment for context called "game pace", which was simply based on total points scored and allowed per game. There were a couple other analysts who had an annual book run for several years. I had them all, and I think I saved one from each guy. It was the same as baseball, that these stats were pretty much as good as could be done, unless PBP type data were to become available. Maybe it is available now, so that defenders can be evaluated on more than just their blocks and steals, for example.